Bio in For Matic Stools

Bioinformatics Tools for
Detection and Clinical

Interpretation of Genomic
Variations
Edited by Ali Samadikuchaksaraei and Morteza Seifi
Bioinformatics Tools for
Detection and Clinical
Interpretation of Genomic
Variations
Edited by Ali Samadikuchaksaraei
and Morteza Seifi
Published in London, United Kingdom

Supporting open minds since 2005
Bioinformatics Tools for Detection and Clinical Interpretation of Genomic Variations
http://dx.doi.org/10.5772/intechopen.77443
Contributors
Xianquan Zhan, Tian Zhou, Tingting Cheng, Miaolong Lu, Gaston K. Mazandu, Emile R. Chimusa,
Ephifania Geza, Milaine Seuneu, Juliano Lino Ferreira, Leila Ferreira, Thelma Sáfadi, Tesfahun Alemu
Setotaw, Osman Ugur Sezerman, Kok-Siong Poon, Evelyn Siew-Chuan Koay, Julian Wei-Tze Tang
© The Editor(s) and the Author(s) 2019

The rights of the editor(s) and the author(s) have been asserted in accordance with the Copyright,
Designs and Patents Act 1988. All rights to the book as a whole are reserved by INTECHOPEN LIMITED.
The book as a whole (compilation) cannot be reproduced, distributed or used for commercial or
non-commercial purposes without INTECHOPEN LIMITED’s written permission. Enquiries concerning
the use of the book should be directed to INTECHOPEN LIMITED rights and permissions department
([email protected]).
Violations are liable to prosecution under the governing Copyright Law.
Individual chapters of this publication are distributed under the terms of the Creative Commons
Attribution 3.0 Unported License which permits commercial use, distribution and reproduction of
the individual chapters, provided the original author(s) and source publication are appropriately
acknowledged. If so indicated, certain images may not be included under the Creative Commons
license. In such cases users will need to obtain permission from the license holder to reproduce
the material. More details and guidelines concerning content reuse and adaptation can be found at
http://www.intechopen.com/copyright-policy.html.
Notice
Statements and opinions expressed in the chapters are these of the individual contributors and not
necessarily those of the editors or publisher. No responsibility is accepted for the accuracy of
information contained in the published chapters. The publisher assumes no responsibility for any
damage or injury to persons or property arising out of the use of any materials, instructions, methods
or ideas contained in the book.
First published in London, United Kingdom, 2019 by IntechOpen

IntechOpen is the global imprint of INTECHOPEN LIMITED, registered in England and Wales,
registration number: 11086078, The Shard, 25th floor, 32 London Bridge Street
London, SE19SG – United Kingdom
Printed in Croatia
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library
Additional hard and PDF copies can be obtained from [email protected]

p. cm.
Print ISBN 978-1-78923-799-3
Online ISBN 978-1-78923-800-6
eBook (PDF) ISBN 978-1-83881-844-9
We are IntechOpen,
the world’s leading publisher of
Open Access books
Built by scientists, for scientists
4,200+ 116,000+ 125M+

Open access books available International authors and editors Downloads
Our authors are among the
151
Countries delivered to
Top 1%
most cited scientists
12.2%
Contributors from top 500 universities
Selection of our books indexed in the Book Citation Index

in Web of Science™ Core Collection (BKCI)
Interested in publishing with us?

Contact [email protected]
Numbers displayed above are based on latest data collected.
For more information visit www.intechopen.com
Meet the editors
Ali Samadikuchaksaraei is a Professor of Medical Biotechnology,

Tissue Engineering and Regenerative Medicine at Iran Univer-
sity of Medical Sciences. He is an expert in detection, analysis,
and interpretation of clinically relevant genomic variations. As
a well-known figure in this field, he is regularly consulted by
the Iranian Academies, scientific organizations, and industrial
sectors.
Dr. Morteza Seifi completed his PhD and postdoctoral training

at the University of Alberta. He employs molecular and cellu-
lar techniques and bioinformatics tools to provide functional
evidence of pathogenicity of genomic variations. Dr. Seifi has re-
ceived several scholarships, awards, and grants, including one of
Canada’s largest and most prestigious endowments for academic
activities, the Izaak Walton Killam Memorial Scholarship.
Contents
Preface XIII
Chapter 1 1
The Bioinformatics Tools for Discovery of Genetic Diversity by Means
of Elastic Net and Hurst Exponent
by Leila Maria Ferreira, Thelma Sáfadi, Tesfahun Alemu Setotaw
and Juliano Lino Ferreira
Chapter 2 15
Bioinformatics Workflows for Genomic Variant Discovery, Interpretation
and Prioritization
by Osman Ugur Sezerman, Ege Ulgen, Nogayhan Seymen
and Ilknur Melis Durasi
Chapter 3 35
Orienting Future Trends in Local Ancestry Deconvolution Models
to Optimally Decipher Admixed Individual Genome Variations
by Gaston K. Mazandu, Ephifania Geza, Milaine Seuneu
and Emile R. Chimusa
Chapter 4 53
Recognition of Multiomics-Based Molecule-Pattern Biomarker for
Precise Prediction, Diagnosis, and Prognostic Assessment in Cancer
by Xanquan Zhan, Tian Zhou, Tingting Cheng and Miaolong Lu
Chapter 5 75
HCV Genotyping with Concurrent Profiling of Resistance-Associated
Variants by NGS Analysis
by Kok-Siong Poon, Julian Wei-Tze Tang and Evelyn Siew-Chuan Koay
Preface
Genomic variations are the basis for phenotypic variations of individual organisms
of the same species. These phenotypic variations could be of clinical importance in
humans and medically relevant organisms. Therefore detection of genomic varia-
tions, and interpretation of their phenotypic effects and pathogenic potentials, has
become a growing field in both biomedical research and clinical medicine.
Bioinformatics Tools for Detection and Clinical Interpretation of Genomic Variations is

an up-to-date compilation of chapters on application of data analysis and mining
tools for identification of clinically important genomic variations.
Chapter 1 discusses the application of non-decimated wavelet transform (NDWT)

coupled with elastic net domains and Hurst exponent in identification of genetic
diversity. Chapter 2 describes a comprehensive workflow for analysis of whole
exome and whole genome sequencing data. It also presents the steps needed for
variant discovery workflow with a particular focus on germline short variants and
germline short insertion and deletion events. Additionally, this chapter outlines
methods for analysis of somatic and structural variations.
Chapter 3 discusses local ancestry deconvolution and dating admixture events and
the possible gaps in the knowledge that lead to the current challenges. Chapter 4
addresses the value of multiomics-based molecular patterns and the concept of
pattern recognition and pattern biomarkers in cancer diagnosis and prognosis. It
also explores the application of these concepts in personalized medicine. Chapter 5
addresses the genetic diversity of the hepatitis C virus and discusses its genotyping
and concurrent variant profiling, as identification of resistance-associated variants
of this virus determines the choice of anti-viral regimes in infected patients.
We would like to thank all the authors for their contributions and time in prepar-
ing this valuable collection. Also, we would like to extend our thanks to Mr. Luka
Cvjetković for his great help during the editing of this book and to IntechOpen for
their commitment and support.
Ali Samadikuchaksaraei, MD, PhD

Departments of Medical Biotechnology and Tissue Engineering
& Regenerative Medicine,
Iran University of Medical Sciences,
Tehran, Iran
Morteza Seifi, PhD

Alberta Children’s Research Institute (ACRI),
Calgary, Alberta,
Canada
Chapter 1
The Bioinformatics Tools for

Discovery of Genetic Diversity by
Means of Elastic Net and Hurst
Exponent
Leila Maria Ferreira, Thelma Sáfadi,
Tesfahun Alemu Setotaw and Juliano Lino Ferreira
Abstract
The genome era allowed us to evaluate different aspects on genetic varia-

tion, with a precise manner followed by a valuable tip to guide the improvement
of knowledge and direct to upgrade to human life. In order to scrutinize these
treasured resources, some bioinformatics tools permit us a deep exploration of
these data. Among them, we show the importance of the discrete non-decimated
wavelet transform (NDWT). The wavelets have a better ability to capture hidden
components of biological data and an efficient link between biological systems and
the mathematical objects used to describe them. The decomposition of signals/
sequences at different levels of resolution allows obtaining distinct characteristics in
each level. The analysis using technique of wavelets has been growing increasingly
in the study of genomes. One of the great advantages associated to this method cor-
responds to the computational gain, that is, the analyses are processed almost in real
time. The applicability is in several areas of science, such as physics, mathematics,
engineering, and genetics, among others. In this context, we believe that using R
software and applied NDWT coupled with elastic net domains and Hurst exponent
will be of valuable guideline to researchers of genetics in the investigation of the
genetic variability.
Keywords: wavelet, genome, NDWT, elastic net, Hurst exponent
1. Introduction
The genome era allowed us to evaluate different aspects on genetic variation,

with a precise manner followed with a valuable tip to guide the improvement of
knowledge and direct to upgrade to human life. In order to scrutinize these trea-
sured resources, some bioinformatics tools permit us a deep exploration of these
data. Among them, we display the significance of the discrete non-decimated
wavelet transform (NDWT). The wavelets they possess improved capability to
identify occult constituents of biological data and do a well-organized connec-
tion amid biological systems and the mathematical items used to designate them.
The decomposition of signals/sequences at diverse stages of resolution allows
1
obtaining different characteristics in each level. The analysis using technique of

wavelets has been growing increasingly in the study of genomes. One of the great
advantages associated to this method corresponds to the computational gain, that
is, the analyses are processed almost in real time. The applicability is in numerous
themes of science, as physics, mathematics, engineering, genetics, meteorology,
and oceanography, among others. The wavelet transform comprehends a technique
of see and represents a signal. This signal is decomposed in resolution intensities,
where each level brings a detailing. Mathematically, it is embodied by a function
oscillating in time or space. As characteristic, it has sliding windows that expand
or compress to capture low- and high-frequency signals. Its starting point arose in
the field of seismic training to designate the instabilities ascending from a seismic
impulse. Among the wavelets techniques, we have the discrete non-decimated
wavelet transform (NDWT), whose main characteristic is that it can work with any
size of signals/sequences. In this procedure, the inductance is paraphrase invari-
ants, to be exact; the selection of origin is irrelevant, provided all the observations
are used in the analysis, a condition that does not happen in the discrete decimated
wavelet transform (DWT). The technique of discrete wavelet transforms is being
used to find gene locations in genomic sequences, detecting long-range correlations,
discovering periodicities in sequences of DNA and analysis of G + C patterns. The
NDWT technique may be applied in any genome type, increasing the promptness
of the analysis, because the analyses with this method are processed almost in real
time. The wavelets have demonstrated to be an efficient method in the analysis of
DNA sequences. This tool is imperative to be applied to elastic net. The main feature
of the elastic net technique is the grouping of correlated variables where the quan-
tity of predictors is greater than the quantity of remarks. Furthermore, the Hurst
exponent allows the evaluation of genome similarities. In the same way, the NDWT
is crucial to evaluate the Hurst exponent. Strictly speaking, the bioinformatics
tool NDWT is a fundamental step to allow the examination of genomic variation
through the other subsequent bioinformatics tools, like elastic net and Hurst expo-
nent, which allow us to understand, interpret, and identify the genome variation in
a certain species.
2. Wavelet
Wavelet analysis, nowadays, is used widely in subjects such as signal processing,

engineering, physics, genetics, mathematics, medical sciences, economics, astron-
omy, etc. The genetic approach of this tool appears to be a valuable and interesting
possibility in science.
Wavelet is miniature wave. Whatsoever their form has a distinct number of
oscillations and lasts through a definite period of time or space. Wavelets hold
countless appropriate properties. Wavelets possess gender categories: there are
father wavelets φ and mother wavelets ψ. The father wavelet fits to 1, and the
mother wavelet fits to 0. Wavelets also arise in different shapes. There are the
discrete ones, the symmetric, the nearly symmetric, and the asymmetric. The key
aspect of wavelet investigation is that it allows the researcher to separate out a vari-
able or signal into its essential multiresolution components [1].
In the last 21 years, more than 2000 articles were published with wavelet tech-
nique in wide-ranging subjects.
Wavelet theory delivers an integrated background for number methods which
had been established autonomously for several signal processing applications [2].
Wavelet concept is established on Fourier analysis [3], in which all function may be
denoted as the sum of sine and cosine functions.
2
The Bioinformatics Tools for Discovery of Genetic Diversity by Means of Elastic Net and Hurst…
DOI: http://dx.doi.org/10.5772/intechopen.82755
Non-decimal wavelet transform (NDWT) possesses ample spectra of applica-

tion, including mammographic imaginings, geology, genomes, applied math-
ematics, applied physics, atmospheric sciences, and economics, among other
applications. In our specific case, we will approach the genomic approach.
When working with the complete genome, which is all the heritable information
of an organism that is set in DNA or, in some viruses, in RNA, this includes both the
genes and the noncoding sequences of a specific species; the main feature we find
is the large volume of data. To elucidate this problem, the technique called wavelets
has emerged as an efficient alternative in data compression, owning one of the main
advantages that this technique offers. However, wavelet functions are also com-
manding apparatuses in signal processing, noise elimination, separation of compo-
nents in the signal, identification of singularities, and detection of self-similarity,
among others.
The goals of this examination address to show how wavelets possibly will be
used in the analysis of genome clustering using the energy and interaction of wave-
let functions with data grouping techniques (elastic net and Hurst exponent).
Structure of the analysis: first it is required to acquire the signal of the genome
that will be analyzed; for this purpose, it is used to the tool called GC content. The
signal if is required to apply a wavelet transform, in this case the NDWT will be
used, working with the Daubechies wavelet with a certain number of null moments.
The amount of decomposition levels will depend on the size of the genome. The
scalogram is calculated using the detail coefficients obtained through the decom-
position levels. The clustering analysis is done using the dendrogram with medium
binding and applying the Mahalanobis distance.
In order to apply the elastic net technique in wavelet transform (NDWT), all
levels of decomposition are used; as a characteristic of this interaction, it is possible
to see the groupings at each of the decomposition levels.
Applying the Hurst exponent technique on the levels of signal decomposition,
each level brings information regarding the amount degree of Hurst exponent
index. All values found for the Hurst exponent are used in the dendrogram with
the mean binding and the distance of Mahalanobis. There are several methods of
estimation of Hurst exponent; the most commonly used is the R/S method.
3. Wavelet transform
Wavelet analysis has arisen as a possible device for spectral investigation owing
to the interval incidence localization which makes it appropriate for multifaceted
and motionless signals. The wavelet transform has added meaningfully in the train-
ing of many processes/signals in virtually all areas of earth science [4].
Wavelet is mathematical function. To be considered a wavelet, it must have the
total area on the function curve equals to zero. The energy of the behavior must be
limited (regularity and well located). Another need in the art is the speed and ease
of calculating the wavelet transform and the inverse transform.
Among various application areas of wavelets are computer vision, data compres-
sion, fingerprint compression at the FBI, data recovery affected by noise, similar
behavior detection, musical tones, astronomy, meteorology, numerical image
processing, and many others.
The wavelet transform rots a function demarcated in the period domain into
another function, well-defined in time domain and frequency domain. It is defined as
( a )dt,
W(a, b) = ∫∞∞ f(t) ___
___ t−b
1 ψ ∗ ___ (1)
√
| | a
3
which is a behavior function of two real parameters, a and b. If we define ψ a,b(t) as
( a ),
ψ a,b(t) = ___
___ t−b
1 ψ ∗ ___ (2)
√ a
| |
we may put another way the transform as the inner output of the functions f(t)
and ψ a,b(t):
∞
W(a, b) = 〈 f(t), ψ a,b(t)〉 = ∫-∞ f(t) ψ ∗a,b(t)dt. (3)
The function ψ(t) which equals ψ 1,0(t) is entitled the mother wavelet, while the
other functions ψ a,b(t) stay called daughter wavelets. The parameter b designates that
the function ψ(t) has been translated on the t axis of a distance equivalent to b, being
then a translation parameter. The parameter causes a change of scale, increasing (if
a > 1) or decreasing (if a < 1) the wavelet formed by the function. Consequently, the
parameter “a” remains known as the scaling parameter.
4. Wavelet analysis
There are abundant types of wavelet transform. Rely on the procedure one can
be desired that others. The wavelet analysis is prepared by the successive procedure
of wavelet transform with several values for the criterion a and b, representing the
decomposition of the signal components located in period and the agreeing to these
parameters. Each wavelet has a better or worse location in the domains of frequency
and of the time, so the analysis can be done with wavelets according to the desired
result. Wavelet analysis brings with it an analysis of where the resolution level is set
by the index a.
Discrete wavelets: among them are the Daubechies wavelet, wavelet of Cohen-
Daubechies-Feauveau (occasionally mentioned as CDF N/P or Daubechies bior-
thogonal wavelets), Beylkin [5], BNC wavelets, Coiflet, Mathieu wavelet, Haar
wavelet, binomial-QMF, Villasenor wavelet, Legendre wavelet, and symlet.
Continuous wavelets: (1) the real-valued wavelets are Mexican hat wavelet,
Hermitian wavelet, beta wavelet, Hermitian hat wavelet, and Shannon wavelet, and
the (2) complex-valued wavelets are Shannon wavelet, Morlet wavelet, complex
Mexican hat wavelet, and modified Morlet wavelet.
In the latest decades, the investigation using method of wavelets has been rising
progressively. One of the great rewards related with this method links to the compu-
tational improvement, that is, the analyses are treated virtually in real time. The
applicability is in numerous areas of science, like physics, mathematics, engineer-
ing, and genetics, among others.
The wavelet transform is a method of sighted and characterizes a signal.
Mathematically, it is characterized by a function wavering in time or space. As a
characteristic, it has sliding windows that increase or bandage to capture low- and
high-frequency signals, respectively [2]. Its origin arose in the field of seismic study
to define the instabilities ascending from a seismic impulse [6].
Among the wavelet techniques, we have the discrete non-decimated wavelet
transform (NDWT), whose main characteristic is that it may work with any extent
of signals/sequences.
In this procedure, the coefficients are translation invariants, that is, the selec-
tion of source is unrelated since all the annotations are done in the investigation,
a condition that does not happen in the discrete decimated wavelet transform
(DWT).
4
In recent period, the discrete wavelet transforms were worn to find gene sites in
sequences of the genome [7], finding long-range correlations, finding periodicities
in sequences of the DNA molecule [8], and also in the scrutiny of G + C patterns [9].
The clustering analysis is often assumed to deal with DNA sequences profi-
ciently. A wavelet-based element vector model was anticipated for grouping of DNA
sequences [10].
The distinction of the discrete NDWT is to retain the similar extent of data in even
and odd decimations on each measure and remain to do the identical on each subse-
quent scale, being D0 the dyadic decimation, D1 the odd decimation, H the high-pass
filter, and L the low-pass filter. Consider, for example, an input path (y1, …,yn). Then,
put on and preserve both D0 Hy and D1 Hy, even and odd indexed of the observation-
filtered wavelets. Each of these sequences is length n/2. Consequently, in whole, the
amount of wavelet coefficients in both decimals on the better scale is 2 × n/2 = n [11].
5. GC content
An important parameter in genetics is the GC content. They are referred as

the percentage of each bases of nitrogen composition of the molecule of DNA
or RNA. We own the adenine, cytosine, guanine, thymine, and uracil. They are
called by the acronyms A, C, G, T, and U, respectively. The last one belongs to RNA
molecule replacing thymine. They are applied to the complete genome or deter-
mined fragment. This concept may be applied in coding or noncoding molecule seg-
ment. The adenine has the same quantity of thymine (DNA) or uracil (RNA). The
cytosine has the same sum of guanine in either RNA or DNA. The amount of GC is
related to high-stability one which value is less than AT or AU. In the opposite is low
stability when this quantity is relatively small compared with AT or AU. This detail
is because GC has three hydrogen bonds, although AU or AT has two of them.
The GC proportion inside a genome is established to be evidently variable. The
DNA coding section is straight proportional to stand-up G + G.
In varied organisms, GC content is found to be too variable, which donate the
dissimilarities in recombination pattern, including association with DNA repair,
selection, and in the alteration of mutational bias patterns. Due to the essence of the
genetic coding, it is nearly incredible for an organism to have a genome with a GC
content pending either 0 or 100%. An organism species with an exceptionally low
GC content is Plasmodium falciparum having about 20% of GC amount, published
at NCBI—available at https://www.ncbi.nlm.nih.gov/bioproject?cmd=Retrieve&do
pt=Overview&list_uids=148.
The GC percentage is the largely used systematic approaches in many pro-
karyotic organisms mainly in bacteria species. Actinobacteria are one example of
uppermost GC bacterial content. Another example is Streptomyces coelicolor being
72% of G + G amount.
Interestingly, the software apparatuses GCSpeciesSorter [12] and TopSort [13]
are used for categorizing species centered on their GC contents.
6. Daubechies wavelet
The Daubechies wavelets, established on the study done by Ingrid Daubechies,

comprise of a series of orthogonal wavelets determining a discrete wavelet transform
and categorized by a greatest amount of disappearing moments for certain given pro-
vision. With every wavelet assembly of this category lies in a scaling function (enti-
tled the father wavelet) that produces an orthogonal multiresolution investigation.
5
Ingrid Daubechies is a Belgian physicist and mathematician. Daubechies was the first
female to be chair of the International Mathematical Union (2011–2014). She is very
well acknowledged for her study using wavelets in image compression.
Daubechies earned the Louis Empain Prize for Physics in 1984, conferred once
every 5 years to a Belgian scientist on the basis of a study done before the age of 29.
In the middle of 1992 and 1997, she stood a partner of the MacArthur Foundation,
in addition in 1993, she was designated to the American Academy of Arts and
Sciences. In 1994, she earned the American Mathematical Society Steele Prize for
explanation for her book Ten Lectures on Wavelets and was requested to provide an
entire talk in Zurich at the International Congress of Mathematicians. In 1997, she
stood granted the AMS Ruth Lyttle Satter Prize available at http://www.ams.org/
profession/prizes-awards/pabrowse#year=1997. In 1998, she was selected to the
United States National Academy of Sciences, which can be visualized at http://nas.
nasonline.org/site/Dir/1753239219?pg=vprof&mbr=1001102&returl=http%3A%
2F%2Fwww.nasonline.org%2Fsite%2FDir%2F1753239219%3Fpg%3Dsrch%26vie
w%3Dbasic&retmk=search_again_link and acquired the Golden Jubilee Award for
Technological Innovation from the IEEE Information Theory Society (https://www.
itsoc.org/honors/golden-jubilee-awards-for-technological-innovation). She turns
into an overseas fellow of the Royal Netherlands Academy of Arts and Sciences in
1999 accessible at https://www.knaw.nl/en/members/foreign-members/4013.
In 2000, Daubechies turns out to be the pioneer lady to obtain the National
Academy of Sciences Award in Mathematics, stated every 4 years for excellence
in published mathematical investigation. The prize honored her for important
findings on wavelets and wavelet growths and designed for her accomplishment in
building wavelet methods a constructive elementary apparatus of applied math-
ematics. This achievement is presented on https://www.knaw.nl/en/members/
foreign-members/4013. She was also conferred the Basic Research Award, German
Eduard Rhein Foundation, which could be visualized on https://web.archive.org/
web/20110718233021/http://www.eduard-rhein-stiftung.de/html/Preistraeger_e.
html and https://web.archive.org/web/20110718234059/http://www.eduard-
rhein-stiftung.de/html/2000/G00_e.html and the NAS Prize in Mathematics
https://web.archive.org/web/20101229195210/http://www.nasonline.org/site/
PageServer?pagename=AWARDS_mathematics.
Generally, the Daubechies wavelet properties stay preferred to have the maxi-
mum sum A of vanishing moments (this does not make sure of indicating the
preeminent levelness) on behalf of assumed provision measurement 2A-1 [3]. It is
present in two designation patterns in routine, DN via the extent or total of blows
and dbA stating to the quantity of vanishing moments. Thus db2 and D4 stand the
equivalent wavelet transform.
Among the 2A-1 thinkable resolution of the arithmetical calculations for the
moment and orthogonal circumstances, the one is elected whose scaling filter has
extreme phase. Wavelet transform remains too easy to place hooked on training
through the debauched wavelet transform. Daubechies wavelets are broadly used
in answering wide-ranging problems, for example, self-homology assets of sign or
fractal difficulties and sign cutoffs, among others.
Daubechies wavelets remain not demarcated in footings of the subsequent
scaling and wavelet functions; actually, they are not probable to inscribe down
in locked procedure.
In the production of a wavelet scaling arrangement, low-pass filter and the
wavelet sequence band-pass filter will standardized to ensure entirety unenliven 2
and summation__of squares unenliven 2. In particular requests, they are standardized
to require sum√2 ; thus one and other arrangements and entirely changes of them by
an even sum of coefficients are orthonormal to each other.
6
The employment of Daubechies wavelets though software such as Mathematica

rope straight mode is available at https://reference.wolfram.com/language/ref/
DaubechiesWavelet.html, a basic execution is humble in MATLAB. This applica-
tion routines periodization to grip the problematic of limited measurement signals.
Other, further refined devices are accessible, but habitually it is not required to use
these as it merely touches the many split ends of the converted signal. The peri-
odization is fulfilled in the onward transform straight in MATLAB vector system
and the inverse transform by means of the circshift() function.
7. Non-decimal wavelet transform
Non-decimal wavelet transform (NDWT) has the benefits of period invariance

and redundancy, paralleled to the typical orthogonal wavelet transformations.
NDWT owns properties beneficial in various wavelet applications. Furthermore,
NDWT matrix is capable to powerfully map a signal arising from an acquirement
field to the wavelet sphere with humble matrix multiplication and deprived of the
prerequisite of the whole quantity of the signal [14].
A widespread version of wavelet transform is a NDWT, which can overwhelm
sensitivity to translations in time and change found in typical [15] orthogonal
wavelet transform. Initially in the 1990s, NDWT arose in scientific literature using
several names for a figure of applications and purposes [16].
A process that approaches nonstop wavelet transform with an iterative algo-
rithm, which evicted to be corresponding to a shift-invariant representation, was
put forward by [17]. Furthermore, a resourceful algorithm was defined with
O(n log2 (n)) complexity for scheming wavelet coefficients that stand shift-
invariant, to be exact, humble repetitious wavelet coefficients at wholly N circulant
shift for an input signal size of N [5, 18]. In addition, a wavelet packet decomposi-
tion for time invariance and applied it to estimation and detection problems was
proposed by Pesquet and collaborators [19] and lengthy finished in the study [20],
uses an over ample wavelet decomposition, which is stated to as discrete wavelet
frame, for arrangement of texture. After that, two other studies [21, 22] tested
translation-invariant transform to verge for noise reduction. Then, the study of
stationary wavelet transform with example applications for local spectra estimation
was published [23]. Finally, an examination of applied translation-invariant wavelet
algorithm for data compression was done [24].
The time-invariance property of NDWT generates a reduced mean square error
and also reduces the Gibbs phenomenon in d-noising applications [21]. Conversely,
the defilement of variance maintenance in NDWT embarrasses the signal restora-
tion [16].
Major benefits of a NDWT matrix are squeezability, calculation promptness,
and tractability in magnitude of an input signal. We previously deliberated the
superior compressibility when NDWT matrices are well-worn for 2-D scale-mixing
transforms.
NDWT possess ample spectra of application, including mammographic imagin-
ings, geology, genomes, physics, atmospheric sciences, and economics, among
other applications.
8. Scalogram
Spectrogram is an ample prevalent tool in signal analysis because it provides

a scattering of signal energy in time-frequency plane. The wavelet spectrogram
7
is broadly known like scalogram [25]. Comprehend a distribution of energy in

timescale plane. The scalogram yields a more or less simply intelligible visual in
two-dimensional representations of signals [26].
The scalogram is a valuable device for the understanding of the wavelet signal
represented. It is like a graph of the square sum of the wavelet coefficients in
different levels. In the occurrence of discrete transformation, it embodies a decom-
position of function energy without timescale. One of its features is the aptitude
to detect periodic components of the signal; either apparatuses will result in peaks
in the scalogram. These apparatuses may be mined from the signal by dividing the
ripple coefficients into different sets, where each of these sets is at the same peak.
High- and low-frequency apparatuses of a signal might be restored by applying a
reverse loop transformation to separate sets [27].
The energy E (j) aimed at the wavelet d coefficients in each level j, which cor-
responds to the scalogram, is represented by
n
2
E( j) = ∑ dj,k para j = 1, … , J (4)
k=0
9. Cluster analysis
Cluster analysis also known as unsupervised classification is a grouping of items

into diverse groups, each of which requisite be assembled rendering to the rules
of programming. This assembly must be handled computationally, without user
intervention.
The term clustering analysis, early termed by [15], actually contains an assort-
ment of different grouping algorithms, all of which address an important issue in
several areas of research: how to organize observed data into structures that make
sense or how to develop taxonomies capable of classifying data observed in differ-
ent classes. Important is to even consider that these assembly must be classes that
occur naturally in the dataset.
Clustering analysis is the designation given to the group of computational
techniques whose purpose is to separate objects into groups, based on the charac-
teristics that these objects have. The basic idea is to put objects in the same group
that are similar in some predetermined criteria. The criterion is usually based on a
dissimilarity function, which function receives two objects and returns the distance
between them. The groups determined by a quality metric should have high internal
and high homogeneity separation (external heterogeneity). This implies that the
elements of a given set should be mutually similar and, preferably, have a high
amount of differences from the elements of other sets [28].
Biologists, for instance, have to organize data observed in structures that make
sense, that is, develop taxonomies. Microbiologists confronted with a variety range
of species of a certain type, for example, must be capable to classify the observed
specimens into clusters before it has been possible to describe these microorganisms
in detail in ways to detach in detail the differences between species and subspecies.
Grouping procedures have been practiced in a huge range of areas. Ref. [29]
already provides a broad overview of several published studies on the use of group-
ing analysis techniques. In the medical field, for example, grouping of diseases by
symptom or cures can lead to very useful taxonomies. In areas of psychiatry, for
example, clustering of syndrome, for instance, paranoia, schizophrenia, and others,
is considered essential for proper therapy. In archeology, conversely, one has also
tried to group civilizations or times of civilizations based on tools of stone, funerary
8
objects, etc. In general, whenever a “mountain” of unknown data is required to be

classified into manageable cells, grouping methods are used.
10. Elastic net
In statistics and specifically in the suitable of linear or logistic regression models,

the elastic net is a standardized regression method that linearly couples the L1 and
L2 punishments of the lasso and ridge approaches. Figure 1 shows the elastic net
typical design.
Lasso is a regression method broadly worn in domains with huge datasets,
such as genomic data, where proficient and agile algorithms are vital [30]. Ridge
regression is a procedure for investigating manifold regression data that arise out
of multicollinearity. When multicollinearity arises, least squares estimates are
unbiased, but their variances are huge so they might be outlying from the accurate
value. In 1970, the investigation of [31] published an article about ridge regression,
approaching the tendentious appraisal for nonorthogonal issues. In 2009, [32]
study examined the ridge regression and their extensions applied to genome-wide
selection into Zea mays L.
R software, available at https://www.r-project.org/, has the packing necessary
to do a wavelet and elastic net based on genome sequence. Furthermore, the elastic
net may be also used with microsatellite (SSR) data. This tool could be used in any
genetic data of all organisms.
The most relevant article about elastic net was published in 2005 [33]. They say
that elastic net is of pronounced interest especially when the predictors’ number is
considerably higher than the sum of observations. This might be useful in real or in
simulation data.
The recent evolution of science brought a fast deeper understanding of the
genome. In this sense, through several methods with varying levels of complexity
added to the computational efficiency at the present days, we may easily compare
organisms based on their genetic dissimilarity. Along these lines, we used accurate
Figure 1.
Elastic net standard scheme.
9
genomic selection methods dropping the penalties of each approaches like in elastic
net, enabling the fitting of a certain statistical model. Therefore, an outstanding
methodology to analyze genome is elastic net domain used in several study, like
[33–36].
Recently, the tuberculosis strain’s differences were evaluated using the elastic
net domain [34]. In that examination, 10 genome sequences of Mycobacterium
tuberculosis with a window size of 10,000 bp were assessed combining the NDWT
and elastic net domain. This study encompasses 10 strains: 2 from drug resistant,
6 from drug susceptible, 1 from multidrug resistant, and finally 1 from exten-
sively drug resistant. The clustering detected on that analysis indicated to be real
adequate.
11. Hurst exponent
Hurst exponent is applied as a degree of long-standing memory of time series.

It associates to the autocorrelations of time series and the degree at which these
decline as the lag between pairs of values intensifications. This coefficient has
started to be established in hydrology, used to understand the variation level of dam
size at Nile River over a long cycle of time. Harold Edwin Hurst was a British engi-
neer that worked with hydrology; for this reason the coefficient has his surname.
Later, this exponent was used in several areas, including fractal geometry, storage
process, trends in financial market analyzing economic time series, mechanics,
physics, mathematics, computation, and finally to the long-ranging dependency in
DNA. Figure 2 displays the values of Hurst exponent and their interpretation in a
long-standing.
Using the genetic data, the Hurst exponent approach is able to build genetic
cluster based on genome sequence. There are a lot of estimation methods of Hurst
exponent: the original and best-known is the alleged rescaled range (R/S) analysis
promoted by [37, 38] and based on previous hydrological findings [39]. Alternatives
include DFA, periodogram regression [40] aggregated variances [41], local Whittle’s
estimator [42], and wavelet analysis [43, 44] both in the time domain and frequency
domain.
In our case, we performed a Hurst exponent in the bacterial strains used
in article [34]. We did many methods of Hurst exponent. Interestingly, the
R/S methodology was the most similar to the cluster obtained on elastic net
domain approach. This data is not shown due to being in a review process to an
International journal currently. Our data agree with the majority of scientific
papers published approaching the Hurst exponent, which so far applying the R/S
method.
Figure 2.
Hurst exponent pattern interpretation of the index value.
10
12. Conclusion
We strongly believe that exploring the genetic variability of any organism using
wavelet coupled with elastic net domain and/or Hurst exponent will be a valuable
and interesting tool. It is not difficult and the free R software could solve easily the
approach. In this way, it gives reliability and robustness in your results. Therefore,
these bioinformatics apparatuses provide more possibility to scrutinize the genetic
divergence of living organisms.
Conflict of interest
The authors do not have conflict of interests.
Author details
Leila Maria Ferreira1, Thelma Sáfadi1, Tesfahun Alemu Setotaw2 and

Juliano Lino Ferreira3*
1 Universidade Federal de Lavras, Lavras, Minas Gerais, Brazil
2 Ethiopian Institute of Agricultural Research, Addis Ababa, Ethiopia
3 Empresa Brasileira de Pesquisa Agropecuária, Bagé, Rio Grande do Sul, Brazil
*Address all correspondence to: [email protected]
© 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms
of the Creative Commons Attribution License (http://creativecommons.org/licenses/
by/3.0), which permits unrestricted use, distribution, and reproduction in any medium,
provided the original work is properly cited.
11
References
[1] Crowley PM. A guide to wavelets [10] Bao J, Yuan RY. A wavelet-based
for economists. Journal of Economic feature vector model for DNA
Surveys. 2007;21:207-267. DOI: clustering. Genetics and Molecular
10.1111/j.1467-6419.2006.00502.x Research. 2015;14:19163-19172. DOI:
10.4238/2015.December.29.26
[2] Percival DB, Walden AT. Wavelet
Methods for Time Series. 1st ed. [11] Nason G. Wavelet Methods in
Cambridge: Analysis Cambridge Statistics with R. 1st ed. New York:
University Press; 2000. 594 p. DOI: Springer-Verlag; 2008. 259 p. DOI:
10.1017/CBO9780511841040 10.1007/978-0-387-75961-6
[3] Dodin G, Vandergheynst P, Levoir [12] Karimi K, Wuitchik D, Oldach M,

P, et al. Fourier and wavelet transform Vize P. Distinguishing species
analysis, a tool for visualizing regular using GC contents in mixed DNA
patterns in DNA sequences. Journal of or RNA sequences. Evolutionary
Theoretical Biology. 2000;206:323-326 Bioinformatics Online. 2018;14:1-4.
DOI: 10.1177/1176934318788866
[4] Chamoli A. Wavelet analysis of
geophysical time series. e-Journal Earth [13] Lehnert E, Mouchka M, Burriesci M,
Science India. 2009;2:258-275 Gallo N, Schwarz J, Pringle J. Extensive
differences in gene expression between
[5] Beylkin G. On the representation symbiotic and aposymbiotic cnidarians
of operators in bases of compactly G3: Genes, genomes. Genetics. 2014;4:
supported wavelets. SIAM Journal on 277-295. DOI: 10.1534/g3.113.009084
Numerical Analysis. 1992;29:1716-1740
[14] Zhou H, Narayanan RM. Microwave
[6] Morlet J, Arens G, Fourgeau E, imaging of non-sparse object using
Giard D. Wave propagation and dual-mesh method and iterative
sampling theory–Part II: Sampling method with adaptive thresholding.
theory and complex waves. IEEE Transactions on Antennas and
Geophysics. 1982;47:222-236. DOI: Propagation. 2018; early access. DOI:
10.1190/1.1441329 10.1109/TAP.2018.2876164
[7] Ning J, Moore CN, Nelson JC. [15] Tryon RC. Cluster Analysis:
Preliminary wavelet analysis of genomic Correlation Profile and Orthometric
sequences. In: Proceedings of the (Factor) Analysis for the Isolation of
IEEE Computer Society Conference Unities in Mind and Personality. 1st ed.
on Bioinformatics CSB ’03. Stanford, Ann Arbor: Edwards Brothers;
California: IEEE; 2003. pp. 509-510 1939. 122 p
[8] Vannucci M, Liò P. Non-decimated [16] Kang M. Non-decimated wavelet

wavelet analysis of biological sequences: transform in statistical assessment
Applications to protein structure and of scaling: Theory and applications
genomics. Sankhyā: The Indian Journal [thesis]. Atlanta: Georgia Institute of
of Statistics, Series B. 2001;63:218-233. Technology; 2016
DOI: 10.2307/25053172
[17] Mallat S. Zero-crossings of a wavelet
[9] Daubechies I. Ten Lectures on transform. IEEE Transactions on
Wavelets. 1st ed. Berlin: Springer-Verlag; Information Theory. 1991;37:1019-1033.
1992. 344 p DOI: 10.1109/18.86995
12
[18] Shensa MJ. The discrete wavelet and understanding continuous

transform: Wedding the a trous and wavelet transforms. In: Combes
mallat algorithms. IEEE Transactions on JM, Grossmann A, Tchamitchian P,
Signal Processing. 1992;40:2464-2482. editors. Wavelets. Inverse Problems
DOI: 10.1109/78.157290 and Theoretical Imaging. Berlin,
Springer; 1990. pp. 2-20. DOI:
[19] Pesquet JC, Krim H, Carfantan 10.1007/978-3-642-75988-8_1
H. Time-invariant orthonormal wavelet
representations. IEEE Transactions on [27] Liò P, Vannucci M. Finding
Signal Processing. 1996;44:1964-1970 pathogenicity islands and gene transfer
events in genoma data. Bioinformatics.
[20] Unser M. Texture classification and 2000;16:932-940. DOI: 10.1093/
segmentation using wavelet frames. bioinformatics/16.10.932
IEEE Transactions on Image Processing.
1995;4:1549-1560 [28] Linden R. Técnicas de agrupamento.
Revista de Sistemas de Informação da
[21] Coifman RR, Donoho DL. FSMA. 2009;4:18-36
Translation-invariant de-noising. In:
[29] Hartigan JA. Clustering Algorithms.
Antoniadis A, Oppenheim G, editors.
Wavelets and Statistics. Lecture Notes 99th ed. New York: John Wiley & Sons;
in Statistics. Vol. 103. New York: 1975. 369 p
Springer; 1995. pp. 1-26. DOI:
[30] Friedman J, Hastie T, Tibshirani
10.1007/978-1-4612-2544-7_9
R. Regularization paths for generalized
linear models via coordinate descent.
[22] Lang M, Guo H, Odegard JE, Burrus
Journal of Statistical Software.
CS, Wells RO Jr. Nonlinear processing
2010;33:1-22
of a shift-invariant discrete wavelet
transform (dwt) for noise reduction. In: [31] Hoerl AE, Kennard RW. Ridge
Szu HH, editor. Wavelet Applications. regression: Biased estimation
2nd ed. Orlando: Proc. SPIE 2491; 1995. for nonorthogonal problems.
pp. 640-651. DOI: 10.1.1.24.4098 Technometrics. 1970;12:55-67
[23] Nason GP, Silverman BW. The [32] Piepho HP. Ridge regression and
stationary wavelet transform and some extensions for genome wide selection in
statistical applications. In: Antoniadis maize. Crop Science. 2009;49:1165-1176.
A, Oppenheim G, editors. Wavelets DOI: 10.2135/cropsci2008.10.0595
and Statistics. 1st ed. New York:
Springer; 1995. pp. 281-299. DOI: [33] Zou H, Hastie T. Regularization
10.1007/978-1-4612-2544-7_17 and variable selection via the
elastic net. Journal of the Royal
[24] Liang J, Parks TW. A translation- Statistical Society, Series B: Statistical
invariant wavelet representation Methodology. 2005;67:301-320. DOI:
algorithm with applications. IEEE 10.1111/j.1467-9868.2005.00503.x
Transactions on Signal Processing.
1995;44:225-232 [34] Ferreira LM, Sáfadi T, Ferreira
JL. Wavelet-domain elastic
[25] Rioul O, Vetterli M. Wavelets net for clustering on genomes
and signal processing. IEEE Signal strains. Genetics and Molecular
Processing Magazine. 1991;8:14-38. Biology. 2018;4:884-892. DOI:
DOI: 10.1109/79.91217 10.1590/1678-4685-GMB-2018-0035
[26] Grossmann A, Kronland- [35] Ogutu JO, Schulz-Streeck T, Piepho

Martinet R, Morlet J. Reading HP. Genomic selection using regularized
13
linear regression models: Ridge [44] Simonsen I, Hansen A, Nes

regression, lasso, elastic net and their OM. Determination of the Hurst
extensions. BMC Proceedings. 2012;6:1-6. exponent by use of wavelet transforms.
DOI: 10.1186/1753-6561-6-S2-S10 Physical Review E. 1998;58:2779-2787.
DOI: 10.1103/PhysRevE.58.2779
[36] Waldmann P, Mészáros G, Gredler
B, Fuerst C, Sölkner J. Evaluation of
the lasso and the elastic net in genome-
wide association studies. Frontiers in
Genetics. 2013;4:270. DOI: 10.3389/
fgene.2013.00270
[37] Mandelbrot BB, Wallis JR. Noah,

Joseph, and operational hydrology.
Water Resources Research.
1968;4:909-918
[38] Mandelbrot BB, Wallis JR.

Robustness of the rescaled range R/S in
the measurement of noncyclic long run
statistical dependence. Water Resources
Research. 1969;5:967-988
[39] Hurst HE. Long-term storage

capacity of reservoirs. Transactions of
the American Society of Civil Engineers.
1951;116:770-799
[40] Geweke J, Porter-Hudak S. The

estimation and application of
long memory time series models.
Journal of Time Series Analysis.
1983;4:221-238. DOI: https://doi.
org/10.1111/j.1467-9892.1983.tb00371.x
[41] Beran J, Terrin N. Estimation of

the long-memory parameter, based on
a multivariate central limit theorem.
Journal of Time Series Analysis.
1994;15:269-278. DOI: 10.1111/j.1467-
9892.1994.tb00192.x
[42] Robinson PM. Gaussian

semiparametric estimation of long-
range dependence. The Annals of
Statistics. 1995;23:1630-1661
[43] Riedi RH. Multifractal processes.

In: Doukhan P, Oppenheim G, Taqqu
MS, editors. Theory and Applications
of Long-Range Dependence. 1st ed.
Boston: Birkhäuser Boston; 2003.
pp. 625-716
14
Chapter 2
Bioinformatics Workflows for

Genomic Variant Discovery,
Interpretation and Prioritization
Osman Ugur Sezerman, Ege Ulgen, Nogayhan Seymen
and Ilknur Melis Durasi
Abstract
Next-generation sequencing (NGS) techniques allow high-throughput detection

of a vast amount of variations in a cost-efficient manner. However, there still are
inconsistencies and debates about how to process and analyse this ‘big data’. To
accurately extract clinically relevant information from genomics data, choosing
appropriate tools, knowing how to best utilize them and interpreting the results
correctly is crucial. This chapter reviews state-of-the-art bioinformatics approaches
in clinically relevant genomic variant detection. Best practices of reads-to-variant
discovery workflows for germline and somatic short genomic variants are presented
along with the most commonly utilized tools for each step. Additionally, methods
for detecting structural variations are overviewed. Finally, approaches and current
guidelines for clinical interpretation of genomic variants are discussed. As empha-
sized in this chapter, data processing and variant discovery steps are relatively well-
understood. The differences in prioritization algorithms on the other hand can be
perplexing, thus creating a bottleneck during interpretation. This review aims to
shed light on the pros and cons of these differences to help experts give more
informed decisions.
Keywords: genomics, NGS, variant discovery
1. Introduction
Whole genome sequencing (WGS) and whole exome sequencing are next-
generation sequencing (NGS) technologies that determine the full and protein-
coding genomic sequence of an organism, respectively. Deep sequencing of
genomes improves understanding of clinical interpretation of genomic variations.
Analyzing NGS data with the aim of understanding the impact and the importance
of genomic variations in health and disease conditions is crucial for carrying the
personalized medicine applications one step further.
One of the main obstacles for reaching the full potential of WES/WGS in
personalized medicine is bioinformatics analysis, which mostly requires strong
computational power. Analysis of WES/WGS data with publicly or commercially
available algorithms and tools require a proper computational infrastructure in
addition to an at least basic understanding of NGS technologies. Second, almost all
publicly available algorithms and tools focus on a single aspect of the entire process
15
and do not provide a workflow that can aid the researcher from start to finish.
Lastly, there are no gold standards for translating WES/WGS into clinical knowl-
edge, since different diseases need different strategies for the basic analysis to
obtain the genomic variants as well as further analyses, including disease-specific
interpretation and prioritization of the variants.
A comprehensive workflow that can be applied for WES/WGS data analysis is
composed of the following steps:
a. Quality control
• Evaluation of the quality of FASTQ data
• Trimming of the low-quality reads and removal of adaptors (if necessary)
b.Sequence alignment
c. Post-alignment processing
• Marking PCR duplicates
• Base quality score recalibration (BQSR)
d.Variant discovery
e. Downstream analyses
• Filtration of genomic variations
Figure 1.
An example single-sample variant discovery workflow. Each step is labelled in the black rectangles. The most
widely used tools for each operation are also presented.
16
Bioinformatics Workflows for Genomic Variant Discovery, Interpretation and Prioritization
• Annotation via a variant annotation tool
• Interpretation/prioritization of genomic variations
An example reads-to-variants workflow is visualized in Figure 1, highlighting

the input and output, a brief description, and the tools that can be utilized in each
step. While we present the most widely used tools, we would like to emphasize that
there are a great variety of tools/algorithms that can be utilized for each process.
Through the rest of this chapter, we give a brief outline of the purpose of each
step and try to provide a basic understanding of a state-of-the-art workflow for the
detection and interpretation of genomic variations. While there are countless
experimental designs, including WES/WGS and targeted (gene panel) sequencing,
the workflow presented here is applicable for all designs, occasionally requiring
slight modifications. We particularly focus on the detection and interpretation of
germline short variants, namely, single nucleotide variations (SNVs) and germline
short insertion or deletion events (indels). However, outlines of analyses for
somatic variants and for structural variations (SVs) are also presented. Finally,
current approaches and tools for clinical interpretation of genomic variations are
discussed.
2. Detection of genomic variations
Detection of genomic variations beginning from raw read data is a multistep task
that can be executed using numerous tools and resources. The workflow outlined in
the introduction section is laid out in detail in this section, including the best
practice recommendations and common pitfalls.
2.1 Acquisition of raw read data: the FASTQ file format
The raw data from a sequencing machine are most widely provided as FASTQ
files, which include sequence information, similar to FASTA files, but additionally
contain further information, including sequence quality information.
A FASTQ file consists of blocks, corresponding to reads, and each block consists
of four elements in four lines (Figure 2).
The first line contains a sequence identifier and includes an optional description
of sequencing information (such as machine ID, lane, tile, etc.). The raw sequence
letters are presented in line 2. The third line begins with a “+” sign and optionally
contains the same sequence identifier. The last line encodes the quality score for the
sequence in line 2 in the form of ASCII characters. While specific scoring measures
might differ among platforms, Phred Score (Qphred = -10log10P, where P being the
probability of misreading any given base) is the most widely used.
2.2 Quality control
In general, the raw sequence data acquired from a sequencing provider is not
immediately ready to be used for variant discovery. The first and most important
Figure 2.
Example FASTQ file format.
17
step of the WES/WGS analysis workflow following data acquisition is the quality
control (QC) step. QC is the process of improving raw data by removing any
identifiable errors from it. By performing QC at the beginning of the analysis,
chances encountering any contamination, bias, error, and missing data are
minimized.
The QC process is a cyclical process, in which (i) the quality is evaluated, (ii) QC
is stopped if the quality is adequate, and (iii) a data altering step (e.g., trimming of
low-quality reads, removal of adapters, etc.) is performed, and then the QC is
repeated beginning from step (i).
The most commonly used tool for evaluating and visualizing the quality of
FASTQ data is FastQC (Babraham Bioinformatics, n.d.), which provides compre-
hensive information about data quality, including but not limited to per base
sequence quality scores, GC content information, sequence duplication levels, and
overrepresented sequences (Figure 3). Alternatives to FastQC include, but are not
limited to, fastqp, NGS QC Toolkit, PRINSEQ , and QC-Chain.
Below, QC approaches for the most commonly encountered data quality issues
are discussed: adapter contamination and low-quality measurements toward the 50
and 30 ends of reads.
Adapters are ligated to the 50 and 30 ends of each single DNA molecule during
sequencing. These adapter sequences hold barcoding sequences, forward/reverse
primers, and the binding sequences to immobilize the fragments to the flow cell and
allow bridge amplification. Since the adapter sequences are synthetic and are not
seen in any genomic sequence, adapter contamination often leads to NGS alignment
errors and an increased number of unaligned reads. Hence, any adapter sequences
need to be removed before mapping. In addition to adapter removal, trimming can
be performed to discard any low-quality reads, which generally occur at the 50 and
30 ends.
There is an abundance of tools for QC, namely, Trimmomatic [1], CutAdapt [2],
AlienTrimmer [3], Skewer [4], BBDuk [5], Fastx Toolkit [6], and Trim Galore [7].
Figure 3.
An example FastQC result.
18
In addition to these stand-alone tools, R packages for QC, such as PIQA and
ShortRead, are also available.
While QC is the most important step of NGS analysis, one must keep in mind
that once basic corrections (such as the ones described above) are made, no amount
of further QC can produce a radically better outcome. QC cannot simply turn bad
data into good data. Moreover, it is also important to remember that because QC
may also introduce error that can affect the analysis, it is vital never to perform
error correction on data that does not need it.
2.3 Sequence alignment
In order to find the exact locations of reads, each must be aligned to a reference
genome. Efficiency and accuracy are crucial in this step because large quantities of
reads could take days to align and a low-accuracy alignment would cause inade-
quate analyses. For humans, the most current and widely used reference sequences
are GRCh37 (hg19) and GRCh38 (hg38). Similar to any bioinformatics problem,
there are a great number of tools for alignment of sequences to the reference
genome, to name a few, BWA [8], Bowtie2 [9], novoalign [10], and mummer [11].
After aligning, a Sequence Alignment Map (SAM) file is produced. This file
contains the reads aligned to the reference. The binary version of a SAM file is
termed a Binary Alignment Map (BAM) file, and BAM files are utilized for random-
access purposes. The SAM/BAM file consists of a header and an alignment section.
The header section contains contigs of aligned reference sequence, read groups
(carrying platform, library, and sample information), and (optionally) data
processing tools applied to the reads. The alignment section includes information on
the alignments of reads.
2.4 Post-alignment processing
One of the key steps in any reads-to-variants workflow is post-alignment data

processing to produce analysis-ready BAM files. This step includes data clean-up
operations to correct for technical biases: marking duplicates and recalibration of
base quality scores.
During the preparation of samples for sequencing, PCR duplicates arise at the
step of PCR amplification of fragments. Since they share the same sequence and the
same alignment position, they can lead to problems in variant detection. For exam-
ple, during SNV calling, false-positive variants may arise as some alleles may be
overrepresented due to amplification biases. To overcome this issue, PCR duplicates
are marked with a certain tag using an algorithm (MarkDuplicates) available in the
tool Picard [12]. Marking duplicates constitutes a major bottleneck since it involves
making a large number of comparisons between all the read pairs. Thus, the major-
ity of the effort in optimizing the runtime of reads-to-variants workflows is focused
on this step.
As aforementioned, NGS platforms provide information on the quality of each
base that they measure in the Phred Score format. The relationship of a Phred Score
with accuracy is straightforward: a Phred Score of 10 represents 90% accuracy, 20
equals 99%, 30 equals 99.9%, and so on. The raw scores produced by the sequencing
machine are prone to technical errors, leading to over- or underestimated base
quality scores. Base quality score recalibration (BQSR) is a machine learning
approach that models these errors empirically and readjusts the base quality scores
accordingly. Through this recalibration, more accurate and reliable base quality
scores are achieved, which in turn improves the reliability of the downstream steps
19
in further analyses. The most widely used tool for BQSR is provided by the Genome
Analysis Toolkit (GATK) [13].
After these post-alignment data processing operations, an analysis-ready BAM
file is obtained.
2.5 Short variant discovery
In this section, approaches for the discovery of germline SNV and indels are
discussed. In the following sections, approaches for the discovery of somatic short
variants and of structural variations are outlined.
Following data processing steps, the reads are ready for downstream analyses,
and the following step is most frequently variant calling. Variant calling is the
process of identifying differences between the sequencing reads, resulting from
NGS experiments and a reference genome. Countless variant callers have been and
are being developed for accomplishing this challenging task as alignment and
sequencing artifacts complicate the process of variant calling. For recent studies
comparing different variant callers, see [14–16]. Methods for detecting short vari-
ants can be broadly categorized into “probabilistic methods” and “heuristic-based
algorithms.” In probabilistic methods, the distribution of the observed data is
modeled, and then Bayesian statistics is utilized to calculate genotype probabilities.
In contrast, in heuristic-based algorithms, variant calls are made based on a number
of heuristic factors, such as read quality cutoffs, minimum allele counts, and bounds
on read depth. Whereas heuristic-based algorithms are not as widely used, they can
be robust to outlying data that violate the assumptions of probabilistic models.
The most widely used state-of-the-art variant callers include, but are not limited
to, GATK-HaplotypeCaller [13], SOAPsnp [17], SAMTools [18], bcftools [18],
Strelka [19], FreeBayes [20], Platypus [21], and DeepVariant [22]. We would like to
emphasize that for WES/WGS, a combination of different variant callers outper-
forms any single method [23].
2.6 Filtration of variants
Following the variant calling step, raw SNV and indels in the Variant Call
Format (VCF) are obtained. These should then be filtered either through applying
hard filters to the data or through a more complex approach such as GATK’s Variant
Quality Score Recalibration (VQSR).
Hard filtering is applied by filtering via thresholds for metrics such as
QualByDepth, FisherStrand, RMSMappingQuality, MappingQualityRankSumTest,
ReadPosRankSumTest, and StrandOddsRatio.
VQSR, on the other hand, relies on machine learning to identify annotation
profiles of variants that are likely to be real. It requires a large training dataset
(minimum 30 WES data, at least one WGS data if possible) and well-curated sets of
known variants. The aim is to assign a well-calibrated probability to each variant
call to create accurate variant quality scores that are then used for filtering.
The accuracy of variant calling is also affected by coverage. Coverage can be
broadly defined as the number of unique reads that include a given nucleotide.
Coverage is affected by the accuracy of alignment algorithms and by the
“mappability” of reads. Coverage can be utilized for both the filtration of variants
and for a general evaluation of the sequencing experiment. Tools for assessing
coverage information include GATK [13], BEDTools [24], Sambamba [25], and
RefCov [26].
20
For validation of variants and for detecting sequencing artifacts, Integrative

Genomics Viewer (IGV) [27] can be used to visualize the processed reads. In
addition to this in silico evaluation, Sanger sequencing can be performed.
2.7 Variant annotation
Variant annotation is yet another critical step in the WES/WGS analysis

workflow. The aim of all functional annotation tools is to annotate information of
the variant effects/consequences, including but not limited to (i) listing which gene
(s)/transcript(s) are affected, (ii) determination of the consequence on protein
sequence, (iii) correlation of the variant with known genomic annotations (e.g.,
coding sequence, intronic sequence, noncoding RNA, regulatory regions, etc.), and
(iv) matching known variants found in variant databases (e.g., dbSNP [28], 1000
Genomes Project [29], ExAc [30], gnomAD [31], COSMIC [32], ClinVar [33], etc.).
The consequence of each variant is expressed through Sequence Ontology (SO)
terms. The severity and impact of these consequences are often indicated using
qualifiers (e.g., low, moderate, high).
Many annotation tools utilize the predictions of SNV/indel deleteriousness pre-
diction methods, to name a few, SIFT [34], PolyPhen-2 [35], LRT [36],
MutationTaster [37], MutationAssessor [38], FATHMM [39], GERP++ [40], PhyloP
[41], SiPhy [42], PANTHER-PSEP [43], CONDEL [44], CADD [45], CHASM [46],
CanDrA [47], and VEST [48].
Annotation can have a strong influence on the ultimate conclusions during
interpretation of genomic variations as incomplete or incorrect annotation
information will result in the researcher/clinician to overlook potentially relevant
findings.
Once the analysis-ready VCF is produced, the genomic variants can then be
annotated using a variety of tools and a variety of transcript sets. Both the choice of
annotation software and transcript set (e.g., RefSeq transcript set [49], Ensembl
transcript set [50]) have been shown to be important for variant annotation [51].
The most widely used functional annotation tools include but are not limited to
AnnoVar [52], SnpEff [53], Variant Effect Predictor (VEP) [54], GEMINI [55],
VarAFT [56], VAAST [57], TransVar [58], MAGI [59], SNPnexus [60], and
VarMatch [61]. Below some of the popular tools are briefly described:
AnnoVar: AnnoVar is one of the most popular tools for annotation of SNV and
indels. AnnoVar takes a simple text-based format that includes chr, start, end, ref,
alt, and optional field(s) as an input. To use AnnoVar, one must convert VCF file
format to the AnnoVar input file format. The tool returns a single annotation for
each variant. If there exists more than one transcript for a specific variant resulting
in different consequences, AnnoVar chooses the transcript according to the gene
definition set by the user.
SnpEff: SnpEff is an open-source tool that annotates variants and predicts their
effects on genes by using an interval forest approach. SnpEff annotates variants
based on their genomic locations such as intronic, untranslated region, upstream,
downstream, splice site, or intergenic regions and predicts coding effects. snpEff
also generates extensive report files and is easily customizable.
VEP: VEP is an open-source, free-to-use toolset for the analysis, annotation, and
prioritization of genomic variants in coding and noncoding regions. VEP is one of
the few annotation tools that annotates variants in regulatory regions.
GEnome MINIng (GEMINI): GEMINI is a flexible software package for
exploring all forms of human genetic variations. Different from most other annota-
tion tools, GEMINI integrates genetic variation with a set of genome annotations.
21
While the abovementioned are all variant annotation tools, it might be wise to put
GEMINI in a different category as it has other built-in tools to make further analysis
of the variants easier.
2.8 Somatic genomic variations
The workflow for identifying somatic short variants (somatic SNV/indels) is

nearly identical to the germline short variant discovery workflow (Figure 4).
However, several differences exist. Firstly, for the discovery of somatic genomic
variations, sequencing both tumor tissue and a matched normal sample (blood,
adjacent normal tissue, etc.) is mostly (but not necessarily) preferred. The QC,
alignment, and post-alignment data processing steps are identical and are
performed for both the tumor and normal data, separately. The main difference is
the variant calling step, where both the tumor and normal processed read data are
utilized to identify somatic SNV/indels, i.e., short variants that are present in the
tumor but not in the normal. Some tools (such as GATK-MuTect2) can utilize
additional information from a panel of normals, a collection of normal samples
(typically larger than 40) that are believed to have no somatic variants, processed in
the same manner at each step and the purpose of which is to capture recurrent
technical artifacts.
Several tools exist for tumor-normal somatic variant calling, to name a few,
GATK - MuTect2 [13], VarScan2 [62], Strelka [19], SomaticSniper [63], SAMtools
[18], SomaticSeq [64], FreeBayes [65], CaVEMan [66], and FaSD-somatic [67]. For
further information on somatic variant calling, we encourage the reader to refer to a
recent and comprehensive review on somatic variant calling algorithms [16].
Filtration of the raw SNV and indels also differs for the somatic workflow.
Several different approaches exist for the filtration of raw somatic variants. Most
Figure 4.
An example somatic variant discovery workflow. Each step is labelled in the black rectangles. Most widely used
tools for each operation are also presented. As can be seen in the diagram, the processing steps until the variant
calling step are performed for both the normal and tumor data, separately.
22
frequently, tumor-specific metrics including the estimation of tumor heterogeneity

and cross-sample contamination are used in addition to the aforementioned metrics
for detection of sequencing/alignment artifacts.
While all the annotation tools, presented in the “variant annotation” section, can
be used to annotate somatic variants, a number of tools that provide cancer-specific
annotation in addition to general annotation are available. Two popular examples
are Oncotator [68] and CRAVAT [69]. Oncotator, a widely used cancer-specific
annotation tool, is often preferred for the annotation of somatic short variants.
Oncotator provides variant- and gene-centric information relevant to cancer
researchers, utilizing resources including but not limited to the Catalogue of
Somatic Mutations in Cancer (COSMIC) [70], the Cancer Gene Census [71], Cancer
Cell Line Encyclopedia [72], The Cancer Genome Atlas (TCGA), and Familial
Cancer Database [73].
2.9 Structural variations
So far, we focused only on the discovery of small-scale genomic variations

(SNVs and indels). There also exist large-scale (1 kb and larger) genomic variations,
which either be copy number variations (CNV) or chromosomal rearrangement
events (including translocations, inversions, and duplications).
2.9.1 Copy number variations
CNV is a frequent form of critical genetic variation that results in an abnormal

number of copies of large genomic regions (either gain or loss events). CNV is
clinically relevant, as they may play vital roles in disease processes, especially
during oncogenesis. It is possible to detect CNVs using WES/WGS data. Several
different approaches exist for this purpose [74]:
i. Paired-end mapping strategy detects CNVs through discordantly mapped

reads. Tools utilizing this approach include BreakDancer, PEMer,
VariationHunter, commonLAW, GASV, and Spanner.
ii. Split read-based methods use the incompletely mapped read from each read
pair to identify small CNVs. Split read-based tools include AGE, Pindel,
SLOPE, and SRiC.
iii. Read depth-based approach detects CNV by counting the number of reads
mapped to each genomic region. Tools using this approach include GATK,
SegSeq, CNV-seq, RDXplorer, BIC-seq, CNAseq, cn.MOPS, jointSLM,
ReadDepth, rSW-seq, CNVnator, CNVnorm, CMDS, mrCaNaVar, CNVeM,
and cnvHMM.
iv. Assembly-based approach detects CNVs by mapping contigs to the reference

genome. Tools using this approach include Magnolya, Cortex assembler, and
TIGRA-SV.
v. Combinatorial approach combines read depth and paired-end mapping

information to detect CNVs. Tools using this approach include NovelSeq,
HYDRA, CNVer, GASVPro, Genome STRiP, SVDetect, inGAP-sv, and SVseq.
In addition to the noise and artifacts caused by WES/WGS, tumor complexity

(the strongest factor being tumor heterogeneity) makes the detection of somatic
23
CNVs more challenging. To overcome this challenge, numerous tools have been
developed. Widely used tools for detecting specifically somatic CNVs include
ADTEx [75], CONTRA [76], cn.MOPS [77], ExomeCNV [78], VarScan2 [62],
SynthEx [79], Control-FREEC [80], GATK [13], and CloneCNA [81].
2.9.2 Chromosomal rearrangements
Chromosomal rearrangements are variations in chromosome structure whose

impact on genetic diversity and disease susceptibility has become increasingly evi-
dent [82]. Per SO, numerous types of rearrangements exist: duplication, deletion,
insertion, mobile element insertion, novel sequence insertion, tandem duplication,
inversion, intrachromosomal breakpoint, interchromosomal breakpoint, transloca-
tion, and complex SVs. Similar to CNV detection, there are multiple approaches for
rearrangement detection: read-pair, split-read, read-depth, and assembly
approaches. The underlying aims of each of these approaches are very similar to
those for CNV detection. As for CNV detection, countless tools have been devel-
oped for the detection of rearrangement variations, including but not limited to
Breakdancer [83], GRIMM [84], LUMPY [85], BreaKmer [86], BreakSeek [87],
CREST [88], DELLY [89], HYDRA [90], MultiBreak-SV [91], Pindel [92],
SoftSearch [93], SVdetect [94], and TIGRA-SV [95].
Lastly, we would like to point out that long reads (enabled by the emergence of
so-called third-generation sequencing technologies) allow for more accurate and
reliable determination of SVs with the development of novel algorithms that spe-
cifically exploits these long reads [96].
3. Clinical interpretation of genomic variations
Perhaps the most challenging process in WES/WGS analysis is the clinical inter-
pretation of genomic variations. While WES/WGS is rapidly becoming a routine
approach for the diagnosis of monogenic and complex disorders and personalized
treatment of such disorders, it is still challenging to interpret the vast amount of
genomic variation data detected through WES/WGS [97].
There exist numerous standardized widely accepted guidelines for the evalua-
tion of genomic variations obtained through NGS such as the American College of
Medical Genetics and Genomics (ACMG), the EuroGentest, and the European
Society of Human Genetics guidelines. These provide standards and guidelines for
the interpretation of genomic variations and include evidence-based recommenda-
tions on aspects including the use of literature and database and the use of in silico
predictors, criteria for variant interpretation, and reporting.
In addition to variant-dependent annotation such as allele frequency (e.g., in
1000 Genomes [29], ExAc [30], gnomAD [31]), the predicted effect on protein and
evolutionary conservation, disease-dependent inquiries such as mode of inheri-
tance, co-segregation of variant with disease within families, prior association of the
variant/gene with disease, investigation of clinical actionability, and pathway-based
analysis are required for the interpretation of genomic variants.
Databases such as ClinVar [33], HGV databases [98], OMIM [99], COSMIC
[100], and CIViC [101] are excellent resources that can aid interpretations of
clinical significance of germline and somatic variants for reported conditions. The
availability of shared genetic data in such databases makes it possible to identify
patients with similar conditions and aid the clinician to make a conclusive diagnosis.
While one may perform interpretation of genomic variations completely manu-
ally after annotation and filtering of variants, there are several tools to aid in
24
interpretation and prioritization of these variants, including Ingenuity Variant Anal-

ysis [102], BaseSpace Variant Interpreter [103], VariantStudio [104], KGGSeq [105],
PhenoTips [106], VarElect [107], PhenoVar [108], InterVar [109], VarSifter [110],
eXtasy [111], VAAST [57], and Exomiser [112]. For personalized oncology purposes,
numerous cancer-specific tools exist, specifically developed to determine driver
genes/mutations as well as to aid in interpretation of somatic variants. Some of the
most widely used somatic interpretation tools are PHIAL, PCGR, and HitWalker2.
Pathway analysis is another powerful component that can enhance the interpre-
tation of genomic variations. Pathway analysis can be broadly defined as a group of
methods incorporating biological information from public databases to simplify
analysis by grouping long lists of genes into smaller sets of related genes (for a
comprehensive review on pathway analysis, see [113]). Pathway analysis improves
the detection of causal variants by incorporating biologic insight. The clinician can
gain a better understanding of the functions of rare genetic variants of unknown
clinical significance in the context of biological pathways. While the gene carrying
the variant may not be related to the phenotype, its associated genes in the pathway
might be causally related to the phenotype at hand. Moreover, through pathway
analysis, the role of multiple variants and their interaction on disease formation can
be discovered.
Countless tools for pathway analysis exist. Some of the widely used pathway
analysis tools are GSEA [114], DAVID [115], IPA [116], SPIA [117], pathfindR
[118], enrichr [119], reactomePA [120], MetaCore [121], and PathVisio [122].
Additionally, many different pathway resources exist, the most popular of which
are Kyoto Encyclopedia of Genes and Genomes [123], Reactome [124],
WikiPathways [125], MSigDb [126], STRINGDB [127], Pathway Commons [128],
Ingenuity Knowledge Base [129], and Pathway Studio [130].
In silico interpretation often fails to provide conclusive evidence for pathoge-
nicity of genomic variations [131]. Furthermore, these in silico interpretations are
mostly only well-supported predictions (this is especially true for VUS). It is there-
fore vital to perform functional validation to understand the functional conse-
quences of genetic variants, provide a conclusive diagnosis, and inform the patient
on the disease course. Functional validation can be performed using different model
systems (e.g., patient cells, model cell lines, model organisms, induced pluripotent
stem cells) and performing the suitable type of assay (e.g., genetic rescue,
overexpression, biomarker analysis).
4. Conclusion
The advancements in NGS, the increasing availability and applicability of WES/

WGS analysis due to decrease in cost, and the development of countless bioinfor-
matics methods and resources enabled the usage of WES/WGS to detect, interpret,
and validate genomic variations in the clinical setting.
As we attempted to describe in this chapter, WES/WGS analysis is challenging,
and there are a great number of tools for each step of variation discovery. There-
fore, one must carefully evaluate the advantages and disadvantages and suitability
of different tools (depending on the specific application) before adapting the “opti-
mal” one into the variation discovery workflow. An optimal and coordinated com-
bination of tools is required to identify the different types of genomic variants,
described here. On the one hand, an efficient analysis strategy needs to adopt one or
more methods for the detection of each type of variant and, on the other hand,
needs to integrate results for the different types of variants into a single compre-
hensive solution.
25
We attempted to describe the best practices for variant discovery, outlining the
fundamental aspects. We hope to have provided a basic understanding of WES/
WGS analysis as we believe awareness of the steps involved as well as the challenges
involved at each step is important to understand how each piece may affect the
downstream steps (and eventually affect interpretation). As emphasized through-
out the chapter, substantial (or even minor) changes at any step can fundamentally
alter the outcomes in the later stages.
While there is no definite gold standard for the interpretation of genomic vari-
ations, we attempted to briefly describe the currently available and widely used
guidelines, tools, and resources for clinical evaluation of genomic variations.
In the following years, with the advancements in bioinformatics, increasing
cooperation between the clinician and bioinformatician and large-scale efforts
(such as IRDiRC [132], TCGA [133], and ICGC [134]), we expect that a greater
focus will be on developing novel tools for clinical interpretation of genomic varia-
tions. Cooperation between multiple disciplines is vital to improve the existing
approaches as well as to develop novel approaches and resources.
Author details
Osman Ugur Sezerman*, Ege Ulgen, Nogayhan Seymen and Ilknur Melis Durasi
Department of Biostatistics and Medical Informatics, Acibadem Mehmet Ali
Aydinlar University, Istanbul, Turkey
26
References
[1] Bolger AM, Lohse M, Usadel B. [13] Mckenna A, Hanna M, Banks E,

Trimmomatic: A flexible trimmer for et al. The genome analysis toolkit: A
illumina sequence data. Bioinformatics. MapReduce framework for analyzing
2014;30(15):2114-2120 next-generation DNA sequencing data.
Genome Research. 2010;20(9):
[2] Martin M. Cutadapt removes adapter 1297-1303
sequences from high-throughput
sequencing reads. EMBnet.Journal. [14] Sandmann S, De graaf AO, Karimi
2011;17(1):10-12 M, et al. Evaluating variant calling tools
for non-matched next-generation
[3] Criscuolo A, Brisse S. AlienTrimmer:
sequencing data. Scientific Reports.
A tool to quickly and accurately trim off
2017;7:43169
multiple short contaminant sequences
from high-throughput sequencing
reads. Genomics. 2013;102(5-6): [15] Bian X, Zhu B, Wang M, et al.
500-506 Comparing the performance of selected
variant callers using synthetic data and
[4] Jiang H, Lei R, Ding SW, Zhu S. genome segmentation. BMC
Skewer: A fast and accurate adapter Bioinformatics. 2018;19(1):429
trimmer for next-generation sequencing
paired-end reads. BMC Bioinformatics. [16] Xu C. A review of somatic single
2014;15:182 nucleotide variant calling algorithms for
next-generation sequencing data.
[5] Available from: https://jgi.doe.gov/da Computational and Structural
ta-and-tools/bbtools/ Biotechnology Journal. 2018;16:15-24.
DOI: 10.1016/j.csbj.2018.01.003
[6] Available from: http://hannonlab.csh
l.edu/fastx_toolkit/ [17] Li R, Li Y, Fang X, et al. SNP
detection for massively parallel whole-
[7] Available from: https://github.com/ genome resequencing. Genome
FelixKrueger/TrimGalore Research. 2009;19(6):1124-1132
[8] Li H, Durbin R. Fast and accurate
[18] Li H, Handsaker B, Wysoker A,
short read alignment with Burrows-
et al. The sequence alignment/map
Wheeler transform. Bioinformatics.
format and SAMtools. Bioinformatics.
2009;25(14):1754-1760
2009;25(16):2078-2079
[9] Langmead B, Salzberg SL. Fast
gapped-read alignment with Bowtie 2. [19] Saunders CT, Wong WS, Swamy S,
Nature Methods. 2012;9(4):357-359 Becq J, Murray LJ, Cheetham RK.
Strelka: Accurate somatic small-variant
[10] Available from: http://novocraft. calling from sequenced tumor-normal
com/ sample pairs. Bioinformatics. 2012;
28(14):1811-1817
[11] Marçais G, Delcher AL, Phillippy
AM, Coston R, Salzberg SL, Zimin A. [20] Available from: https://arxiv.org/ab
MUMmer4: A fast and versatile genome s/1207.3907
alignment system. PLoS Computational
Biology. 2018;14(1):e1005944 [21] Rimmer A, Phan H, Mathieson I,
et al. Integrating mapping-, assembly-
[12] Available from: http://broadinstitute and haplotype-based approaches for
.github.io/picard/ calling variants in clinical sequencing
27
applications. Nature Genetics. 2014; mutations in cancer. Nucleic Acids

46(8):912-918 Research. 2019;47(D1):D941-D947
[22] Poplin R, Chang P-C, Alexander D, [33] Landrum MJ, Lee JM, Riley GR,
et al. A universal SNP and small-indel et al. ClinVar: Public archive of
variant caller using deep neural relationships among sequence variation
networks. Nature Biotechnology. 2018; and human phenotype. Nucleic Acids
36(10):983-987. DOI: 10.1038/nbt.4235 Research. 2014;42(Database issue):
D980-D985
[23] Bao R, Huang L, Andrade J, et al.
Review of current methods, [34] Ng PC, Henikoff S. SIFT: Predicting
applications, and data management for amino acid changes that affect protein
the bioinformatics analysis of whole function. Nucleic Acids Research. 2003;
exome sequencing. Cancer Informatics. 31(13):3812-3814
2014;13(Suppl 2):67-82
[35] Adzhubei IA, Schmidt S, Peshkin L,
[24] Quinlan AR, Hall IM. BEDTools: A et al. A method and server for predicting
flexible suite of utilities for comparing damaging missense mutations. Nature
genomic features. Bioinformatics. 2010; Methods. 2010;7(4):248-249
26(6):841-842
[36] Chun S, Fay JC. Identification of
[25] Tarasov A, Vilella AJ, Cuppen E, deleterious mutations within three
Nijman IJ, Prins P. Sambamba: Fast human genomes. Genome Research.
processing of NGS alignment formats. 2009;19(9):1553-1561
Bioinformatics. 2015;31(12):2032-2034
[37] Schwarz JM, Rödelsperger C,
[26] Available from: http://gmt.genome. Schuelke M, Seelow D. MutationTaster
wustl.edu/gmt-refcov evaluates disease-causing potential of
sequence alterations. Nature Methods.
[27] Robinson JT, Thorvaldsdóttir H, 2010;7:575-576
Winckler W, et al. Integrative genomics
viewer. Nature Biotechnology. 2011; [38] Reva B, Antipin Y, Sander C.
29(1):24-26 Predicting the functional impact of
protein mutations: Application to cancer
[28] Sherry ST, Ward MH, Kholodov M, genomics. Nucleic Acids Research. 2011;
et al. dbSNP: The NCBI database of 39(17):e118
genetic variation. Nucleic Acids
Research. 2001;29(1):308-311 [39] Shihab HA, Gough J, Cooper DN,
et al. Predicting the functional,
[29] Auton A, Brooks LD, Durbin RM, molecular, and phenotypic
et al. A global reference for human consequences of amino acid
genetic variation. Nature. 2015; substitutions using hidden Markov
526(7571):68-74 models. Human Mutation. 2013;34(1):
57-65
[30] Lek M, Karczewski KJ, Minikel EV,
et al. Analysis of protein-coding genetic [40] Davydov EV, Goode DL, Sirota M,
variation in 60,706 humans. Nature. Cooper GM, Sidow A, Batzoglou S.
2016;536(7616):285-291 Identifying a high fraction of the human
genome to be under selective constraint
[31] Available from: https://gnomad.b using GERP++. PLoS Computational
roadinstitute.org/ Biology. 2010;6(12):e1001025
[32] Tate JG, Bamford S, Jubb HC, et al. [41] Pollard KS, Hubisz MJ, Rosenbloom
COSMIC: The catalogue of somatic KR, Siepel A. Detection of nonneutral
28
substitution rates on mammalian taxonomic expansion, and functional

phylogenies. Genome Research. 2010; annotation. Nucleic Acids Research.
20(1):110-121 2016;44(D1):D733-D745
[42] Garber M, Guttman M, Clamp M, [50] Zerbino DR, Achuthan P, Akanni

Zody MC, Friedman N, Xie X. W, et al. Ensembl 2018. Nucleic Acids
Identifying novel constrained elements Research. 2018;46(D1):D754-D761
by exploiting biased substitution
patterns. Bioinformatics. 2009;25(12): [51] Mccarthy DJ, Humburg P, Kanapin
i54-i62 A, et al. Choice of transcripts and
software has a large effect on variant
[43] Tang H, Thomas PD. PANTHER- annotation. Genome Medicine. 2014;
PSEP: Predicting disease-causing genetic 6(3):26
variants using position-specific
evolutionary preservation. [52] Wang K, Li M, Hakonarson H.
Bioinformatics. 2016;32(14):2230-2232 ANNOVAR: Functional annotation of
genetic variants from next-generation
[44] González-pérez A, López-bigas N. sequencing data. Nucleic Acids
Improving the assessment of the Research. 2010;38:e164
outcome of nonsynonymous SNVs with
a consensus deleteriousness score, [53] Cingolani P, Platts A, Wang le L,
Condel. American Journal of Human et al. A program for annotating and
Genetics. 2011;88(4):440-449 predicting the effects of single
nucleotide polymorphisms, SnpEff:
[45] Rentzsch P, Witten D, Cooper GM, SNPs in the genome of Drosophila
Shendure J, Kircher M. CADD: melanogaster strain w1118; iso-2; iso-3.
Predicting the deleteriousness of Fly (Austin). 2012;6(2):80-92
variants throughout the human genome.
Nucleic Acids Research. 2019;47(D1): [54] Mclaren W, Gil L, Hunt SE, et al.
D886-D894 The Ensembl variant effect predictor.
Genome Biology. 2016;17(1):122
[46] Wong WC, Kim D, Carter H,
Diekhans M, Ryan MC, Karchin R. [55] Paila U, Chapman BA, Kirchner R,
CHASM and SNVBox: Toolkit for Quinlan AR. GEMINI: Integrative
detecting biologically important single exploration of genetic variation and
nucleotide mutations in cancer. genome annotations. PLoS
Bioinformatics. 2011;27(15):2147-2148 Computational Biology. 2013;9(7):
e1003153
[47] Mao Y, Chen H, Liang H, Meric-
bernstam F, Mills GB, Chen K. CanDrA: [56] Desvignes JP, Bartoli M, Delague V,
Cancer-specific driver missense et al. VarAFT: A variant annotation and
mutation annotation with optimized filtration system for human next
features. PLoS One. 2013;8(10):e77945 generation sequencing data. Nucleic
Acids Research. 2018;46(W1):W545-
[48] Carter H, Douville C, Yeo G, W553
Stenson PD, Cooper DN, Karchin R.
Identifying Mendelian disease genes [57] Hu H, Huff CD, Moore B, Flygare S,
with the variant effect scoring tool. Reese MG, Yandell M. VAAST 2.0:
BMC Genomics. 2013;14(3):1-16 Improved variant classification and
disease-gene identification using a
[49] O'leary NA, Wright MW, Brister JR, conservation-controlled amino acid
et al. Reference sequence (RefSeq) substitution matrix. Genetic
database at NCBI: Current status, Epidemiology. 2013;37(6):622-634
29
[58] Zhou W, Chen T, Chong Z, et al. algorithm for cancer genome

TransVar: A multilevel variant sequencing data. Bioinformatics. 2014;
annotator for precision genomics. 30(17):2498-2500
Nature Methods. 2015;12(11):1002-1003
[68] Ramos AH, Lichtenstein L, Gupta
[59] Leiserson MD, Gramazio CC, Hu J, M, Lawrence MS, Pugh TJ, Saksena G,
Wu HT, Laidlaw DH, Raphael BJ. et al. Oncotator: Cancer variant
MAGI: Visualization and collaborative annotation tool. Human Mutation. 2015;
annotation of genomic aberrations. 36(4):E2423-E24E9. pmid: 25703262
Nature Methods. 2015;12(6):483-484
[69] Douville C, Carter H, Kim R, et al.
[60] Dayem Ullah AZ, Oscanoa J, Wang CRAVAT: Cancer-related analysis of
J, Nagano A, Lemoine NR, Chelala C. variants toolkit. Bioinformatics. 2013;
SNPnexus: Assessing the functional 29(5):647-648
relevance of genetic variation to
facilitate the promise of precision [70] Forbes SA, Beare D, Boutselakis H,
medicine. Nucleic Acids Research. 2018; et al. COSMIC: Somatic cancer genetics
46(W1):W109-W113 at high-resolution. Nucleic Acids
Research. 2016;45(D1):D777-D783
[61] Sun C, Medvedev P. VarMatch:
Robust matching of small variant [71] Futreal PA, Andrew Futreal P, Coin
datasets using flexible scoring schemes. L, Marshall M, Down T, Hubbard T,
Bioinformatics. 2017;33(9):1301-1308 et al. A census of human cancer genes.
Nature Reviews. Cancer. 2004;4:177-183
[62] Koboldt DC, Zhang Q , Larson DE,
et al. VarScan 2: Somatic mutation and [72] Barretina J, Caponigro G,
copy number alteration discovery in Stransky N, et al. The cancer cell line
cancer by exome sequencing. Genome encyclopedia enables predictive
Research. 2012;22(3):568-576 modelling of anticancer drug sensitivity.
Nature. 2012;483(7391):603-607.
[63] Larson DE, Harris CC, Chen K, et al. Published 2012 Mar 28. DOI: 10.1038/
SomaticSniper: Identification of somatic nature11003
point mutations in whole genome
sequencing data. Bioinformatics. 2012; [73] Sijmons RH. Identifying Patients
28(3):311-317 with Familial Cancer Syndromes. 2010
Feb 27 [Updated 2010 Feb 27]. In:
[64] Fang LT, Afshar PT, Chhibber A, Riegert-Johnson DL, Boardman LA,
Mohiyuddin M, Fan Y, Mu JC, et al. An Hefferon T, et al., editors. Cancer
ensemble approach to accurately detect Syndromes [Internet]. Bethesda (MD):
somatic mutations using SomaticSeq. National Center for Biotechnology
Genome Biology. 2015;16:197 Information (US); 2009. Available
from: https://www.ncbi.nlm.nih.gov/
[65] Available from: http://arxiv.org/abs/ books/NBK45295/
1207.3907
[74] Zhao M, Wang Q , Wang Q , Jia P,
[66] Stephens PJ et al. The landscape of Zhao Z. Computational tools for copy
cancer genes and mutational processes number variation (CNV) detection
in breast cancer. Nature. 2012;486: using next-generation sequencing data:
400-404 Features and perspectives. BMC
Bioinformatics. 2013;14(11):S1
[67] Wang W, Wang P, Xu F, Luo R,
Wong MP, Lam T-W. FaSD-somatic: A [75] Amarasinghe KC, Li J, Hunter SM,
fast and accurate somatic SNV detection et al. Inferring copy number and
30
genotype in tumour exome data. BMC Protocols in Bioinformatics. 2014;45:

Genomics. 2014;15:732 15.6.1-15.611
[76] Hooghe B, Hulpiau P, van Roy F, De [84] Tesler G. GRIMM: Genome

Bleser P. ConTra: A promoter alignment rearrangements web server.
analysis tool for identification of Bioinformatics. 2002;18(3):492-493
transcription factor binding sites across
species. Nucleic Acids Research. 2008; [85] Layer RM, Chiang C, Quinlan AR,
36:W128-W132 Hall IM. LUMPY: A probabilistic
framework for structural variant
[77] Klambauer G, Schwarzbauer K, discovery. Genome Biology. 2014;15(6):
Mayr A, et al. cn.MOPS: Mixture of R84
Poissons for discovering copy number
variations in next-generation [86] Abo RP, Ducar M, Garcia EP, et al.
sequencing data with a low false BreaKmer: Detection of structural
discovery rate. Nucleic Acids Research. variation in targeted massively parallel
2012;40(9):e69 sequencing data using kmers. Nucleic
Acids Research. 2014;43(3):e19
[78] Sathirapongsasuti JF, Lee H, Horst
BA, et al. Exome sequencing-based [87] Zhao H, Zhao F. BreakSeek: A
copy-number variation and loss of breakpoint-based algorithm for full
heterozygosity detection: ExomeCNV. spectral range INDEL detection. Nucleic
Bioinformatics. 2011;27(19):2648-2654 Acids Research. 2015;43(14):6701-6713
[79] Silva GO, Siegel MB, Mose LE, et al.
[88] Wang J, Mullighan CG, Easton J,
SynthEx: A synthetic-normal-based
et al. CREST maps somatic structural
DNA sequencing tool for copy number
variation in cancer genomes with base-
alteration detection and tumor
pair resolution. Nature Methods. 2011;
heterogeneity profiling. Genome
8(8):652-654. Published 2011 Jun 12.
Biology. 2017;18(1):66
DOI: 10.1038/nmeth.1628
[80] Boeva V, Popova T, Bleakley K,
et al. Control-FREEC: A tool for [89] Rausch T, Zichner T, Schlattl A,
assessing copy number and allelic Stütz AM, Benes V, Korbel JO. DELLY:
content using next-generation Structural variant discovery by
sequencing data. Bioinformatics. 2012; integrated paired-end and split-read
28(3):423-425 analysis. Bioinformatics. 2012;28(18):
i333-i339
[81] Yu Z, Li A, Wang M. CloneCNA:
Detecting subclonal somatic copy [90] Miller PL, Blumenfrucht SJ, Rose
number alterations in heterogeneous JR, et al. HYDRA: A knowledge
tumor samples from whole-exome acquisition tool for expert systems that
sequencing data. BMC Bioinformatics. critique medical workup. Medical
2016;17:310 Decision Making. 1987;7(1):12-21
[82] Sedlazeck FJ, Dhroso A, Bodian DL, [91] Ritz A, Bashir A, Sindi S, Hsu D,
Paschall J, Hermes F, Zook JM. Tools for Hajirasouliha I, Raphael BJ.
annotation and comparison of structural Characterization of structural variants
variation. F1000Res. 2017;6:1795 with single molecule and hybrid
sequencing approaches. Bioinformatics.
[83] Fan X, Abbott TE, Larson D, Chen 2014;30(24):3458-3466
K. BreakDancer: Identification of
genomic structural variation from [92] Ye K, Schulz MH, Long Q , Apweiler
paired-end read mapping. Current R, Ning Z. Pindel: A pattern growth
31
approach to detect break points of large [101] Griffith M, Spies NC, Krysiak K,
deletions and medium sized insertions et al. CIViC is a community
from paired-end short reads. knowledgebase for expert
Bioinformatics. 2009;25(21):2865-2871 crowdsourcing the clinical
interpretation of variants in cancer.
[93] Hart SN, Sarangi V, Moore R, et al. Nature Genetics. 2017;49(2):170-174
SoftSearch: Integration of multiple
sequence features to identify [102] Available from: www.ingenuity.c
breakpoints of structural variations. om
PLoS One. 2013;8(12):e83356
[103] Available from: https://www.
[94] Zeitouni B, Boeva V, Janoueix- illumina.com/products/by-type/inf
Lerosey I, et al. SVDetect: A tool to ormatics-products/basespace-variant-
identify genomic structural variations interpreter.html
from paired-end and mate-pair
[104] Available from: https://www.bioz.c
sequencing data. Bioinformatics. 2010;
om/result/VariantStudio%20variant/
26(15):1895-1896
product/Illumina
[95] Chen K, Chen L, Fan X, Wallis J, [105] Li MX, Gui HS, Kwan JS, Bao SY,
Ding L, Weinstock G. TIGRA: A Sham PC. A comprehensive framework
targeted iterative graph routing for prioritizing variants in exome
assembler for breakpoint assembly. sequencing studies of Mendelian
Genome Research. 2014;24(2):310-317 diseases. Nucleic Acids Research. 2012;
40(7):e53
[96] Guan P, Sung WK. Structural
variation detection using next- [106] Girdea M, Dumitriu S, Fiume M,
generation sequencing data: A et al. PhenoTips: Patient phenotyping
comparative technical review. Methods. software for clinical and research use.
2016;102:36-49 Human Mutation. 2013;34(8):1057-1065
[97] Sayitoğlu M. Clinical interpretation [107] Stelzer G, Plaschkes I, Oz-levi D,

of genomic variations. Turkish Journal et al. VarElect: The phenotype-based
of Haematology. 2016;33(3):172-179 variation prioritizer of the GeneCards
Suite. BMC Genomics. 2016;17(Suppl
[98] Freeman PJ, Hart RK, Gretton LJ, 2):444
Brookes AJ, Dalgleish R.
VariantValidator: Accurate validation, [108] Trakadis YJ, Buote C, Therriault JF,
mapping, and formatting of sequence Jacques PÉ, Larochelle H, Lévesque S.
variation descriptions. Human PhenoVar: A phenotype-driven
Mutation. 2017;39(1):61-68 approach in clinical genomics for the
diagnosis of polymalformative
[99] Hamosh A, Scott AF, Amberger J, syndromes. BMC Medical Genomics.
Bocchini C, Valle D, McKusick VA. 2014;7:22. Published 2014 May 12. DOI:
Online Mendelian Inheritance in Man 10.1186/1755-8794-7-22
(OMIM), a knowledgebase of human
genes and genetic disorders. Nucleic [109] Li Q , Wang K. InterVar: Clinical
Acids Research. 2002;30(1):52-55 interpretation of genetic variants by the
2015 ACMG-AMP guidelines. American
[100] Bamford S, Dawson E, Forbes S, Journal of Human Genetics. 2017;
et al. The COSMIC (Catalogue of 100(2):267-280
Somatic Mutations in Cancer) database
and website. British Journal of Cancer. [110] Teer JK, Green ED, Mullikin JC,
2004;91(2):355-358 Biesecker LG. VarSifter: Visualizing and
32
analyzing exome-scale sequence [119] Kuleshov MV, Jones MR, Rouillard

variation data on a desktop computer. AD, et al. Enrichr: A comprehensive
Bioinformatics. 2011;28(4):599-600 gene set enrichment analysis web server
2016 update. Nucleic Acids Research.
[111] Sifrim A, Popovic D, Tranchevent 2016;44(W1):W90-W97
LC, Ardeshirdavani A, Sakai R, Konings P,
et al. eXtasy: Variant prioritization by [120] Yu G, HeReactomePA Q-Y. An R/
genomic data fusion. Nature Methods. Bioconductor package for reactome
2013;10(11):1083-1084. DOI: 10.1038/ pathway analysis and visualization.
nmeth.2656 Molecular BioSystems. 2016;12:477-479
[112] Smedley D, Jacobsen JO, Jäger M,

[121] Available from: https://clarivate.c
et al. Next-generation diagnostics and om/products/metacore/
disease-gene discovery with the
Exomiser. Nature Protocols. 2015;
[122] Kutmon M, Van iersel MP, Bohler
10(12):2004-2015
A, et al. PathVisio 3: An extendable
[113] García-Campos MA, Espinal- pathway analysis toolbox. PLoS
Enríquez J, Hernández-Lemus E. Computational Biology. 2015;11(2):
Pathway analysis: State of the art. e1004085
Frontiers in Physiology. 2015;6:383. doi:
10.3389/fphys.2015.00383 [123] Kanehisa M, Goto S. KEGG: Kyoto
encyclopedia of genes and genomes.
[114] Powers RK, Goodspeed A, Pielke- Nucleic Acids Research. 2000;28(1):
Lombardo H, Tan AC, Costello JC. 27-30
GSEA-InContext: Identifying novel and
common patterns in expression [124] Croft D, O'Kelly G, Wu G, et al.
experiments. Bioinformatics. 2018; Reactome: A database of reactions,
34(13):i555-i564 pathways and biological processes.
Nucleic Acids Research. 2010;39
[115] Dennis G, Sherman BT, Hosack
(Database issue):D691-D697
DA, et al. DAVID: Database for
annotation, visualization, and integrated
[125] Kelder T, van Iersel MP, Hanspers
discovery. Genome Biology. 2003;4(9):
K, et al. WikiPathways: Building
R60
research communities on biological
[116] Yu J, Gu X, Yi S. Ingenuity
pathways. Nucleic Acids Research. 2011;
pathway analysis of gene expression 40(Database issue):D1301-D1307
profiles in distal Nerve stump following
nerve injury: Insights into Wallerian [126] Liberzon A, Subramanian A,
degeneration. Frontiers in Cellular Pinchback R, Thorvaldsdóttir H,
Neuroscience. 2016;10:274. Published Tamayo P, Mesirov JP. Molecular
2016 Dec 6. DOI: 10.3389/ signatures database (MSigDB) 3.0.
fncel.2016.00274 Bioinformatics. 2011;27(12):1739-1740
[117] Tarca AL, Draghici S, Khatri P, [127] Szklarczyk D, Morris JH, Cook H,
et al. A novel signaling pathway impact et al. The STRING database in 2017:
analysis. Bioinformatics. 2008;25(1): Quality-controlled protein-protein
75-82 association networks, made broadly
accessible. Nucleic Acids Research.
[118] Ulgen E, Ozisik O, Sezerman OU. 2016;45(D1):D362-D368
pathfindR: An R Package for Pathway
Enrichment Analysis Utilizing Active [128] Cerami EG, Gross BE, Demir E,
Subnetworks. bioRxiv. 2018 et al. Pathway commons, a web resource
33
for biological pathway data. Nucleic

Acids Research. 2010;39(Database
issue):D685-D690
[129] Available from: http://www.inge

nuity.com/
[130] Nikitin A, Egorov S, Daraselia N,

Mazo I. Pathway studio—The analysis
and navigation of molecular networks.
Bioinformatics. 2003;19(16):2155-2157
[131] Rodenburg RJ. The functional

genomics laboratory: Functional
validation of genetic variants. Journal of
Inherited Metabolic Disease. 2018;
41(3):297-307
[132] Austin CP, Cutillo CM, Lau LPL,

et al. Future of rare diseases research
2017–2027: An IRDiRC perspective.
Clinical and Translational Science. 2018;
11(1):21-27
[133] Available from: https://cancerge

nome.nih.gov/
[134] Zhang J, Baran J, Cros A, et al.

International cancer genome
consortium data portal—A one-stop
shop for cancer genomics data.
Database: The Journal of Biological
Databases and Curation. 2011;2011:
bar026
34
Chapter 3
Orienting Future Trends in Local

Ancestry Deconvolution Models to
Optimally Decipher Admixed
Individual Genome Variations
Gaston K. Mazandu, Ephifania Geza, Milaine Seuneu
and Emile R. Chimusa
Abstract
Rapid advances in sequencing and genotyping technologies have significantly

contributed to shaping the area of medical and population genetics. Several thou-
sand genomes are completed with millions of variants identified in the human
deoxyribonucleic acid (DNA) sequences. These genomic variations highly influence
changes in phenotypic manifestations and physiological functions of different indi-
viduals or population groups. Of particular importance are variations introduced by
admixture event, contributing significantly to a remarkable phenotypic variability
with medical and/or evolutionary implications. In this case, knowledge of local
ancestry estimates and date of admixture is of utmost importance for a better
understanding of genomic variation patterns throughout modern human evolution
and adaptive processes. In this chapter, we survey existing local ancestry
deconvolution and dating admixture event models to identify possible gaps that still
need to be filled and orient future trends in designing more effective models, which
account for current challenges and produce more accurate and biological relevant
estimates.
Keywords: genomic variations, admixture, local ancestry, dating admixture event,

linkage disequilibrium
1. Introduction
Today, advances in high-throughput technologies have generated huge amounts

of human genomics data in public domains. These data are useful for medical and
population genetics to understand the population history, human evolution and
demographics, susceptibility to disease, and response to drug. Over time, humanity
has experienced the exchange of genetic materials across populations, mainly due to
population migrations [1], which have led to wide human genetic variations as
results of interbreeding or mating between different populations previously iso-
lated. These genetic variations observed in the human deoxyribonucleic acid (DNA)
sequences are caused by inheritance processes, such as mutation and recombina-
tion. Generally, the mating process yields the genetic recombination break points,
35
introduces some variations, and creates mixed DNA segments. As a consequence,

current human populations are admixed [2, 3] with specific genomes displaying a
mosaic of segments originating from different ancestral populations [1, 2, 4], wide
phenotypic variations, divergent genetic ancestry, and different traits observed
among individuals in worldwide population groups. Thus, it is critical to understand
the dynamics related to the origin of these variations, the evolution process, and its
consequences in human heredity and health.
Studying admixture patterns in human populations consists of characterization
of admixture features in human populations, including admixture mapping and
date to admixture events. Admixture mapping combines both the identification of
genetic variants underlying the ethnic difference in disease risk and inference of
ancestry estimates associated with these genetic variants. Estimation of ancestry is
commonly known as genetic ancestry inference, which is either global or local
ancestry inference. Global ancestry inference estimates the overall proportion con-
tributed by each ancestral population to the admixed genome; while, local ancestry
deconvolution (local ancestry inference) estimates the number of copies from a
particular population at a given site [5]. Together, admixture mapping and date to
admixture events provide a better understanding of the genetic variation features
throughout modern human evolution, the demographics, and adaptive processes of
human populations. Currently, analyzing admixture patterns has become central to
genomics research, contributing to a wide range of biomedical applications. Current
advance in technologies is facilitating the movement of people worldwide, thus
influencing the complexity of population admixture dynamics and leading to multi-
faceted admixture events. On the other hand, the determination of local ancestry
through genotyping and microarray datasets has empowered the approaches for
dating mutation, selection, and admixture events [6, 7].
The significance of the local ancestry inference topic is viewed through the
research interests it has raised over the last two decades. Several models exist for
local ancestry deconvolution, including ANCESTRYMAP [8], ADMIXMAP [9],
SABER [10], LAMP [11], LAMPLD/LAMPHAP [12], SUPPORTMIX [13], EILA
[14], LOTER [15], etc. Figure 1 displays the implementation dynamics of different
local ancestry deconvolution models graphically, indicating the time each model
was introduced. Local ancestry inference is relevant in personalizing medicines,
understanding complex diseases, localizing missing sequences in reference genomes
and understanding the population history and demographics. Subsequently, several
studies have particularly been focusing on dating past admixture events, relevant to
population migrations, heritable genes associated to some diseases, and responses to
Figure 1.
The evolution of local ancestry deconvolution since 2003 to 2017.
36
Orienting Future Trends in Local Ancestry Deconvolution Models to Optimally Decipher…
Figure 2.
A partial worldwide admixture painting map. The figure shows several worldwide admixed populations
with patterns identified through published paper on population structure from 2008 to 2018. The population
migrations within and between continents have resulted in different admixed populations ranging from one-
to five-way admixtures.
treatment [16]. The date of admixture in a given population can be predicted by

analyzing the ancestral track, break-points, and linkage disequilibrium (LD) [17].
Also, distinction between date of admixture events is made with the use of LD and
ancestral tracts in the admixed genomes [17]. Nowadays, there are several models
for predicting the age of an admixture event, which are classified into two main
groups: LD-based approaches and haplotype-based approaches [17, 18]. These
models use information from genomes of several population groups around the
world as representative or equivalent ancient populations known to influence the
migration and/or admixture processes, yielding observed admixed population pat-
terns worldwide (Figure 2).
In this chapter, we survey current models for deconvoluting local ancestry and
dating admixture events and explore computational techniques used in these
models. We highlight advances made so far in this genomic era and opportunities
behind these models and challenges or gaps that still need to be addressed. This
informs users and researchers on the current state of research, and orient future
trends in designing more effective models, which account for current challenges
and produce more accurate and biological relevant estimates. In the subsequent
sections, we provide an overview of existing methods used for inferring local
ancestry estimates and dating admixture events.
2. Overview of admixture feature inference models
In this section, we survey current models used to elucidate admixture patterns,

including local ancestry estimates (deconvolution) and dating admixture events.
These models assume that the T genotyped sites are biallelic and the genotype infor-
mation of the K reference candidate ancestral and admixed populations are considered
known. Ancestry at different sites or windows follows a Markov chain. Recombina-
tion is assumed to occur at every generation resulting in Poison recombination points
with a rate which depends on both the recombination rate, r, and number of genera-
tions since admixture, g, and individuals are independent of each other.
37
2.1 Local ancestry inference models
As pointed out previously, existing local ancestry inference models can be cate-
gorized into two main groups based on whether the model makes use of admixture/
background linkage disequilibrium (LD) or not.
2.1.1 LD-based models for local ancestry inference
LD-based models account for LD in local ancestry deconvolution, and due to the
importance of LD in disease mapping, the first local ancestry methods fall into this
category. They assume that ancestry along an admixed individual genome follows a
first order Markov chain. This means that the immediate past state captures all the
information on past states [19]. As a result, LD-based models assume that, at every
site, the observed admixed genotypes are generated by the unobserved ancestry,
and hence, Hidden Markov Model (HMM) and its extensions are used to infer the
unobserved (hidden) states. Thus, to deconvolute ancestry along the admixed
genome, these models have three model parameters, namely the initial, transition
and observation, or emission probability models. Due to uncertainty and the num-
ber of parameters involved, LD-based methods use Markov Chain Monte Carlo
(MCMC), forward-backward, or Viterbi algorithms to determine the hidden ances-
try sequence for a given individual. Falush et al. and Patterson et al. modeled
ancestry switch between ancestry populations at a given site, X t ∈f1; …; K g, by
PðX 1 ¼ kjq; rÞ ¼ qk , (1)

0
0
P X t ¼ kjX t�1 ¼ k ; q; r ¼ δ k ¼ k e�dt r þ 1 � e�dt r qk for 1 < t ≤ T (2)
representing the
first marker,
and the transition probability between consecutive
0
markers with δ k ¼ k is the indicator function and dt the genetic distance
between sites t and t � 1, above and qk the proportion of ancestry contributed by

candidate ancestral population k such that q ¼ q1 ; …; qk is a vector of ancestry
inherited from each ancestral population. On haploid data, the probability of a
recombination event is 1 � e�dt r , meaning that the probability of no recombination
is e�dt r [8, 20]. LD-based methods can be subdivided into admixture LD-based and
admixture and background LD methods. Note that admixture LD occurs when
ancestry at nearby markers is inherited together and background LD is the LD
within ancestral populations, and it depends highly on population history
(i.e, generated by genetic drift and population bottlenecks).
2.1.1.1 Admixture LD-based models
Admixture LD-based methods are models that account for LD that resulted from
the admixture process. They do not model background LD. Admixture LD-based
methods include the early methods, for example, STRUCTURE V2 [20],
ANCESTRYMAP [8], and ADMIXMAP [9], which are based on the Bayesian
framework. Early methods rely on markers that show significant difference in
frequency between ancestral populations (AIMs). Admixture LD-based models
assume that markers are independent and the global and ancestral allele frequencies
are known. They integrate HMM with MCMC, and their switch model and initial
and transition models are as in Eqs. (1) and (2), respectively. Since LD-based
methods do not model background LD, their observation model depends on only
38
the allele frequency of the ancestry at that site. For instance, assuming K = 2,
Patterson et al. defined the emission probability by
8 !
>
> 2 � �2�na
>
>
> pnka 1 � pk if na ¼ 0 or 1
>
> na
>
<0 � �� 1
PðY t ¼ yjX t ¼ na Þ ¼ 2 1 � p1 1 � p2 (3)
>
> B � � � �C
> B
> p 1�p þp 1�p C
>
> @ 2 2 1 2 A if na ¼ 2
>
>
:
p1 p2
where y and na ∈f0; 1g are numbers of reference alleles of an admixed individual at

t, and that of alleles from population 1, respectively. pk is the allele frequency of
population k∈f1; 2g at the site t, such that when na ¼ 0, pk ¼ p1 while pk ¼ p2 when
na ¼ 2. Nowadays, technological, statistical, and computational advances avail
enormous amounts of high density SNP data. Although high density SNPs violate
the independence assumption due to background LD [21], they contain more infor-
mation than in AIMs [22]. To loosen the independence assumption and minimize
noise and systematic biases from unmodelled LD, more advanced local ancestry
inference methods emerged [22]. These methods include SEQMIX [23], PCADMIX
[24], and SUPPORTMIX [13].
SUPPORTMIX [11] models only admixture LD by combining support vector
machines (SVMs) and HMM. It was proposed in 2012 to improve on the computa-
tional time and address the challenge of a few typed or nonexistent reference
panels, which overall improve multi-way local ancestry deconvolution.
SUPPORTMIX is the first model to allow the learning of ancestral surrogates given a
pool of reference panels. As a result, it is capable to train ancestral populations that
are bigger in size than those that are mixed. Since SVMs can handle huge datasets,
SUPPORTMIX is faster than early methods. It uses the rich haplotype information.
Also proposed in 2012, PCADMIX [24] divides the genome into contiguous win-
dows of SNPs as in SUPPORTMIX. It leverages principal component analysis from
proxy ancestral haplotypes to model admixture LD under a standard HMM. Similar
to SUPPORTMIX, PCADMIX is fast and requires phased data. Nevertheless,
SUPPORTMIX and PCADMIX do not model phase switch errors, and as a result, in
2013, SEQMIX [23] was proposed. Unlike all other admixture LD-based methods,
SEQMIX is based on exome sequence, reads data, and uses HMM. SEQMIX models
only admixture LD and prunes SNPs in background LD. As a result, to reduce noise
and systematic biases from using all SNPs [10] whilst not fully modeling LD (back-
ground), admixture and background LD methods emerged [22].
2.1.1.2 Admixture and background LD models
Since the biological data often have some dependences that violate the indepen-
dence assumption in standard HMM, admixture LD-based methods are often not
realistic. To relax the independence assumption, the HMM is extended to either
Markov HMM, factorial HMM, hierarchical HMM, or two-layer HMM or other
multivariate statistical models such as multivariate normal distribution (MVN) and
a rich ancestral haplotype data are used unlike early methods. This is the case for
SABER [10], SWITCH [25], HAPAA [26], HAPMIX [4], MULTIMIX [27], ALLOY
[28], and ELAI [29]. MHMMs were the first HMM extension in local ancestry. They
were first implemented in SABER and later in SWITCH. SABER was the first
method to model background LD in the genetic ancestry inference. MHMM
assumes that the current observed haplotype depends on both the current ancestry
39
and the immediate past observation. The difference in the MHMM and admixture
LD HMM-based is that when ancestry switches between sites t � 1 and t, then the
MHMM observation model depends on the joint distribution of allele frequencies at
the two sites [6, 30], defined as follows [10]:
� 0
� � 0
�
P Y t ¼ cjY t�1 ¼ d; X t ¼ k; X t0 ¼ k ¼ Bt c; d; k ; k ,
8
� � <B ~ 0 ðc; dÞ for k0 ¼ k (4)
0 k ,t
P Y t ¼ cjY t�1 ¼ d; X t ¼ k; X t0 ¼ k ¼
: B 0 ðcÞ otherwise
k ,t
where B ~ ðc; dÞ is the probability of having alleles at marker t provided there was
k, t
allele d at t � 1 and Bk, t ðcÞ the allele frequency of alleles at marker t have for origins
the population k. However, if the ancestry does not switch, then the observation
model is like that of models in Section 2.1.1.1. The transition model of the SABER
model accounts for the differences in admixture times that are in the real case of
continuous gene flow where populations contribute their genetic material to the
admixture in different generations [10]. Tang et al. defined the probability of
switching from ancestry k at t to k at t as
8
>
> g 2i
< qi
>
∑Kk¼1 qk g k
� gi , for i ¼ j,
Aij ¼ gi gj (5)
>
>
>
: qj , otherwise
∑Kk¼1 qk g k
where g k is the admixture time when population k started to contribute to the

admixture.
However, SABER has a large parameter set, and does not explicitly model back-
ground LD as it models background LD using first order Markov chain [22]; other
methods such as SWITCH were proposed. SWITCH takes into recombination even
if it does not result in an ancestry switch, emerged. In contrast to SABER, SWITCH
conditions the MHMM on recombination. Similar to early methods, probability of
recombination depends on the admixture generations, genetic distance between
consecutive SNPs, and the recombination rate. Thus, if the transition probability
model in SWITCH is marginalized over recombination, then it is similar to Eq. (2)
for two-way and Eq. (5) for multi-way. Although SWITCH models background LD
and estimates recombination rates, the authors recommended richer MHMM or
other different models that would outperform the SWITCH and SABER pairwise
models [25]. As a result, methods that use both large- and small-scale HMM,
referred to as the HHMM, were introduced.
2.1.2 Non-LD-based local ancestry inference models
Non-LD methods neither model background nor admixture LD. They either
remove SNPs in LD which is the case for LAMP [11] and WINPOP [31], or use all
SNPs (linked and unlinked SNPs) without modeling LD; this is the case for EILA
[14], RFMIX [32], and LOTER [15]. Since MHMMs have a large number of param-
eters and do not model LD explicitly, an algorithmic approach that divides genome
into windows of SNPs, LAMP [11], emerged in 2008. LAMP is fast and robust, and
can infer local ancestry even without proxy ancestral genotypes. This is the case for
40
two-way admixtures. It uses the naive Bayes classifier and a clustering algorithm
known as the iterative conditional modes. LAMP estimates the most probable
ancestry at a site by applying the majority vote for each SNP [11]. Although accu-
racy is comprised, LAMP does not suffer from challenges of HMM and extension.
As a result, LAMP underperforms in closely related populations, and hence it was
extended to WINPOP [31], a dynamic programming algorithm. Unlike LAMP,
WINPOP assumes at least one recombination event within each window and varies
the window length depending on the genetic distance between populations. Hence,
WINPOP and LAMP outperform other methods in closely and distantly related
populations, respectively. Both LAMP and WINPOP assume unlinked markers and
discards SNPs in LD.
As the admixed sequence data availability increases, Maples et al. proposed a
discriminative approach to estimate local ancestry, RFMIX [32]. A discriminative
approach estimates the posterior probability directly and not via the joint probabil-
ity distribution. In contrast to generative ancestry inference models, RFMIX uses
the information contained in admixed individuals. This is advantageous in cases of
genotyped few reference panels. This is the case for Native Americans [32]. RFMIX
uses conditional random fields (CRFs) parametrized on random forests. It outper-
forms in multi-way admixtures maybe due to modeling phase switch errors. In
2013, EILA [14], a multivariate statistic based method, was proposed particularly to
increase inference power through addressing three common challenges in local
ancestry. Addressed challenges are the independence of SNP assumption, difficul-
ties in identifying break points, and the use of three genotype values. Instead of raw
genotypes, EILA uses a numerical value between 0 and 1. The score determines how
close SNPs are to the ancestral populations. Breakpoints are a challenge to identify,
but EILA identifies them by fused quantile regression facilitating the use of esti-
mates in admixture dating. Finally, k-means classifiers are used to infer ancestry
using all genotyped SNPs [14].
Recently, a software package that deconvolves local ancestry in multi-way
admixtures for a wide range of species, LOTER [15], was proposed. LOTER can
account for phase errors in two-way admixture only. It facilitates the local ancestry
inference process and its application in non-model species [15]. Unlike other
methods, LOTER needs no biological such as admixture time and recombination
rate or statistical parameters such as, number of hidden states and misfit probabil-
ities to deconvolve ancestry [15]. Although it uses the Li and Stephen’s copying
model [33] as in LAMPLD/LAMPHAP, LOTER is a nonprobabilistic approach for-
mulated from an optimization problem. Its solution is obtained through dynamic
programming.
Finally, different existing LD and non-LD-based local ancestry inference models
are summarized in Table 1 extracted from Geza et al. [34].
2.2 Models for dating admixture events in a genome
Several models are now available to determine the date of admixture events in
a given admixed genome. Breakpoints of haplotypes are used by some models
while others focus on the ancestry blocks. Models based on ancestry blocks for
dating admixture are formulated using either an empirical criteria or variants
associated with a specific population. In order to determine the average length of
the admixture block, these methods then assign ancestry on predefined windows
using either wavelet transformation or conditional random fields [35]. On the
other hand, there are models requiring rapid decrease in haplotype block sizes to
estimate the date of the admixture event [36]. This suggests that, in general,
41
42
Software Multi- Account LD model Biological/statistical parameters Reference Admixed Year of
way LD populations populations publication
STRUCTURE V2* ✓ ✓ HMM Markers, and ancestry proportions Unphased Unphased August 2003
ANCESTRYMAP* ✗ ✓ HMM Physical map, recombination and ancestry Unphased Unphased May 2004
proportions
ADMIXMAP* ✓ ✓ HMM Physical map and ancestry proportions Unphased Unphased May 2004
SABER ✓ ✓ MHMM Physical map or recombination distance Phased/unphased Phased/unphased July 2006
“LAMP” ✓ ✗ ✗ Admixture generations, LD threshold, and physical Unphased Unphased February 2008
map
HAPAA ✓ ✓ HHMM Admixture generations and genetic divergence Phased Phased February 2008
SWITCH ✓ ✓ MHMM Recombination rate Phased Phased February 2008
GEDI-ADMX ✓ ✓ Fixed size FHMM Admixed and ancestral SNPs (physical map) Phased Unphased May 2009
WINPOP ✓ ✗ ✗ Recombination, admixture generations, LD threshold, Unphased Unphased June 2009

and physical map
HAPMIX ✗ ✓ HHMM Genetic map mutation rate and admixed and ancestral Phased Unphased June 2009
SNPs
CHROMOPAINTER ✓ ✓ Co-ancestry matrix Recombination rate Phased Phased January 2012
LAMPLD ✓ ✓ HHMM Number of hidden states, window size and physical Phased Unphased May 2012
map
SUPPORTMIX* ✓ ✓ HMM Admixture generations and genetic map Phased Phased June 2012
PCADMIX* ✓ ✓ Windows of blocks Genetic map and window size Phased Phased August 2012
of SNPs
mSPECTRUM ✓ ✓ SNPs, mutation and recombination rate Phased Phased August 2012
MULTIMIX ✓ ✓ MVN Genetic map, legend file and misfitting probabilities Phased/unphased Phased/unphased November
2012
43
Software Multi- Account LD model Biological/statistical parameters Reference Admixed Year of
way LD populations populations publication
ALLOY ✓ ✓ Non-homogeneous Markers, ancestral proportions, admixture Phased Phased February 2013
VLMC generations, and genetic map
RFMIX ✓ ✗ ✗ Genetic map, window size, and admixture Phased Phased August 2013
generations
EILA ✓ ✗ ✗ Physical map Unphased (no missing Unphased (no missing November
values) values) 2013
SEQMIX ✓ ✗ ✗ Genetic map Unphased Unphased November

2013
ELAI ✓ ✓ Two layer HMM Admixture generations, lower and upper cluster Phased/unphased Phased/unphased May 2014
LOTER ✓ ✗ ✗ — Phased Phased November

2017
Table 1.
Existing 20 ancestry deconvolution tools: ✓ indicates the ability of the software to perform a specified task, ✗ indicates the inapplicability of the task by a particular tool. Unless explicitly
specified, LD refers to background LD.
models used for dating admixture events can be subdivided in two main classes
[17, 18], namely those based on LD and those based on the haplotype distribution,
as mentioned earlier.
2.2.1 LD-based models for dating admixture events
An admixture event is mainly characterized by the transfer of genes from

the ancestral populations to the admixed ones. This leads to the appearance of
linkage disequilibrium with regard to the ancestral populations. However, this
LD formed often decreases with time. Also, the rate of decay of LD is a
function of recombination and the proportion of the admixture [35]. Inversely,
many methods employ this rate to calculate the time since the admixture
event occurs.
In 2011, Moorjani et al. introduced a method to determine the weighted correla-
tion for a pair of SNPs [36]. This correlation coefficient is further used to measure
the LD with ancestral populations [37]. The time of admixture is then determined
by analyzing the correlation with respect to the genetic distance, and also fitting
using a least squares method the decay of the correlation [35]. This method got
improved in 2011 by Loh et al. [18]. The major improvements are in terms of
computation. Loh et al. employed instead a fast Fourier transform and other faster
techniques to determine the optimal distance to the fitting curve. This method has
another advantage that it reduces considerable biases in the estimation of the time
of admixture [18, 36]. Later, Loh et al.’s method was improved by Pickrell et al. [38]
by introducing the notion of mixture exponential decay in order to take into
account the admixture events in the given admixed population history. It mainly
focuses on the decay of the LD.
2.2.1.1 Multiple weighted correlation coefficient
Let us consider three ancestral populations k1 , k2 , and k3 , and Q the admixed

population. Let us denote by ω1�2 , ω1�3 , and ω2�3 three weighted linkage disequi-
librium scores computed based on all possible pairs of SNPs between the three
ancestral populations: k1 � k2 , k1 � k3 , and k2 � k3 , respectively, in the admixed
population Q calculated using the method proposed by Loh et al. According to
Prickrell et al., the multiple weighted correlation coefficient is [38],
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ω22�3 þ ω21 � 2ω2�3 ω1�2 ω1�3
Ck1 �k2 , k1 �k2 , k2 �k3 ¼ : (6)
1 � ω22�3
The date of admixture between population k1 and k3 is

( δn
w0 þ w1 e�n1 100 , for one admixture event � Dð1Þ ,
Dk1 , k2 , k1 k2 �k2 k3 ¼ δn δn
(7)
w0 þ w1 e �n1 100
þ w2 e�n2 100 , in the case of two admixture events � Dð2Þ ,
with n1 and n1 the number of generations; δn the genetic distance; w1 and w2

stand for the value of the multiple weighted LD; and w0 the affine term. Dð1Þ is the
date of admixture of population Q in the case of admixture either between k1 � k2
or k2 � k3 . On the other hand, if it is assumed that two admixture events took place
between k1 � k3 and either k1 � k2 or k2 � k3 , the date of the admixed population is
given by Dð2Þ .
44
2.2.2 Haplotype distribution-based models for dating admixture events
Among the haplotype-based approaches, there is the likelihood method intro-

duced in 2009 by Price et al. [4]. It basically determines the number of breakpoints
using Hidden Markov Model. It is also able to determine the number of alleles at a
particular site inherited from a given ancestor in a population. This is done in two
steps. First, the method consists in identifying haplotype from the proxy ancestry
populations, and secondly, the origin of each haplotype bock is identified by com-
paring their likelihood for one ancestral population versus the others. Considering
an admixed genome, the likelihood of an observed allele is given by

θu Pðtvw ¼ 0Þ þ ð1 � θu ÞPðtvw ¼ 1Þ, if u ¼ v,
Huvw ðhÞ ¼ (8)
θ3 Pðtvw ¼ 0Þ þ ð1 � θ3 ÞPðtvw ¼ 1Þ, otherwise
with θu , u∈f1; 2; 3g the mutation parameter is; h represents the haplotype site in the
chromosomal offspring; the function tvw is an indicator function. It takes the value 1
if individual w coming from offspring x has the same haplotype with the ancestral
population v and 0 otherwise; and P is the probability to inherit a pair of haplotype
[4]. The number of generations since admixture is given by
C
G¼ (9)
4γ ð1 � γ Þζ
where ζ is the total Morgan length, γ the proportion of admixture, and C the
observed number of breakpoints [4].
On the other hand, Pugach et al. [17] employed the wavelet transform to design
a haplotype block approach. The aim of this method is to derive the time of admix-
ture of a given population using the simple hybrid isolation model. It proceeds in
two main steps. First, it obtains a signal of admixture from the admixed data using
the principal component technique. The second step consists in deriving the date of
admixture using the signal obtained in the first step [17].
Pool and Nielsen also built a haplotype-based approach. It used precautionary
ancestral populations to infer the date of admixture from the genome of an admixed
population [39]. It assumed that after a number of generation g, the distribution of
the ancestral haplotypes follows exponential distribution given by
f ðλ; gÞ ¼ ge�λg (10)
where λ is the length of haplotypes. Also, the mean of this distribution is known and
is equal to 1g.
Further methods include that of Gravel developed in 2012 for the identification
of multiple ancestral populations in a given admixture dataset [40]. Also, Jin et al.
[41] came up with a similar method to explain admixture dynamics. The method
incorporates several models including gradual admixture (GA), hybrid isolated
(HI), and continuous gene flow (CGF) models [41], which can be extended to
GA-Isolation (GA-I) and CGF-Isolation (CGF-I) by considering isolation after
admixture [42]. Hellenthal et al. [43] on the other hand built up on the work of
Lawson et al. [44] on dating admixture. This method particularly considers the
genome of an admixed individual to be a set chunk DNA coming from other
individuals. The scheme of this method is mainly made of two stages. The first stage
consists in dividing the genome into chunks and matching each of them to the
proper ancestral individual. This stage is achieved with the help of Hidden Markov
45
Model. The second stage consists in identifying haplotypes and determining their
respective ancestral population [43, 44]. Moreover, the admixture event and its
date are derived by fitting the decay of the ancestral haplotype with an exponential
distribution curve. Moreover, Ni et al. developed a method based on the observation
that the date of admixture events is related to the model used. Their method
consists in using the likelihood ratio test to identify the best model for the inference
of the date of admixture. Furthermore, they are able to estimate several admixture
events with the given optimal model [35].
Finally, different existing models and tools for dating admixture events are
summarized in Table 2 extracted from Chimusa et al. [35].
Tool Category Admixture Priori Multi- Online link

model proxy way
ancestral events
raw data
ROLLOFF LD-based model HI Yes No https://github.com/

DReichLab/Ad
mixTools/
ALDER HI Yes No http://cb.csail.mit.edu/

cb/alder/
MALDER HI Yes Yes https://github.com/joe

pickrell/malder/
CAMer HI, GA, Yes Yes https://github.com/da

CGF, GA-I, vid940408/CAMer
CGF-I
IMAAPs HI, GA, Yes Yes http://www.picb.ac.c

CGF, GA-I, n/PGG/resource.php
CGF-I
StepPCO Haplotype/ HI Yes Yes https://bioinf.eva.mpg.

ancestry block size de/download/Ste
distribution-based pPCO/
model
Adware HI, Dual- Yes Yes https://cran.r-project.
admixture org/web/packages/ad
wave/index.html
HAPMIX HI Yes Yes http://genetics.med.ha

rvard.edu/reichlab/Re
ich_Lab/Software.h
tml/
MultiWaveIner HI Yes Yes https://github.com/

xyang619/MultiWave
Infer/ or
http://www.picb.ac.c
n/PGG/resource.php
GLOBBERTROTTER HI, GA, No Yes https://github.com/

CGF maarjalepamets/huma
n-admixture/
Tracts HI, GA, No Yes https://github.com/sg

CGF ravel/tracts/
Ancestry_HMM HI No No https://github.com/
russcd/
Table 2.
Existing dating admixture genomic tools.
46
3. Challenges and perspectives
3.1 Case of local ancestry inference models
Although several models exist to deconvolve local ancestry, most studies that
evaluate such models showed that deviations in local ancestry estimates still exist in
multi-way admixtures. Deviations in local ancestry also result from genetic drift,
miscalling true ancestry, and genotyping errors. However, the signals from these
factors affect the whole genome while that of unmodelled natural selection affects
particular regions. For example, Chen et al. using four local ancestry inference
models to scan for disease-related loci through admixture mapping showed that
although all of them are LD based and divide the genome into windows of contin-
uous SNPs, MULTIMIX and LAMPLD estimates differed in almost 20% of the
analyzed SNPs. This results from the differences in the biological and statistical
parameters they require and the mathematical approaches they use. Another asso-
ciation study by Chimusa et al. [45] also pointed out that admixture mapping is still
limited by inaccuracies in multi-way local ancestry deconvolution when they eval-
uated one LD-based and one non-LD-based local ancestry models, WINPOP and
LAMPLD.
Inaccuracies in local ancestry estimates may result from the use of statistical or
biological parameters in the estimation process, which are not always accurate when
provided. It could also be due to the dependence of models on reference panels
which for some populations are few or even not sampled for others. This is the case
for the Native Americans. More so for other admixed populations, their history is
not well known. When applied to ancient admixtures, existing methods may yield
spurious estimates as they were designed for recent admixtures. Existing methods
do not account for natural selection; hence, some deviations exist in regions that are
under selection [45]. Also, most of them are benchmarked for three-way
admixtures.
Since each model was introduced to address a particular challenge that models
before it faced, it is clearly expected that no model or tool can achieve the best
performance in all admixture scenarios and not trading estimate accuracy with
computational speed. Using existing studies by Geza et al. [34], more than 50% of
studies that either introduced a model or evaluated methods for association map-
ping showed that LAMPLD/LAMPHAP outperforms most LD-based methods. And
the only LD-based method than outperformed LAMPLD is ELAI; however, this is
the only study that assessed ELAI with other models. In cases where LD-based
models were compared to non-LD-based models, RFMIX outperformed LAMPLD
in three cases highlighted in [34], while another separate study aiming to determine
the place of admixture of an admixed population RFMIX also outperformed. This
could be because RFMIX can deconvolve ancestry in closely related populations
[46]. However, a recent assessment between RFMIX and LOTER resulted in LOTER
outperforming in ancient admixtures [15].
Generally, each model is implemented as a tool in local ancestry deconvolution,
existing as individual scripts requiring unique inputs and producing unique outputs.
This challenges researchers with a limited computational background; thus, there
is lack of a unified framework which can require a standard easy to manipulate
input files and output results in a way that is easy to process for further application.
In conclusion, for informed decisions on models and algorithms, existing models or
tools should be assessed within a unified framework. This will allow them to be
tested on different admixture scenarios and also incorporating most state-of-the-art
LD and non-LD based models.
47
3.2 Case of the dating admixture models
The evolution of human populations and the history of the mixture of these
populations have been deciphered using statistical and computational methods.
These methods have been found to perform well when dealing with single point
admixture event in two-way admixed populations [35]. However, as any method,
they not only have advantages but also pitfalls regarding the estimation of admix-
ture dates in some cases. It is challenging to fit to real admixed populations (for
more than 3-way admixture context) in the existing models dating admixture
events due to several reasons, including reliance to optimal local ancestry estimates
and accurate ancestry breakpoints. This suggests that there is still a need for
designing an integrative or a new model to dating admixture events for current
multi-way admixed populations to further advance our understanding of human
demographics and movement, and facilitate admixture mapping and estimation of
the age of a disease locus contributing to disease risk.
In addition, it have been discovered that the mixture exponential decay model
over-estimates the date of older admixture events [35] and was suggested to detect
at most three admixture events. As mentioned earlier, Ni et al. [47] dealt with the
optimization of the method used in dating admixture estimation. They took into
account several models but the evaluation of their technique is not effective in the
estimation of ancient and multi admixture events [35, 47]. On the other hand,
several practical considerations can further limit these approaches including the use
of proxy ancestry populations in the estimations which could bias the accuracy of
the result. This is the case when dealing for instance with low sample size and
inappropriate proxy ancestral populations [35]; the requirement of having accurate
LD patterns, ancestry haplotypes distribution, and a big sample size of the
admixed population. Thus, there is a need for an adequate model for inferring
different dates of admixture events and matching real admixture history using
proxy ancestry-based methods [35].
4. Conclusions
Currently, more than 20 models exist and are implemented as software to

deconvolve local ancestry and 12 tools for dating admixture events. In this chapter,
we discussed in detail and summarized the most commonly used models, the model
assumptions, statistical and biological parameters they require, and existing chal-
lenges. This discussion highlights the need for designing more effective models,
which account for current challenges and produce more accurate and biologically
relevant estimates. Furthermore, it provides useful information for the implemen-
tation of practical tools, which consider current medical and population genetic
demands. More importantly, this may guide users in the choice of appropriate tools
for specific applications and can assist software developers in designing more
advanced tools for local ancestry deconvolution and dating admixture events.
Acknowledgements
Some of the authors are supported in part by the National Institutes of Health
(NIH) Common Fund [grant numbers U24HG006941 (H3ABioNet) and
1U01HG007459–01 (SADaCC)]. One of the authors is fully funded by the Organi-
zation for Women in Science for the Developing World (OWSD) and Swedish
International Development Cooperation Agency (Sida). The content of this
48
publication is solely the responsibility of the authors and does not necessarily
represent the official views of the funders.
The authors declare that they have no competing interest.
Author details
Gaston K. Mazandu1,2,3*, Ephifania Geza1,3, Milaine Seuneu1,2

and Emile R. Chimusa2
1 African Institute for Mathematical Sciences (AIMS), Cape Town, South Africa
2 Division of Human Genetics, Department of Pathology, Faculty of Health

Sciences, Institute of Infectious Disease and Molecular Medicine, University of
Cape Town (UCT), Cape Town, South Africa
3 Computational Biology Division, Department of Integrative Biomedical Sciences,

University of Cape Town, South Africa
49
References
[1] Cavalli-Sforza LL, Feldman MW. The [9] Hoggart CJ, Shriver MD, Kittles RA,
application of molecular genetic Clayton DG, McKeigue PM. Design and
approaches to the study of human analysis of admixture mapping studies.
evolution. Nature Genetics. 2003;33: The American Journal of Human
266-275 Genetics. 2004;74(5):965-978
[2] A. Koehl, Estimating Ancestry and [10] Tang H, Coram M, Wang P, Zhu X,
Genetic Diversity in Admixed Risch N. Reconstructing genetic
Populations. The University of New ancestry blocks in admixed individuals.
Mexico. Thesis 2016 The American Journal of Human
Genetics. 2006;79(1):1-12
[3] Yang JJ, Cheng C, Devidas M, Cao X,
Fan Y, Campana D, et al. Ancestry and [11] Sankararaman S, Sridhar S, Kimmel
pharmacogenomics of relapse in acute G, Halperin E. Estimating local ancestry
lymphoblastic leukemia. Nature in admixed populations. The American
Genetics. 2011;43(3):237-241 Journal of Human Genetics. 2008;82(2):
290-303
[4] Price AL, Tandon A, Patterson N,
Barnes KC, Rafaels N, Ruczinski I, et al. [12] Baran Y, Pasaniuc B, Sankararaman
Sensitive detection of chromosomal S, Torgerson DG, Gignoux C, Eng C,
segments of distinct ancestry in et al. Fast and accurate inference of local
admixed populations. PLoS Genetics. ancestry in latino populations.
2009;5(6):e1000519 Bioinformatics. 2012;28(10):1359-1367
[5] Thornton TA, Bermejo JL. Local and [13] Omberg L, Salit J, Hackett N, Fuller
global ancestry inference and J, Matthew R, Chouchane L, et al.
applications to genetic association Inferring genome-wide patterns of
analysis for admixed populations. admixture in qataris using fifty-five
Genetic Epidemiology. 2014;38(S1): ancestral populations. BMC Genetics.
S5-S12 2012;13(1):49
[6] Liu Y, Nyunoya T, Leng S, et al. [14] Yang JJ, Li J, Buu A, Williams LK.
Softwares and methods for estimating Efficient enference of local ancestry.
genetic ancestry in human populations. Bioinformatics. 2013;29(21):2750-2756
Human Genomics. 2013;7(1):1
[15] Dias-Alves T, Mairal J, Blum MG.
[7] Bhatia G, Patterson N, Pasaniuc B, Loter: A software package to infer local
et al. Genome-wide comparison of ancestry for a wide range of species.
African-ancestry populations from care Molecular Biology and Evolution. 2018;
and other cohorts reveals signals of 35(7):msy126
natural selection. American
Journal of Human Genetics. 2011;89: [16] Cheng R, Lim J, Samocha K, et al.
368-381 Genome-wide association studies and
the problem of relatedness among
[8] Patterson N, Hattangadi N, Lane B, advanced intercross lines and other
Lohmueller KE, Hafler DA, Oksenberg highly recombinant populations.
JR, et al. Methods for high-density Genetics. 2010;185:1033-1044
admixture mapping of diseases genes.
The American Journal of Human [17] Pugach I, Matveyev R, Wollstein A,
Genetics. 2004;74(5):979-1000 et al. Dating the age of admixture via
50
wavelet transform analysis of genome- origin using HAPAA. Genome Research.

wide data. Genome Biology. 2011;12:R19 2008;18(4):676-682
[18] Loh P-R, Lipson M, Patterson N, [27] Churchhouse C, Marchini J.

et al. Inferring admixture histories of Multiway admixture deconvolution
human populations using linkage using phased or unphased ancestral
disequilibrium. Genetics. 2013;193: panels. Genetic Epidemiology. 2013;
1233-1254 37(1):1-12
[19] Murphy KP. Machine Learning: A [28] Rodriguez JM, Bercovici S, Elmore
Probabilistic Perspective. Cambridge, M, Batzoglou S. Ancestry inference in
Massachusetts, London: MIT press; 2012 complex admixtures via variable-length
Markov chain linkage models. Journal of
[20] Falush D, Stephens M, Pritchard JK. Computational Biology. 2013;20(3):
Inference of population structure using 199-211
multilocus genotype data: Linked loci
and correlated allele frequencies. [29] Guan Y. Detecting structure of
Genetics. 2003;164(4):1567-1587 haplotypes and local ancestry. Genetics.
2014;196(3):625-642
[21] Chen X, Ishwaran H. Random
forests for genomic data analysis. [30] Padhukasahasram B. Inferring
Genomics. 2012;99(6):323-329 ancestry from population genomic data
and its applications. Frontiers in
[22] Seldin MF, Pasaniuc B, Price AL. Genetics.5:204
New approaches to disease mapping in
admixed populations. Nature Reviews [31] Paşaniuc B, Sankararaman S,
Genetics. 2011;12(8):523-528 Kimmel G, Halperin E. Inference of
locus-specific ancestry in closely related
[23] Hu Y, Willer C, Zhan X, Kang HM, populations. Bioinformatics. 2009;
Abecasis G. Accurate local-ancestry 25(12):i213-i221
inference in exome-sequenced admixed
individuals via off-target sequence [32] Maples BK, Gravel S, Kenny EE,
reads. The American Journal of Human Bustamante CD. RFMix: A
Genetics. 2013;93(5):891-899 discriminative modeling approach for
rapid and robust local-ancestry
[24] Brisbin A, Bryc K, Byrnes J, inference. The American
Zakharia F, Omberg L, Degenhardt J, Journal of Human Genetics. 2013;93(2):
et al. PCAdmix: Principal components- 278-288
based assignment of ancestry along each
chromosome in individuals with [33] Li N, Stephens M. Modeling linkage
admixed ancestry from two or more disequilibrium and identifying
populations. Human Biology. 2012; recombination hotspots using single-
84(4):343 nucleotide polymorphism data.
Genetics. 2003;165(4):2213-2233
[25] Sankararaman S, Kimmel G,
Halperin E, Jordan MI. On the inference [34] Geza E, Mugo J, Mulder NJ,
of ancestries in admixed populations. Wonkam A, Chimusa ER, Mazandu GK.
Genome Research. 2008;18(4):668-675 A comprehensive survey of models for
dissecting local ancestry deconvolution
[26] Sundquist A, Fratkin E, Do CB, in human genome. Briefings in
Batzoglou S. Effect of genetic Bioinformatics. 2018. DOI: 10.1093/bib/
divergence in identifying ancestral bby044
51
[35] Chimusa ER, Defo J, Thami PK, using dense haplotype data. PLoS
Awany D, Mulisa DD, Allali I, et al. Genetics. 2012;8:e1002453
Dating admixture events is unsolved
problem in multi-way admixed [45] Chimusa ER, Zaitlen N, Daya M,
populations. Briefings in Bioinformatics. Møller M, van Helden PD, Mulder NJ,
2018:1-58. https://doi.org/10.1093/bib/ et al. Genome-wide association study of
bby112 ancestry-specific tb risk in the South
African coloured population. Human
[36] Moorjani P, Thangaraj K, Patterson Molecular Genetics. 2014;23(3):796-809
N, et al. Genetic evidence for recent
population mixture in India. Human [46] Xue J, Lencz T, Darvasi A, Pe'er I,
Genetics. 2013;93:422-438 Carmi S. The time and place of
European admixture in Ashkenazi
[37] Moorjani P, Patterson N, Jewish history. PLoS Genetics. 2017;
Hirschhorn J, et al. The history of 13(4):e1006644
African gene flow into Southern
Europeans, Levantines, and Jews. PLoS [47] Ni X, Yuan K, Yang X, et al.
Genetics. 2011;7:e1001373 Inference of multiple-wave admixtures
by length distribution of ancestral
[38] Pickrell J, Reich D. Toward a new tracks. Heredity (Edinb). 2018;121:52-63
history and geography of human genes
informed by ancient DNA. Trends in
Genetics. 2014;30:377-389
[39] Pool J, Nielsen R. Inference of

historical changes in migration rate
from the lengths of migrant tracts.
Genetics. 2009;181:711-719
[40] Gravel S. Population genetics

models of local ancestry. Genetics. 2012;
191:607-619
[41] Jin W, Li R, Zhou Y, et al.

Distribution of ancestral chromosomal
segments in admixed genomes and its
implications for inferring population
history and admixture mapping. Human
Genetics. 2014;22:930
[42] Zhou Y, Qiu H, Xu S. Modeling

continuous admixture using admixture-
induced linkage disequilibrium.
Scientific Reports. 2017;7:43054
[43] Hellenthal G, Busby G, Band G,

et al. A genetic atlas of human
admixture history. Science. 2014;434:
747-751
[44] Lawson D, Hellenthal G, Myers S,

et al. Inference of population structure
52
Chapter 4
Recognition of Multiomics-Based
Molecule-Pattern Biomarker for
Precise Prediction, Diagnosis, and
Prognostic Assessment in Cancer
Xanquan Zhan, Tian Zhou, Tingting Cheng and Miaolong Lu
Abstract
Cancer is a complex whole-body chronic disease, is involved in multiple causes,

multiple processes, and multiple consequences, which are associated with a series
of molecular alterations in the different levels of genome, transcriptome, proteome,
metabolome, and radiome, with between-molecule mutual interactions. Those
molecule-panels are the important resources to recognize the reliable molecular
pattern biomarkers for precise prediction, diagnosis, and prognostic assessment in
cancer. Pattern recognition is an effective methodology to identify those molecule-
panels. The rapid development of computation biology, systems biology, and
multiomics is driving the development of pattern recognition to discover reliable
molecular pattern biomarkers for cancer treatment. This book chapter addresses the
concept of pattern recognition and pattern biomarkers, status of multiomics-based
molecular patterns, and future perspective in prediction, diagnosis, and prognostic
assessment of a cancer.
Keywords: cancer, multiomics, genomics, transcriptomics, proteomics,

metabolomics, radiomics, molecule-pattern biomarker, pattern recognition
1. Introduction
Cancer is a leading cause of death worldwide, with increasing morbidity and

mortality. Studies indicated that the number of new cancer case per year will be
19.3 million by 2025, and more than half of cancer cases and mortality occur in
developing countries and the proportion tendency is estimated to increase by 2025
[1]. Cancer is a complex process involving multiple causes, multiple processes, and
multiple consequences, which are associated with a series of molecular alterations in
the different levels of genome, transcriptome, proteome, metabolome, and radiome,
with between-molecule mutual interactions. Cancer arises when normal cells’ orderly
processes controlled multiplication and life span were interfered. It is also reported
that person’s genetic makeup and lifestyle factors such as diet, alcohol, smoke, and
physical activity, influence the rate at which cancer develops and progresses.
Alterations or mutations of genetic substance of the cells are the main cause of
changes in cellular behavior. Dysregulation of the normal cellular procedure in can-
cer for cell fission, differentiation, apoptosis and proliferation is due to alterations
53
in multiple genes expression and leading to an imbalance between cell replication

and cell death, which is beneficial for growth of tumor cell population [2, 3]. With
cancer progresses, the genetic drift of the cell population generates cell hetero-
geneity with characteristics involved in cell antigenicity, invasiveness, metastatic
potential, rate of cell proliferation, differentiation state and response to chemo-
therapeutic agents [4–6]. A study showed that the mutation of two to eight driver
genes is sufficient for an emblematical cancer occurrence. The passenger genes are
not oncogenic and mutation of passenger genes is unable to cause occurrence of a
cancer [7, 8]. Therefore, attention should be paid to a panel of genetic mutations,
named gene pattern mutation. Depending on the genetic central dogma, gene
pattern mutation may lead to a series of alterations of messenger RNA (mRNA) and
protein expressions. With the use of this pattern, the condition of low sensitivity
of a single-tumor marker or low specificity of a large number of samples is reduced
when diagnosis models are set based on differentially expressed proteins or peptides
between tumor tissues and normal tissues [9].
A cancer biomarker is defined as a substance or biological process that can
indicate the presence of cancer in the body, which is important for people to monitor
personal health [10]. Physical examinations (e.g., blood pressure), biological and
genetic tests, along with others that can be objectively detected and used as indica-
tors of pathogenic processes and alterations which may present as a result of treat-
ment, are regarded as biomarkers [11, 12]. All the alterations in the levels of DNA,
RNA, protein, and metabolite between cancer patients and healthy people could be
called biomarkers, and therefore in terms of source, biomarkers usually are assorted
into different categories including genetic biomarker, epigenetic biomarker, protein
biomarker, metabolite biomarker and immunological biomarker and so on [13].
Generally, biomarkers used in clinic survey and diagnosis are from the four ways:
(i) metabolites of tumor cells, (ii) abnormal differentiation of cellular gene
products, (iii) tumor necrosis and exfoliation of tumor cells release into the blood
circulation, and (iv) cell reactive products of tumor host cells [9]. Most of cancer
biomarkers are detected in the tumor tissue or in blood. In order to maximize
usefulness and minimize cost of screening or early detection, it is advantageous to
be able to measure biomarkers in body fluid, which can be obtained using minimally
invasive samples, such as blood, urine, sputum or stool [10]. Biomarkers play an
important role in cancer for precise prediction, diagnosis and prognostic assessment.
Thereby, with the development of biomarkers, they have far-reaching significances
for people to recognize and treat cancer as follows: (i) the understanding of molecu-
lar mechanisms of diseases, (ii) identification of possible new disease pathways, (iii)
prediction models of complex diseases, (iv) the determination of the level of biologi-
cal activity of the disease, (v) refinement of disease phenotypes that may respond
differently to specific treatments, (vi) the monitoring of treatment responses,
and (vii) the potential application of precision medicine [14, 15]. However, it still
remains a problem that biomarkers were detected after occurrence of cancer. With
the fast development of image technology, radiomics is generated and can well solve
that above problem. Quantitative analysis of imaging characteristics provides not
only the tumor phenotype but also the underlying genotype information so that
one can better diagnose and prognostic assessment for cancer patients [16]. A single
tumor biomarker is insufficient and unreliable for precise prediction, diagnosis
and prognostic assessment in cancer. The multi-parameter systematic strategies for
predictive, preventive, and personalized medicine (PPPM) in cancer [4] emphasized
that those molecule-panels, all of the differences and molecular alterations in the
genome, transcriptome, proteome, metabolome, and radiome, with between-
molecule mutual interactions, are the important resources to identify and recognize
the reliable molecular pattern biomarkers for precise prediction, diagnosis, and
54
Recognition of Multiomics-Based Molecule-Pattern Biomarker for Precise Prediction, Diagnosis…
prognostic assessment in cancer. Pattern recognition is an effective methodology to

identify those molecule-panels. In fact, pattern recognition means that recognize
molecule-pattern biomarkers, in other words, to use a set of patterns that consist of
several biomarkers to improve the accuracy and specificity of prediction, preven-
tion, diagnosis, treatment, and prognostic assessment of tumor [9].
The rapid development of computation biology, systems biology, and mul-
tiomics is driving the development of pattern recognition to discover reliable
molecular pattern biomarkers for cancer treatment. This book chapter addresses the
concept of pattern recognition and pattern biomarkers, status of multiomics-based
molecular patterns, and future perspective in prediction, diagnosis, and prognostic
assessment of a cancer.
2. Pathophysiological basis of molecule-pattern biomarker in cancer
Cancer is a complex whole-body chronic disease, which results in a series of

molecular alterations and associated with signal transduction system, cell cycle,
proliferation, differentiation and apoptosis [17, 18]. Many factors are related to
occurrence and development of a cancer.
Genomic instability plays a key role in cancer development and progression. It
provides a way to make a cell or subset of cells gain an ability of selective advantage
than adjacent cells, achieving outgrowth and advantages in the tissue microenviron-
ment. Genomic instability can generate aneuploid cells. Aneuploidy influences
on the transcriptome and proteome and further results in proteotoxic stress and
activation of the endoplasmic reticulum stress response. Consequently, aneuploidy
can regulate features of the cells and the microenvironment [19]. In normal cells,
the quality of reproduction of the genome at each stage of the cell cycle is protected
by checkpoints. The existence of aneuploid cells in cancer exactly suggested one
or more checkpoints are failed. The genomic heterogeneity might provide growth
advantages for cancer “tissue” under selection pressure, such as hypoxia, immunity,
and treatment-related challenges [1]. Genomic instability in cancer causes a serious
challenge for cancer treatment.
Genetic mutations that cause cell dysfunction in most of cases support the
development and progression of cancer. Moreover, the interaction between cancer
cells and their environment, known as the tumor microenvironment, and their
mutually interacted regulatory factors, can affect disease initiation and progres-
sion. The tumor microenvironment is composed of stromal cells, extracellular
matrix (ECM), and signaling molecules that communicate with cancer cells. The
stromal cells including endothelial cells, pericytes, fibroblasts, and immune cells,
along with the surrounding ECM, constitute a supporting matrix for the tumor and
regulate the tumor microenvironment. Angiogenesis and metastasis, two pivotal
hallmarks of cancer, are modulated by the composition of the tumor microenviron-
ment. Furthermore, the tumor microenvironment is not only affected by signals
from tumor cells, but also stromal components through influencing cancer cell
function to promote tumor progression and metastasis [20, 21]. Therefore, tumor
microenvironment also is an important aspect for cancer therapy.
Tumor heterogeneity is another momentous feature of malignant tumor and
plays a vital role in development, progression, and treatment of cancer [22–26]. On
the one hand, in most of cancer cases, heterogeneity is found that not only from
same kind of tumor among different patients, but also in all tumor progression
phases of the identical individual patients [27]. The genetic instability of tumor cell
is tightly related to tumor progression and heterogeneity and leads to the presence
of variations [28, 29]. On the other hand, tumor heterogeneity is relevant to the
55
individual differences between tumor patients. For example, the function of liver
and kidney, age, physical condition, psychological status and personal lifestyle
factors, are also another important factors which affect on the tumor progres-
sion and treatment [30]. A number of treatment plans of patients were designed
according to the doctor’s experiences and adopted same therapy model for different
cancer patients in clinic. Due to ignore tumor heterogeneity, the “one-size-fits-all”
therapeutic model resulted in the expected curative effect could not completely
be achieved [4]. Thereby, tumor heterogeneity is becoming an important factor to
hinder the effective treatment and cancer research.
Molecular mechanisms of initiation and progression of a cancer do not just exist
one kind of intracellular signal pathway [31]. Several researches have indicated that
phosphoinositide 3-kinase/protein kinase B (PI3K/Akt), mitogen-activated protein
kinase (MAPK) and signal transducer, and activator of transcription 3 (STAT3)
pathways were activated in obesity-associated colon cancer. Mammalian target
of rapamycin (mTOR) as a down-stream of both PI3K/Akt and MAPK is highly
activated [32]. Activated mTOR in proper order inhibits the PI3K/Akt pathway
and further activates the STAT3 pathway [33]. In case that mTOR is inhibited, the
activity of PI3K/Akt may obviously increase owing to the feedback inhibition of
mTOR on PI3K activity [34]. Therefore, it is necessary to simultaneously suppress
the expressions of mTOR and PI3K for the treatment of obesity-related cancer [4].
Hence one can see that the interaction and interrelationship of multiple signaling
pathways is essential to pay more attention to study, and a single signaling molecule
or biomarker is unreliable for the prediction, diagnosis, and treatment of cancer.
So far, there are many kinds of treatments for cancer including surgery, radio-
therapy, and systemic treatments including cytotoxic chemotherapy, hormonal
therapy, immunotherapy, and targeted therapies [35]. Personalized or individual-
ized variations are related to human healthcare, and the relationship is shown
(Figure 1). Three primary stages, prediction/prevention, early-stage diagnosis/
early-stage therapy, and late-stage diagnosis/late-stage therapy are involved
in human healthcare. Personalized or individualized variations can be used as
biomarkers for prediction, and further the assessment of preventive response
reflects the results of preventive treatments. Personalized or individualized
variations also can be regarded as diagnostic biomarkers and further for cancer
therapy. The assessment of therapeutic response, known as prognostic assessment,
consists in early-stage therapy and late-stage therapy, and reveals the influence of
therapeutic intervention. Of the three stages, prediction/prevention is the most
significant part due to make people keep on a healthy condition and be treated in
time once cancer occurs. Early-stage diagnosis/therapy also is better approach to
block and repress the progression of cancer while the preventive strategy failed.
Late-stage diagnosis/therapy is also named clinical diagnosis and treatment of a
cancer. Unluckily, most of cancer cases were found in late stage. In order to avoid
aforementioned problem and improve people’s health level, many researchers
concentrate on exploration of biomarkers on prediction/prevention and early-
stage diagnosis/therapy for cancer [4]. According to functional classification,
biomarkers are divided into two categories (Table 1): (i) serving for the mecha-
nism and therapeutic targets, and (ii) devoting to prediction, diagnostic test, and
prognosis assessment. The first kind of biomarkers is relevant to the initiation and
development of disease, and directly indicates the mechanism and pathogenesis of
the disease. Commonly, it is pivotal site in cell signal pathways, like P53 in naso-
pharyngeal carcinoma (NPC) [36]. Another kind of biomarkers does not need to
be causal to the occurrence and development of the disease, but requires to be pro-
vided with specificity and a certain number of changes to be easily detected. Based
on Bayes’ rule, three or more key molecules can form molecule-pattern biomarker
56
Figure 1.
Variations involved in each aspect of healthcare. Reproduced from Hu et al. [4], with permission from BioMed
Central open access article, copyright 2013.
Types Definition Application
Type I This type of biomarker exist a causal relationship with disease, Contribute to the
associate with the initiation and development of disease, and mechanism and therapeutic
can directly address the pathogenesis of disease. targets of disease.
Type II This type of biomarker does not need a causal relationship Contribute to the
with the occurrence and development of disease, but requires prediction, diagnosis, and
specificity and a certain amount of change to be easily detected. prognostic assessment.
Table 1.
Concept and categories of biomarkers [9].
to improve the accuracy of cancer diagnosis and therapy [9, 37]. In summary, due
to the complex pathophysiological basis of cancer, recognition of molecule-pattern
biomarker for precise prediction, diagnosis, and prognostic assessment in cancer
is an urgent demand to study and further close to realize precision medicine (PM)
and PPPM.
3. Methodology of recognition of multiomics-based pattern biomarkers

in cancer
Based on central dogma, genetic changes influence the RNA expression, and
cause the alterations of proteins, along with taken into account the changes of
metabolite and tumor heterogeneity, all above variations in genome, transcrip-
tome, proteome, metabolome, and radiome are measured with corresponding
omics methodology including genomics, transcriptomics, proteomics, metabolo-
mics, and radiomics. Multiomics-generated biomarkers can make up integrative
molecule-pattern biomarkers and pattern recognition for cancer treatment. This
section mainly addresses the previous mentioned five omics approaches com-
bined with computation biology and systems biology contribute to the develop-
ment of cancer precise medicine (Figure 2) [9].
57
Figure 2.
Different levels of omics-based pattern biomarkers. Modified from Cheng and Zhan [9], with permission from
Springer open access article, copyright 2017.
3.1 Genomics
The development of genomics drives the understanding and cognition of

cancer. The development of gene sequencing technology is a significant advance-
ment in the field of scientific research. First, based on the method of the previous
“plus and minus,” Sanger modified and invented the “dideoxy method” for DNA
sequencing in 1977 [38–40]. Sanger sequencing acquired many achievements and
completed a great work “Human Genome Project.” Nevertheless, high cost and low
throughput are disadvantages of Sanger sequencing technology [41, 42]. The limit
of Sanger sequencing promotes the progression and generation of new sequencing
technology. The second-generation sequencing technology has many advantages
including higher speed and throughput, higher degree of parallelism, effective uti-
lization of reagents and so on. However, problems still exist, such as the reduction
of accuracy of sequencing and relatively higher cost [43, 44]. Due to the presence
of shortcomings of the second-generation sequencing technology, the next-
generation sequencing (NGS) requires to be discovered. The third-generation of
sequencing technology is found to make up for the deficiency of second-generation.
For example, PacBio RS and Oxford Nanopore sequencing not only possess funda-
mental character of the single molecule sequencing, do not need any polymerase
chain reaction (PCR) process, availably avoid the PCR bias caused by the system
error, and well improve the read length, but also keep the high-throughput and low
cost of the second-generation technology [45]. Research demonstrates that accu-
mulation of genomic alternations leads to the occurrence of cancer, which involves
small insertions and deletions, base substitutions, copy number alterations (CNA),
chromosomal rearrangements, and microbial infections [46]. Besides, a number of
polymorphic CNAs have been discovered in the human genome [47]. DNA micro-
arrays, also named as “gene chip” or “DNA chip,” obtained a great success that
could monitor tens of thousands of one time expression and hundreds of thou-
sands of genes. Single nucleotide polymorphisms (SNPs) are the most common
form of DNA variation in the human genome, approximately occurring one time
every 100–300 bases [48]. Many studies suggested SNPs might affect the activity
58
of metabolism-related key enzyme, therefore generating effects on tumor progres-

sion and drug efficacy. However, with in-depth research, scientists indicated one
SNP or a simple CNA could not influence the whole development of the individual
process of cancer. The occurrence of a cancer is a result of changes of multiple
sites, thus current study is shifting towards several genetic mutation patterns
[9]. In addition, breakthrough progress has been made in strategies for obtaining
DNA information of tumor tissues. Currently, a novel method found to collect
DNA information of tumor tissue is called circulating tumor cell (CTC), which
is a general term for all tumor cells in peripheral blood [49]. Compared to tumor
tissue samples, blood specimens possess more advantages such as less invasive, easy
to acquire, and can be collected repeatedly. It is a typical source of specimens and
convenient to operate in clinical practice, so that significantly improves the value
of aforementioned method [9]. Circulating tumor DNA (ctDNA) means a tumor
cell body that is apoptotic by shedding or released into the circulatory system,
and rapid development of gene sequencing results in that it is able to detect in the
blood [50]. Therefore, ctDNAs are possible to find key mutation sites and served as
biomarkers. Over the past few years, liquid biopsy combined with ctDNA analysis
is helpful and beneficial for the molecular diagnosis and monitoring of cancer.
Moreover, BEAMing (emulsion, amplification, beads, and magnetics) and CAPP-
seq (cancer personalized profiling by deep sequencing) are discovered and used to
quantify ctDNA in blood [51, 52]. Furthermore, there are several unknown things
about ctDNA including its size, existing form, mechanisms of released into blood
stream, and its degradation rate in blood [53]. In summary, the development of
genomics provides the method, important information about genome, and impact-
ful biomarkers for diagnosis of cancer and drives the progress of cancer genomics.
3.2 Transcriptomics
Based on the genetic central rule, DNA through self-replication and transcripts
to form the mRNAs, and finally translates to be a protein. The mRNA is served as a
bridge between gene and protein in biological process and linked genome and phe-
notype. Once variation of gene sequence of mRNA occurs, the amino acid sequence
of the protein will be correspondingly altered. Therefore, the understanding of
transcriptomics is important for addressing functional elements of the genome and
cognizing the development of cancer. The key goal of transcriptomics is to classify
all types of transcripts, reveal the transcriptional structure of the genes, and quan-
tify the expression levels of each transcript during development and under different
conditions. Nowadays, many methods are generated to be used for the study of tran-
scriptome, such as hybridization-or sequence-based approaches [54]. In general, the
way of nucleic acids with hybridization-based is incubation of fluorescently labeled-
complementary DNA (cDNA) from reverse transcription of different mRNAs with
a microarray contained genes of interest, then digitized with a dedicated scanner
and image analysis and finally gene name, clone identifier, and intensity values are
acquired [55]. Furthermore, genomic tiling microarrays are found to provide a more
unerring opinion of the transcriptional activities within a genome [56]. Howbeit,
there are some disadvantages, like relying on the current knowledge of genome
sequence, high background levels owing to cross-hybridization, and both back-
ground and saturation of signals resulted in a limited dynamic range of detection
[57, 58]. Sequence-based strategy is able to detect cDNA sequence but not depend on
the probes. With the development of high-throughput DNA sequencing technique of
NGS, a new method used for mapping and quantifying transcriptome is occurred,
named RNA-seq. It possesses a lot of advantages, for instance, high throughput,
high sensitivity, high resolution, and no reconstructions. RNA-seq is able to analyze
59
the whole transcriptome of any species, including detection of unknown genes or

transcripts, exact identification of the cleavage site, and a variable SNP or untrans-
lated region (UTR region) [16]. In another hand, the research field of noncoding
RNA (ncRNA) should be paid more attention. The ncRNAs consist of tRNA, rRNA,
snoRNA, snRNA, piRNA, miRNA, and lncRNA [59]. Of them, miRNA and lncRNA
are familiar and studied more. MicroRNAs, with a sequence of approximately
21 bp, are a kind of small ncRNAs, which take part in multiple cellular functions
including proliferation, differentiation, metabolism and apoptosis [60]. In general,
TaqMan-based real-time quantitative PCR (RT-qPCR) with separate microRNA-
specific primers and probes is used to detect the expression levels of microRNAs.
The expression of microRNAs is frequently dysregulated in a cancer-specific manner
so that microRNAs are potential to be biomarkers for cancer detection. Many studies
demonstrated the microRNAs as biomarkers for prediction, diagnosis, and prog-
nosis for cancer [9]. However, current studies on the function of miRNAs have not
yet been fully understood, previous studies of miRNAs have found different types
of miRNAs and their effects on oncogenesis and gene expression level of miRNA as
antioncogene. In addition, it is predicted that about 30% of protein-encoding genes
are regulated by miRNAs [61, 62]. lncRNAs execute multiple functions in cells and
are reported as biomarkers in many types of cancers, like breast, lung, gastric, liver,
and prostate cancers [63]. The lncRNAs play a vital role in recognition and treatment
of cancer. Up to now, the biological effects of lncRNAs are still incompletely clear,
but they have already been found to be prolific regulators of many cell processes.
Several lncRNAs overlap with gene promoters, thus transcription of these lncRNAs
might interfere with nucleosome-deleted regions and histone modifications of
nucleosomes in those promoters [64, 65]. Moreover, detection of lncRNA is easily
influenced by anticoagulant such as EDTA, and lncRNA is lightly degraded by other
substance of the blood so that it cannot be preserved for a long time. More researches
are necessary to solve these problems in the future [9].
3.3 Proteomics
Proteins are most direct phenotype characteristics of DNA in biological system.

Proteins are related to multiple cellular mechanisms including cell motility, cell
growth, cell signaling, and protein metabolic process [66]. The study of proteome
is beneficial to the understanding of cancer. The aim of proteomics is to identify
proteins and construct protein pathways and networks to characterize information
and ultimately understand the functional relevance of proteins in cells or organisms
[67]. The proteome is one of the most complex omes among genome, transcriptome,
and proteome. The amount of human proteins and their variants or protein species
are approximately reached to billions [4]. Furthermore, one gene is corresponded
to multiple proteins, known as one gene-multiple proteins model, not one gene-one
protein model so that the complexity of proteome is conceivable [68, 69]. So far,
only the sequence and copy number of DNAs and RNAs in a genome are able to
measure with current technologies. However, a lot of information can be acquired
in a proteome, including amino acid sequence, copy number, splicing, variants,
post-translational modifications (PTMs), spatial conformation, and spatial re-
distribution [16]. Proteomics mainly applies to the detection, identification, and
quantification of the protein in a defined system (cell, tissue, organ, and organelles).
Of detection technologies, gel and gel-free methods are used [68, 69]. Two-
dimensional gel electrophoresis (2DGE), two-dimensional difference in gel electro-
phoresis (2D DIGE), and one-dimensional gel electrophoresis (1DGE) are mainly
involved in gel-based methods [69, 70]. When ones want to detect a certain variants
of a given protein or a kind of PTM with gel-based methods, a specific antibody is
60
necessary to be used [70–72]. Gel-free methods primarily have hydrophobic interac-

tion chromatography (HIC) to separate large bio-molecules, like proteins, C4 or C5
reverse phase liquid chromatography (RPLC) with 300 Å pore-size particles, capil-
lary electrophoresis (CE)-electrospray ionization-mass spectrometry (CE-ESI-MS),
multiplexed gel-eluted liquid fraction entrapment electrophoresis (mGELFrEE;
size-based separation) with 8 parallel glass gel column, and weak-cation exchange
chromatography (WCX) in combination with HIC in a single column with a single
phase (2D-LC; from WCX to HIC mode) [73–80]. Mass spectrometry (MS) plays
an important role in identification of protein variants and PTMs, because the amino
acid sequence of complete proteins, splicing sites and PTM-sites are able to be deter-
mined with MS [69, 71, 81, 82]. Tandem mass spectrometry (MS/MS) can detect
amino acid sequence of a protein, and directly authenticate the errors of amino
acid sequence, variations, and modifications, which causes character of PTMs and
protein variants with different types of mass spectrometers, for instance, matrix-
assisted laser desorption ionization-time of flight-time of flight (MALDI-TOF-TOF),
LTQ Orbitrap system, triple TOF 5600 or 6600 systems and Fourier transform ion
cyclotron resonance (FTICR) with different types of ion fragmentation models
including electron capture dissociation (ECD), electron transfer dissociation (ETD)
and collision induced dissociation (CID). Different types of samples and research
objectives should use identification techniques that are appropriate for them [80].
Quantification of protein is necessary to clarify their biological significance, which
is detected with three main methods, including 2DGE-based quantitative methods,
label-free quantitative techniques like sequential window acquisition of all theo-
retical mass spectra (SWATH) and selected/multiple reaction monitoring (SRM/
MRM), and stable isotope-labeled quantitative approaches including isobaric tags
for relative and absolute quantification iTRAQ [80]. Furthermore, combined with
structural proteomics maybe is better for understanding the biological functions
in biological systems [83, 84]. Also, the study of the protein-protein interaction
analysis and cell signal pathways has become a hot topic. The identification of
protein-protein interactions is meaningful for understanding signal transduction
mechanisms and establishing intracellular signaling networks [4]. Under patho-
logical conditions, the body can secrete several special proteins owing to the other
mRNA synthesis and alternative chromosomal genetic variations involved cancer,
diabetes and Alzheimer disease [85]. Therefore, protein is able to be a biomarker and
proteomics is an important strategy for the study of cancer.
3.4 Metabolomics
Metabolites and proteins are equally important to understand cancer. Metabolites

are small molecules (<1 KDa) produced by metabolism, which can provide func-
tional information that is not directly available from the genome and proteome
in cellular and tissue states [86, 87]. Metabolites are derived from lipids, sugars,
proteins, and nucleic acids in a given biological system, cell, tissue, or body-fluid
[88–90]. The alteration in metabolites is relevant to multiple factors, such as genetic,
environment, internal, external, drug, and dietary factors. These metabolic profiles
are related to the whole biochemical processes that are the starting, intermediate
or final products and provide complex interactions information between the genes
and the environment of a given condition [91, 92]. Metabolites may be capable of
reflecting physiological and pathological processes and monitoring the progres-
sion of a disease, and are helpful to predict, diagnose, and treat [93]. Therefore,
metabolomics is a methodology used to study metabolome, refers to identification
of biochemical and molecular features of metabolome, among different metabo-
lite interactions between genetic/environmental factors and metabolites, and to
61
assessment of biochemical mechanisms associated with a given conditions like

different pathophysiological processes [94]. Generally, two strategies, targeted and
untargeted methods, are mainly employed to detect variations in a metabolome [95, 96].
Targeted metabolomics method concentrates on quantification of the variations of
the hypothesis-driven known metabolite profiling (like metabolites that are pro-
duced from one or more unknown pathways) between or among groups, followed
by multivariate statistical analysis and establishment of mathematical model [95,
97]. Up to now, the familiar techniques used for targeted metabolomics are the triple
quadrupole mass spectrometry (QqQ-MS) in the SRM/MRM modes with optimized
sample extraction and liquid chromatography-mass spectrometry (LC-MS) condi-
tions [98, 99]. The untargeted metabolomics is different from targeted method,
which shows in these aspects, such as no hypothesis-driven strategy, and the whole
comprehensive study variations of metabolome in a biological system without bias
for exploration of metabolite biomarkers for impactful prediction, diagnosis, and
prognostic assessment [80, 96]. The current techniques used to qualify and quantify
the metabolomic variations are nuclear magnetic resonance (NMR)-based methods
and mass spectrometry (MS)-based methods [88, 100–102]. NMR-based methods
involve one-dimensional NMR (1D-NMR), two-dimensional NMR (2D-NMR), and
three-dimensional NMR (3D-NMR). The way to provide chemical structural and
molecular environment information is utilizing the interaction of spin active nuclei
(13C, 1H, 31P, 19F) with electromagnetic fields [100, 101]. NMR-based methods pos-
sess many advantages, including nondestruction of sample, minimal sample prepa-
ration, high reproducibility, relative high throughput, availability of databases, and
availability of molecular dynamic and compartmental information with diffusional
methods. However, overlapping of metabolites, low sensitivity, and high instrumen-
tation cost are its disadvantages [103]. MS-based methods include direct injection
coupled with MS (DIMS), LC-MS, gas chromatography coupled with MS (GC-MS),
capillary electrophoresis coupled with MS (CE-MS), and ion mobility coupled with
MS (IM-MS) [80]. Aforementioned five MS-based methods have its advantages and
limitations, and proper combination helps ones to better study. In clinic, in order to
measure variations in a metabolome, the biological samples are extremely complex,
including cell, tissue extracts and body-fluid. Serum/plasma and urine are com-
monly used body-fluid for metabolomics analysis in all diseases because they are
very easily acquired and prepared, and almost no injury for patients [88, 104, 105].
Additionally, many researches have reported that cerebrospinal fluid (CSF), saliva,
exhaled air, tears, and synovial fluid are likely to be regarded as biomarkers for a
specific disease [80]. Metabolites are important source of biomarkers, and metabo-
lomics methods reasonably adopted are beneficial to predict, diagnose, and evaluate
for cancer.
3.5 Radiomics
Medical imaging technologies, including computed tomography (CT), positron

emission tomography (PET), and magnetic resonance imaging (MRI), are vital to
diagnose and check after treatment for cancer. Medical images provide ones with a
number of information about tumors, which include location and volume of tumor,
probable measurements of diameter, the overall and marginal morphology of the
lesion, the relationship with surrounding tissues, internal heterogeneity, CT and
PET/CT values, MRI signal height and other values. This information is instructive
for the diagnosis of tumors and the decision-making of clinical treatment. However,
it is not able to accurately reflect the morphological and behavioral complexities of
a tumor, with limitation in the assessment of treatment sensitivity and prognosis
[106]. With the rapid development of technology, emerging discipline- radiomics
62
has occurred. Based on excellent computer technology and advanced statistical

methods, radiomics achieves high-throughput extraction and conversion of quanti-
tative features of medical data, and make it serve for clinic decision of cancer [16].
Radiomics has enabled medical imaging to achieve a qualitative to quantitative
transition and provides guidance for clinical treatment, and a large amount of data
has the potential to develop into biomarkers that contribute to further research in
cancer.
4. Application of pattern biomarker for PPPM or PM in cancer
Based on the development of multiomics technology, a series of molecular

alterations in the levels of genome, transcriptome, proteome, metabolome, and
radiome are possible to be detected and measured, which also offer many kinds
of potential biomarkers to ones and are beneficial to well understand and study
for cancer. In order to improve the treatment effect and approach PPPM or PM in
cancer, the methodology of recognition of multiomics-based molecule-pattern
biomarker is presented. The concept “pattern biomarker” refers to several biomark-
ers make up a pattern for precise prediction, diagnosis, and prognostic assessment
in cancer, which can be derived from genome, transcriptome, proteome, metabo-
lome, or radiome, and each pattern biomarker is able to be used as a biomarker
for recognition, therapy, and other-related research of cancer. Many researches
prove that the use of more biomarkers can increase the accuracy of understand-
ing for cancer. For instance, based on somatic cell gene copy number aberrations,
the alterations of gene expression analyzed with genomic and transcriptomic data
and long-term clinical outcomes indicated several potentially important targeted
therapeutic response-related events and mentioned a novel molecular classification
of breast cancer patients [107]. Genomic combined with proteomic data analysis
revealed that PI3K pathway aberrations are popular in hormone receptor-positive
breast cancer, which provides new idea for clinically targeted therapy [108]. Tissue
transcriptomics and urine metabolomics integrated analysis identified four urinary
biomarkers that are more reliable compared to biomarkers derived from single omics
Figure 3.
Application of pattern biomarker in personalized medicine or precision medicine.
63
[109]. Comprehensive analysis of transcriptomic and proteomic data suggested a

highly significant enrichment of gonadotropin-releasing hormone (GnRH) signal-
ing pathway that was not deciphered with single omics dataset in glioblastomas,
which proved the necessity of multiomics research [110]. In addition, the failure of
sorafenib-treated HCCs was employed with an integrated quantitative proteomics
and phosphoproteomics analysis, and found that the targeted drug can effectively
inhibit its target kinase in Raf-Erk-Rsk pathway, but the downstream targets of
Rsk-2 (eIF4B, filamin-A, and so on) were not affected, suggesting that they may be
replaced by another active pathway and lead to treatment failure [111]. However,
there are also many challenges needed to be faced. Considering the tumor heteroge-
neity, individual difference, different stages of tumor development, the recurrence
of tumor, and so on, one designs an ideal model for prediction, prognosis, and
prognostic assessment of cancer in order to further realize PPPM or PM (Figure 3).
5. Conclusion
Cancer is a complex whole-body chronic disease, is involved in multiple causes,

multiple processes, and multiple consequences. On the contrary, the complexity of
cancer exactly provides ones with more opportunities for PPPM or PM in cancer.
The rapid development of genomics, transcriptomics, proteomics, metabolomics,
and radiomics in combination with advanced computation biology and systems
biology drives the development of pattern recognition to find reliable and effective
molecular pattern biomarkers for cancer treatment, and further achieves PPPM or
PM. Multiomics integration analysis is beneficial to better understand cell malig-
nant transformation and tumor progression, clarify molecular mechanisms of a
cancer, discover novel biomarkers and targeted drugs, and improve the effect of
targeted therapies.
Acknowledgements
The authors acknowledge the financial supports from the Xiangya Hospital
Funds for Talent Introduction (to X.Z.), the Hunan Provincial “Hundred Talent
Plan” program (to X.Z.), the National Natural Science Foundation of China
(Grant No. 81572278 and 81272798 to X.Z.), China “863” Plan Project (Grant No.
2014AA020610-1 to X.Z.), and the Hunan Provincial Natural Science Foundation of
China (Grant No. 14JJ7008 to X.Z.).
We declare that we have no financial and personal relationships with other

people or organizations.
Author’s contributions
Z.T. analyzed references and wrote manuscript draft of the book chapter.
C.T. M.L. participated in collection of references and analysis of data. X.Z. con-
ceived the concept, designed the book chapter, and critically revised/wrote the
book chapter, coordinated and was responsible for the correspondence work and
financial support.
64
Acronyms and abbreviations
CAPP-seq cancer personalized profiling by deep sequencing

cDNA complementary DNA
CE capillary electrophoresis
CE-MS capillary electrophoresis coupled with MS
CE-ESI-MS capillary electrophoresis-electrospray
ionization-massspectrometry
CID collision induced dissociation
CNA copy number alterations
CSF cerebrospinal fluid
CT computed tomography
CTC circulating tumor cell
ctDNA circulating tumor DNA
DIMS direct injection coupled with MS
ECM extracellular matrix
ECD electron capture dissociation
ETD electron transfer dissociation
FTICR Fourier transform ion cyclotron resonance
GC-MS gas chromatography coupled with MS
GnRH gonadotropin-releasing hormone
HIC hydrophobic interaction chromatography
lncRNAs long noncoding RNAs
IM-MS ion mobility coupled with MS
iTRAQ isobaric tags for relative and absolute quantification
LC-MS liquid chromatography-mass spectrometry
MAPK mitogen-activated protein kinase
mGELFrEE multiplexed gel-eluted liquid fraction entrapment
electrophoresis
mRNA messenger RNA
MS mass spectrometry
MALDI-TOF-TOF matrix-assisted laser desorption ionization-time of flight-
time of flight
MRI magnetic resonance imaging
mTOR mammalian target of rapamycin
ncRNA noncoding RNA
NGS next-generation sequencing
NMR nuclear magnetic resonance
NPC nasopharyngeal carcinoma
1DGE one-dimensional gel electrophoresis
PCR polymerase chain reaction
PET positron emission tomography
PM precision medicine
PPPM predictive, preventive, and personalized medicine
PI3K/Akt phosphoinositide 3-kinase/protein kinase B
PTMs post-translational modifications
QqQ-MS quadrupole mass spectrometry
RPLC reverse phase liquid chromatography
RT-qPCR real-time quantitative PCR
SNPs single nucleotide polymorphisms
STAT3 signal transducer, and activator of transcription 3
SWATH sequential window acquisition of all theoretical mass
spectra
65
SRM/MRM selected/multiple reaction monitoring

1D-NMR one-dimensional NMR
2D-NMR two-dimensional NMR
3D-NMR three-dimensional NMR
2DGE two-dimensional gel electrophoresis
2D DIGE two-dimensional difference in gel electrophoresis
MS/MS Tandem mass spectrometry
UTR region untranslated region
WCX weak-cation exchange chromatography
Author details
Xanquan Zhan*, Tian Zhou, Tingting Cheng and Miaolong Lu

Key Laboratory of Cancer Proteomics of Chinese Ministry of Health, Xiangya
Hospital, Central South University, Changsha, China
66
References
[1] Block KI, Gyllenhaal C, Lowe L, [9] Cheng T, Zhan X. Pattern recognition
Amedei A, ARMR A, Amin A, for predictive, preventive, and
et al. A broad-spectrum integrative personalized medicine in cancer. The
design for cancer prevention and EPMA Journal. 2017;8:51-60. DOI:
therapy. Seminars in Cancer Biology. 10.1007/s13167-017-0083-9
2015;35(Suppl):S276-S304. DOI:
10.1016/j.semcancer.2015.09.007 [10] Wagner PD, Srivastava S. New
paradigms in translational science
[2] Friedl P, Alexander S. Cancer research in cancer biomarkers.
invasion and the microenvironment: Translational Research. 2012;159:
Plasticity and reciprocity. Cell. 343-353. DOI: 10.1016/j.trsl.2012.01.015
2011;147:992-1009. DOI: 10.1016/j.
cell.2011.11.016 [11] Canonica GW, Bachert C, Hellings P,
Ryan D, Valovirta E, Wickman M, et al.
[3] Maximo V, Lima J, Prazeres H, Soares Allergen immunotherapy (AIT): A
P, Sobrinho-Simoes M. The biology and prototype of precision medicine. World
the genetics of Hurthle cell tumors of Allergy Organization Journal. 2015;8:31.
the thyroid. Endocrine-Related Cancer. DOI: 10.1186/s40413-015-0079-7
2012;19:R131-R147. DOI: 10.1530/
ERC-11-0354 [12] Biomarkers Definitions Working
Group. Biomarkers and surrogate
[4] Hu R, Wang X, Zhan X. Multi-
endpoints: Preferred definitions
parameter systematic strategies for
and conceptual framework. Clinical
predictive, preventive and personalised
Pharmacology and Therapeutics.
medicine in cancer. The EPMA Journal.
2001;69:89-95. DOI: 10.1067/
2013;4:2. DOI: 10.1186/1878-5085-4-2
mcp.2001.113989
[5] Kang M, Buckley YM, Lowe AJ.
[13] Zhai XH, Yu JK, Yang FQ, Zheng S.
Testing the role of genetic factors
Identification of a new protein
across multiple independent invasions
biomarker for colorectal cancer
of the shrub scotch broom (Cytisus
diagnosis. Molecular Medicine Reports.
scoparius). Molecular Ecology.
2012;6:444-448. DOI: 10.3892/
2007;16:4662-4673
mmr.2012.923
[6] Jobling MA. The impact of recent
events on human genetic diversity. [14] Taylor DR, Pavord ID. Biomarkers
Philosophical Transactions of the Royal in the assessment and management of
Society of London. Series B, Biological airways diseases. Postgraduate Medical
Sciences. 2012;367:793-799. DOI: Journal. 2008;84:628-634; quiz 633.
10.1098/rstb.2011.0297 DOI: 10.1136/pgmj.2008.069864
[7] Vogelstein B, Papadopoulos N, [15] Manolio TA. Genomewide

Velculescu VE, Zhou S, Diaz LA Jr, association studies and assessment of
Kinzler KW. Cancer genome landscapes. the risk of disease. The New England
Science. 2013;339:1546-1558. DOI: Journal of Medicine. 2010;363:166-176.
10.1126/science.1235122 DOI: 10.1056/NEJMra0905980
[8] Hoth M. CRAC channels, calcium, [16] Lu M, Zhan X. The crucial role of
and cancer in light of the driver and multiomic approach in cancer research
passenger concept. Biochimica et and clinically relevant outcomes. The
Biophysica Acta. 2016;1863:1408-1417. EPMA Journal. 2018;9:77-102. DOI: doi.
DOI: 10.1016/j.bbamcr.2015.12.009 org/10.1007/s13167-018-0128-8
67
[17] Gonzalez-Angulo AM, Iwamoto T, classification of human clinically

Liu S, Chen H, Do KA, Hortobagyi nonfunctional pituitary adenomas
GN, et al. Gene expression, molecular identified by gene expression profiling
class changes, and pathway analysis and proteomic analyses. Cancer
after neoadjuvant systemic therapy for Research. 2005;65:10214-10222
breast cancer. Clinical Cancer Research.
2012;18:1109-1119. DOI: 10.1158/1078- [25] Samuel N, Hudson TJ. Translating
0432.CCR-11-2762 genomics to the clinic: Implications
of cancer heterogeneity. Clinical
[18] Nosho K, Baba Y, Tanaka N, Shima Chemistry. 2013;59:127-137. DOI:
K, Hayashi M, Meyerhardt JA, et al. 10.1373/clinchem.2012.184580
Tumour-infiltrating T-cell subsets,
molecular changes in colorectal [26] Almendro V, Marusyk A, Polyak K.
cancer, and prognosis: Cohort study Cellular heterogeneity and molecular
and literature review. The Journal of evolution in cancer. Annual Review
Pathology. 2010;222:350-366. DOI: of Pathology. 2013;8:277-302. DOI:
10.1002/path.2774 10.1146/annurev-pathol-020712-163923
[19] Sheltzer JM, Torres EM, Dunham [27] Julien S, Merino-Trigo A, Lacroix L,
MJ, Amon A. Transcriptional Pocard M, Goéré D, Mariani P, et al.
consequences of aneuploidy. Characterization of a large panel of
Proceedings of the National Academy patient-derived tumor xenografts
of Sciences of the United States of representing the clinical heterogeneity
America. 2012;109:12644-12649. DOI: of human colorectal cancer. Clinical
10.1073/pnas.1209227109 Cancer Research. 2012;18:5314-5328.
DOI: 10.1158/1078-0432.CCR-12-0372
[20] Gould CM, Courtneidge SA.
Regulation of invadopodia by the tumor [28] Damia G, D'Incalci M. Genetic
microenvironment. Cell Adhesion & instability influences drug response
Migration. 2014;8:226-235 in cancer cells. Current Drug Targets.
2010;11:1317-1324
[21] Hanahan D, Weinberg RA.
Hallmarks of cancer: The next [29] Marusyk A, Almendro V, Polyak K.
generation. Cell. 2011;144:646-674. Intra-tumour heterogeneity: A looking
DOI: 10.1016/j.cell.2011.02.013 glass for cancer? Nature Reviews.
Cancer. 2012;12:323-334. DOI: 10.1038/
[22] Zhan X, Desiderio DM. The nrc3261
use of variations in proteomes to
predict, prevent, and personalize [30] George O, Koob GF. Individual
treatment for clinically nonfunctional differences in prefrontal cortex function
pituitary adenomas. The EPMA and the transition from drug use to
Journal. 2010;1:439-459. DOI: 10.1007/ drug dependence. Neuroscience and
s13167-010-0028-z Biobehavioral Reviews. 2010;35:232-247.
DOI: 10.1016/j.neubiorev.2010.05.002
[23] Longo DL. Tumor heterogeneity
and personalized medicine. The [31] Zhan X, Desiderio DM. Signaling
New England Journal of Medicine. pathway networks mined from human
2012;366:956-957. DOI: 10.1056/ pituitary adenoma proteomics data.
NEJMe1200656 BMC Medical Genomics. 2010;3:13.
DOI: 10.1186/1755-8794-3-13
[24] Moreno CS, Evans CO, Zhan X,
Okor M, Desiderio DM, Oyesiku NM. [32] Laplante M, Sabatini DM. mTOR
Novel molecular signaling and signaling in growth control and disease.
68
Cell. 2012;149:274-293. DOI: 10.1016/j. [41] Tran B, Dancey JE, Kamel-Reid S,

cell.2012.03.017 McPherson JD, Bedard PL, Brown AM,
et al. Cancer genomics: Technology,
[33] Chen J. Multiple signal pathways discovery, and translation. Journal of
in obesity-associated cancer. Obesity Clinical Oncology. 2012;30:647-660.
Reviews. 2011;12:1063-1070. DOI: DOI: 10.1200/JCO.2011.39.2316
10.1111/j.1467-789X.2011.00917.x
[42] Metzker ML. Sequencing
[34] Janku F, Wheler JJ, Westin SN, technologies—The next generation.
Moulder SL, Naing A, Tsimberidou AM, Nature Reviews. Genetics. 2010;11:
et al. PI3K/AKT/mTOR inhibitors in 31-46. DOI: 10.1038/nrg2626
patients with breast and gynecologic
malignancies harboring PIK3CA [43] Shendure J, Ji H. Next-generation
mutations. Journal of Clinical Oncology. DNA sequencing. Nature Biotechnology.
2012;30:777-782. DOI: 10.1200/ 2008;26:1135-1145. DOI: 10.1038/
JCO.2011.36.1196 nbt1486
[35] Palumbo MO, Kavan P, Miller WH [44] Ansorge WJ. Next-generation

Jr, Panasci L, Assouline S, Johnson N, DNA sequencing techniques. New
et al. Systemic cancer therapy: Biotechnology. 2009;25:195-203. DOI:
Achievements and challenges that lie 10.1016/j.nbt.2008.12.009
ahead. Frontiers in Pharmacology.
2013;4:57. DOI: 10.3389/fphar.2013. [45] Niedringhaus TP, Milanova D, Kerby
00057 MB, Snyder MP, Barron AE. Landscape
of next-generation sequencing
[36] Liu FF. Novel gene therapy technologies. Analytical Chemistry.
approach for nasopharyngeal 2011;83:4327-4341. DOI: 10.1021/
carcinoma. Seminars in Cancer Biology. ac2010857
2002;12:505-515
[46] Meyerson M, Gabriel S, Getz G.
[37] Cheon S. Probability concepts Advances in understanding cancer
and distributions for analyzing large genomes through second-generation
biological data. In: Lee JK, editor. sequencing. Nature Reviews. Genetics.
Statistical Bioinformatics for Biomedical 2010;11:685-696. DOI: 10.1038/nrg2841
and Life Science Researchers. Hoboken:
Willey; 2010. pp. 7-56 [47] Pique-Regi R, Monso-Varona
J, Ortega A, Seeger RC, Triche TJ,
[38] Sanger F, Nicklen S, Coulson Asgharzadeh S. Sparse representation
AR. DNA sequencing with chain- and Bayesian detection of genome copy
terminating inhibitors. Proceedings number alterations from microarray
of the National Academy of Sciences data. Bioinformatics. 2008;24:309-318.
of the United States of America. DOI: 10.1093/bioinformatics/btm601
1977;74:5463-5467
[48] Sherry ST, Ward MH, Kholodov M,
[39] Sanger F, Coulson AR. A rapid Baker J, Phan L, Smigielski EM, et al.
method for determining sequences in dbSNP: The NCBI database of genetic
DNA by primed synthesis with DNA variation. Nucleic Acids Research.
polymerase. Journal of Molecular 2001;29:308-311
Biology. 1975;94:441-448
[49] Sorenson GD, Pribish DM, Valone
[40] Sanger F. Determination FH, Memoli VA, Bzik DJ, Yao SL.
of nucleotide sequences in Soluble normal and mutated DNA
DNA. Bioscience Reports. 1981;1:3-18 sequences from single-copy genes in
69
human blood. Cancer Epidemiology, 4-microRNA signature for clear cell

Biomarkers & Prevention. 1994;3:67-71 renal cell carcinoma metastasis and
prognosis. PLoS One. 2012;7:e35661.
[50] Bettegowda C, Sausen M, Leary RJ, DOI: 10.1371/journal.pone.0035661
Kinde I, Wang Y, Agrawal N, et al.
Detection of circulating tumor [59] Alahari SV, Eastlack SC, Alahari SK.
DNA in early- and late-stage human Role of long noncoding RNAs in
malignancies. Science Translational neoplasia: Special emphasis on prostate
Medicine. 2014;6:224ra24. DOI: 10.1126/ cancer. International Review of Cell and
scitranslmed.3007094 Molecular Biology. 2016;324:229-254.
DOI: 10.1016/bs.ircmb.2016.01.004
[51] Diehl F, Schmidt K, Choti MA,
Romans K, Goodman S, Li M, et al. [60] Reid JF, Sokolova V, Zoni E, Lampis
Circulating mutant DNA to assess A, Pizzamiglio S, Bertan C, et al. miRNA
tumor dynamics. Nature Medicine. profiling in colorectal cancer highlights
2008;14:985-990. DOI: 10.1038/nm.1789 miR-1 involvement in MET-dependent
proliferation. Molecular Cancer Research.
[52] Newman AM, Bratman SV, To J, 2012;10:504-515. DOI: 10.1158/1541-7786
Wynne JF, Eclov NC, Modlin LA,
et al. An ultrasensitive method for [61] Li Y, Cao H, Jiao Z, Pakala SB,
quantitating circulating tumor DNA Sirigiri DN, Li W, et al.
with broad patient coverage. Nature Carcinoembryonic antigen interacts
Medicine. 2014;20:548-554. DOI: with TGF-{beta} receptor and inhibits
10.1038/nm.3519 TGF-{beta} signaling in colorectal
cancers. Cancer Research. 2010;70:
[53] Cheng F, Su L, Qian C. Circulating
8159-8168. DOI: 10.1158/0008-5472
tumor DNA: A promising biomarker in
the liquid biopsy of cancer. Oncotarget.
[62] Liu M, Li CF, Chen HS, Lin LQ,
2016;7:48832-48841. DOI: 10.18632/
Zhang CP, Zhao JL, et al. Differential
oncotarget.9453
expression of proteomics models
[54] Wang Z, Gerstein M, Snyder M.
of colorectal cancer, colorectal
RNA-Seq: A revolutionary tool for benign disease and healthy controls.
transcriptomics. Nature Reviews. Proteome Science. 2010;8:16. DOI:
Genetics. 2009;10:57-63. DOI: 10.1038/ 10.1186/1477-5956-8-16
nrg2484
[63] Houseley J, Rubbi L, Grunstein M,
[55] Duggan DJ, Bittner M, Chen Y, Tollervey D, Vogelauer M. A ncRNA
Meltzer P, Trent JM. Expression modulates histone modification and
profiling using cDNA microarrays. mRNA induction in the yeast GAL gene
Nature Genetics. 1999;21(1 Suppl):10-14 cluster. Molecular Cell. 2008;32:
685-695. DOI: 10.1016/j.molcel.2008.
[56] Yazaki J, Gregory BD, Ecker JR. 09.027
Mapping the genome landscape using
tiling array technology. Current Opinion [64] Pauli A, Valen E, Lin MF, Garber M,
in Plant Biology. 2007;10:534-542 Vastenhouw NL, Levin JZ, et al.
Systematic identification of long
[57] Mishra PJ. MicroRNA noncoding RNAs expressed during
polymorphisms: A giant leap towards zebrafish embryogenesis. Genome
personalized medicine. Personalized Research. 2012;22:577-591. DOI:
Medicine. 2009;6:119-125 10.1101/gr.133009.111
[58] Wu X, Weng L, Li X, Guo C, Pal [65] Ponting CP, Oliver PL, Reik W.
SK, Jin JM, et al. Identification of a Evolution and functions of long
70
noncoding RNAs. Cell. 2009;136: [74] Cummins PM, O'Connor

629-641. DOI: 10.1016/j.cell.2009.02.006 BF. Hydrophobic interaction
chromatography. Methods in Molecular
[66] Karley D, Gupta D, Tiwari A. Biology. 2011;681:431-437. DOI:
Biomarker for cancer: A great promise 10.1007/978-1-60761-913-0_24
for future. World Journal of Oncology.
2011;2:151-157. DOI: 10.4021/wjon352w [75] Hong G, Gao M, Yan G, Guan X,
Tao Q, Zhang X. Optimization of two-
[67] Horgan RP, Kenny LC. dimensional high performance liquid
‘Omic’ technologies: Genomics, chromatographic columns for highly
transcriptomics, proteomics and efficient separation of intact proteins. Se
metabolomics. The Obstetrician and Pu. 2010;28:158-162
Gynaecologist. 2011;13:189-195
[76] Staub A, Zurlino D, Rudaz S,
[68] Stastna M, Van Eyk JE. Analysis of
Veuthey JL, Guillarme D. Analysis
protein isoforms: Can we do it better? of peptides and proteins using sub-2
Proteomics. 2012;12:2937-2948. DOI: μm fully porous and sub 3-μm shell
10.1002/pmic.201200161 particles. Journal of Chromatography.
A. 2011;1218:8903-8914. DOI: 10.1016/j.
[69] Zhan X, Giorgianni F, Desiderio
chroma.2011.07.051
DM. Proteomics analysis of growth
hormone isoforms in the human
[77] Tran JC, Doucette AA. Multiplexed
pituitary. Proteomics. 2005;5:1228-1241
size separation of intact proteins in
[70] Kohler M, Thomas A, Püschel K, solution phase for mass spectrometry.
Schänzer W, Thevis M. Identification Analytical Chemistry. 2009;81:
of human pituitary growth hormone 6201-6209. DOI: 10.1021/ac900729r
variants by mass spectrometry. Journal
of Proteome Research. 2009;8: [78] Sikanen T, Aura S, Franssila S,
1071-1076. DOI: 10.1021/pr800945b Kotiaho T, Kostiainen R. Microchip
capillary electrophoresis-electrospray
[71] Peng F, Li J, Guo T, Yang H, Li M, ionization-mass spectrometry of intact
Sang S, et al. Nitroproteins in human proteins using uncoated Ormocomp
astrocytomas discovered by gel microchips. Analytica Chimica
electrophoresis and tandem mass Acta. 2012;711:69-76. DOI: 10.1016/j.
spectrometry. Journal of the American aca.2011.10.059
Society for Mass Spectrometry.
2015;26:2062-2076. DOI: 10.1007/ [79] Geng X, Ke C, Chen G, Liu P, Wang
s13361-015-1270-3 F, Zhang H, et al. On-line separation
of native proteins by two-dimensional
[72] Ono M, Matsubara J, Honda K, liquid chromatography using a single
Sakuma T, Hashiguchi T, Nose H, column. Journal of Chromatography. A.
et al. Prolyl 4-hydroxylation of 2009;1216:3553-3562. DOI: 10.1016/j.
alpha-fibrinogen: A novel protein chroma.2009.01.085
modification revealed by plasma
proteomics. The Journal of Biological [80] Zhan X, Long Y, Lu M. Exploration
Chemistry. 2009;284:29041-29049. of variations in proteome and
DOI: 10.1074/jbc.M109.041749 metabolome for predictive diagnostics
and personalized treatment
[73] Goheen SC, Engelhorn SC. algorithms: Innovative approach
Hydrophobic interaction high- and examples for potential clinical
performance liquid chromatography of application. Journal of Proteomics.
proteins. Journal of Chromatography. 2018;188:30-40. DOI: 10.1016/j.
1984;317:55-65 jprot.2017.08.020
71
[81] Guo T, Wang X, Li M, Yang H, discovery. Mass Spectrometry Reviews.

Li L, Peng F, et al. Identification 2017;36:115-134. DOI: 10.1002/
of glioblastoma phosphotyrosine- mas.21455
containing proteins with two-
dimensional western blotting [89] Dunn WB, Broadhurst DI, Atherton
and tandem mass spectrometry. HJ, Goodacre R, Griffin JL.
BioMed Research International. Systems level studies of mammalian
2015;2015:134050. DOI: 10.1155/2015/ metabolomes: The roles of mass
134050 spectrometry and nuclear magnetic
resonance spectroscopy. Chemical
[82] Zhan X, Desiderio DM. The human Society Reviews. 2011;40:387-426. DOI:
pituitary nitroproteome: Detection 10.1039/b906712b
of nitrotyrosyl-proteins with two-
dimensional western blotting, and [90] Nicholson JK, Lindon JC, Holmes
amino acid sequence determination with E. 'Metabonomics': Understanding
mass spectrometry. Biochemical and the metabolic responses of living
Biophysical Research Communications. systems to pathophysiological stimuli
2004;325:1180-1186 via multivariate statistical analysis of
biological NMR spectroscopic data.
[83] Zhan X, Wang X, Desiderio DM. Xenobiotica. 1999;29:1181-1189
Mass spectrometry analysis of
nitrotyrosine-containing proteins. Mass [91] Nicholson JK. Global systems
Spectrometry Reviews. 2015;34: biology, personalized medicine and
423-448. DOI: 10.1002/mas.21413 molecular epidemiology. Molecular
Systems Biology. 2006;2:52
[84] Hyung SJ, Ruotolo BT. Integrating
mass spectrometry of intact protein [92] Mirsaeidi M, Banoei MM, Winston
complexes into structural proteomics. BW, Schraufnagel DE. Metabolomics:
Proteomics. 2012;12:1547-1564. DOI: Applications and promise in
10.1002/pmic.201100520 mycobacterial disease. Annals of
the American Thoracic Society.
[85] Deschoolmeester V, Baay M, 2015;12:1278-1287. DOI: 10.1513/
Specenier P, Lardon F, Vermorken JB. AnnalsATS.201505-279PS
A review of the most promising
biomarkers in colorectal cancer: One [93] Everett JR. Pharmacometabonomics
step closer to targeted therapy. The in humans: A new tool for personalized
Oncologist. 2010;15:699-731. DOI: medicine. Pharmacogenomics.
10.1634/theoncologist.2010-0025 2015;16:737-754. DOI: 10.2217/pgs.15.20
[86] Holmes E, Wilson ID, Nicholson JK. [94] Tebani A, Abily-Donval L,

Metabolic phenotyping in health and Afonso C, Marret S, Bekri S. Clinical
disease. Cell. 2008;134:714-717. DOI: metabolomics: The new metabolic
10.1016/j.cell.2008.08.026 window for inborn errors of metabolism
investigations in the post-genomic
[87] Patti GJ, Yanes O, Siuzdak G. era. International Journal of Molecular
Innovation: Metabolomics: The apogee Sciences. 2016;17. DOI: 10.3390/
of the omics trilogy. Nature Reviews. ijms17071167
Molecular Cell Biology. 2012;13:263-269.
DOI: 10.1038/nrm3314 [95] Siskos AP, Jain P, Römisch-Margl W,
Bennett M, Achaintre D, Asad Y, et al.
[88] Khamis MM, Adamko DJ, El-Aneed Interlaboratory reproducibility of a
A. Mass spectrometric based approaches targeted metabolomics platform for
in urine metabolomics and biomarker analysis of human serum and plasma.
72
Analytical Chemistry. 2017;89:656-665. protocols based on LC-MS, GC-MS and

DOI: 10.1021/acs.analchem.6b02930 CE-MS for nontargeted metabolomics
of biological tissues. Bioanalysis.
[96] Mizuno H, Ueda K, Kobayashi Y, 2014;6:1657-1677. DOI: 10.4155/
Tsuyama N, Todoroki K, Min JZ, et al. bio.14.119
The great importance of normalization
of LC-MS data for highly-accurate [103] Markley JL, Brüschweiler R, Edison
non-targeted metabolomics. Biomedical AS, Eghbalnia HR, Powers R, Raftery D,
Chromatography. 2017;31:e3864. DOI: et al. The future of NMR-based
10.1002/bmc.3864 metabolomics. Current Opinion in
Biotechnology. 2017;43:34-40. DOI:
[97] Kitteringham NR, Jenkins RE, 10.1016/j.copbio.2016.08.001
Lane CS, Elliott VL, Park BK. Multiple
reaction monitoring for quantitative [104] Dunn WB, Broadhurst D, Begley P,
biomarker analysis in proteomics Zelena E, Francis-McIntyre S, Anderson
and metabolomics. Journal of N, et al. Procedures for large-scale
Chromatography. B, Analytical metabolic profiling of serum and
Technologies in the Biomedical and Life plasma using gas chromatography and
Sciences. 2009;877:1229-1239. DOI: liquid chromatography coupled to
10.1016/j.jchromb.2008.11.013 mass spectrometry. Nature Protocols.
2011;6:1060-1083. DOI: 10.1038/
[98] Zhou J, Yin Y. Strategies for nprot.2011.335
large-scale targeted metabolomics
quantification by liquid [105] Want EJ, Wilson ID, Gika H,
chromatography-mass spectrometry. Theodoridis G, Plumb RS,
The Analyst. 2016;141:6362-6373 Shockcor J, et al. Global metabolic
profiling procedures for urine
[99] Guo B, Chen B, Liu A, Zhu W,
using UPLC-MS. Nature Protocols.
Yao S. Liquid chromatography-mass
2010;5:1005-1018. DOI: 10.1038/
spectrometric multiple reaction
nprot.2010.50
monitoring-based strategies for
expanding targeted profiling towards
[106] Kumar V, Gu Y, Basu S, Berglund
quantitative metabolomics. Current
A, Eschrich SA, Schabath MB, et al.
Drug Metabolism. 2012;13:1226-1243
Radiomics: The process and the
[100] Kruk J, Doskocz M, Jodłowska E,
challenges. Magnetic Resonance
Zacharzewska A, Łakomiec J, Czaja K, Imaging. 2012;30:1234-1248. DOI:
et al. NMR techniques in metabolomic 10.1016/j.mri.2012.06.010
studies: A quick overview on examples
[107] Curtis C, Shah SP, Chin SF,
of utilization. Applied Magnetic
Resonance. 2017;48:1-21. DOI: 10.1007/ Turashvili G, Rueda OM, Dunning MJ,
s00723-016-0846-9 et al. The genomic and transcriptomic
architecture of 2,000 breast tumours
[101] Marchand J, Martineau E, Guitton reveals novel subgroups. Nature.
Y, Dervilly-Pinel G, Giraudeau P. 2012;486:346-352. DOI: 10.1038/
Multidimensional NMR approaches nature10983
towards highly resolved, sensitive
and high-throughput quantitative [108] Stemke-Hale K, Gonzalez-Angulo
metabolomics. Current Opinion in AM, Lluch A, Neve RM, Kuo WL,
Biotechnology. 2017;43:49-55. DOI: Davies M, et al. An integrative genomic
10.1016/j.copbio.2016.08.004 and proteomic analysis of PIK3CA, PTEN,
and AKT mutations in breast cancer.
[102] Naz S, Moreira dos Santos DC, Cancer Research. 2008;68:6084-6091.
García A, Barbas C. Analytical DOI: 10.1158/0008-5472.CAN-07-6854
73
[109] Nam H, Chung BC, Kim Y,

Lee K, Lee D. Combining tissue
transcriptomics and urine metabolomics
for breast cancer biomarker
identification. Bioinformatics.
2009;25:3151-3157. DOI: 10.1093/
bioinformatics/btp558
[110] Jayaram S, Gupta MK, Raju R,

Gautam P, Sirdeshmukh R. Multi-omics
data integration and mapping of altered
kinases to pathways reveal gonadotropin
hormone signaling in glioblastoma.
OMICS International. 2016;20:736-746
[111] Dazert E, Colombi M, Boldanova T,

Moes S, Adametz D, Quagliata L,
et al. Quantitative proteomics and
phosphoproteomics on serial tumor
biopsies from a sorafenib-treated HCC
patient. Proceedings of the National
Academy of Sciences of the United
States of America. 2016;113:1381-1386.
DOI: 10.1073/pnas.1523434113
74
Chapter 5
HCV Genotyping with Concurrent

Profiling of Resistance-Associated
Variants by NGS Analysis
Kok-Siong Poon, Julian Wei-Tze Tang
and Evelyn Siew-Chuan Koay
Abstract
Determination of viral characteristics including genotype (GT), subtype

(ST) and resistance-associated variants (RAVs) profile is important in assigning
direct-acting antivirals regimes in HCV patients. To help achieve the best clini-
cal management of HCV patients, a routine diagnostic laboratory should aim at
reporting accurate viral GT/ST and RAVs using a reliable diagnostic platform of
choice. A laboratory study was conducted to evaluate performance characteristics
of a new commercial next-generation sequencing (NGS)-based HCV genotyping
assay in comparison to another widely used commercial line probe assay for HCV
genotyping. Information on RAVs from deeply sequenced NS3, NS5A and NS5B
regions in samples classified as HCV 1a and 1b was harnessed from the fully auto-
mated software. Perfect (100%) concordance at HCV genotype level was achieved
in GT2 (N = 13), GT3 (N = 55) and GT5 (N = 7). NGS refined the ST assignment in
GTs 1, 4 and 6, and resolved previously indeterminate GTs reported by line probe
assay. NGS was found to have consistent intra- and inter-run reproducibility in
terms of genotyping, subtyping and RAVs identification. Detection of infections
with multiple HCV GTs or STs is feasible by NGS. Deep sequencing allows sensitive
identification of RAVs in the GT 1a and 1b NS3, NS5A and NS5B regions, but the list
of target RAVs is not exhaustive.
Keywords: resistance-associated variants, next-generation sequencing,

hepatitis C, HCV genotyping, NGS
1. Introduction
Due to the genetic diversity of the hepatitis C virus (HCV), its accurate genotyp-
ing is still currently challenging despite the use of modern molecular techniques. In
addition to the six widely-recognised HCV genotypes, a newly identified genotype
(GT) 7 was reported in 2015 [1]. Molecular methods including reverse hybridization,
real-time PCR and Sanger sequencing are commonly utilised for HCV genotyping and
subtyping in clinical laboratories. HCV genotype and subtype (ST) have been the crit-
ical factors in decision-making for administering interferon-based therapies for the
past decade [2]. According to the latest AASLD guidelines [3], determination of viral
characteristics including GT, ST and resistance-associated variants (RAVs) profile is
important in assigning direct-acting antivirals (DAAs) regimes in HCV patients.
75
To help achieve the best clinical management of HCV patients, a routine diag-
nostic laboratory should aim at minimising reporting out non-informative HCV
genotyping results which are due to inherent limitations of the diagnostic platform
of choice. In general, about 2–8.5% of HCV positive samples have been reported to
carry “indeterminate” GTs by several commercial assays [4–9]. To tackle uncertain-
ties in determining HCV GT and ST, Sanger sequencing could be utilised to resolve
indeterminate or discordant GTs or ST results produced by commercial assays
[10, 11]. Despite the ability to provide definitive genotyping information most of
the time, unfavourable features of Sanger sequencing including low throughput,
time-consuming procedures and relatively high costs, pose a barrier to it becoming
routinely adopted as a first-line genotyping method. With the advent of next-gener-
ation sequencing (NGS), limitations of probe-based genotyping assays and Sanger
sequencing for HCV genotyping can be overcome. NGS provides a high-resolution
means for direct sequence-based interrogation of the HCV genome. Moreover, NGS
also allows concurrent profiling of RAVs where such value-added feature is highly
relevant for the clinical management of HCV infection with appropriate use of DAAs.
In the present study, the Sentosa SQ HCV genotyping assay (hereinafter referred
to as Vela NGS) (Vela Diagnostics, Singapore) which primarily interrogates the
NS5B region of HCV GTs 1–6 by ion torrent-based NGS technology, was evaluated
in comparison to the VERSANT HCV Genotype 2.0 Assay (hereinafter referred
to as LiPA) (Siemens Healthineers, Erlangen, Germany). HCV indeterminate GTs
previously reported in clinical samples by LiPA were resolved using Vela NGS assay
with further confirmation by Sanger sequencing. Information on RAVs was also
harnessed from deeply sequenced NS3, NS5A and NS5B regions in samples classi-
fied as HCV 1a and 1b using Vela NGS.
2. Study design
2.1 Clinical samples
This study was performed on residual sera or plasma from 222 clinical specimens
previously received for routine genotyping using the VERSANT HCV Genotype 2.0
Line Probe Assay (Siemens Healthineers, Erlangen, Germany). All samples were stored
at -80°C post-LiPA analysis and were only thawed prior to re-analysis by NGS and
Sanger sequencing. All samples were de-identified for anonymisation purposes, and
hence, the treatment histories remain unknown and cannot be traced. These were all
residual samples, which would otherwise be discarded, and were used for the purposes
of assay validation only. In such situations, ethics approval is not normally required, as
all samples could not be linked back to the original patients after anonymisation.
2.2 NGS by Sentosa SQ HCV genotyping assay
In this study, NGS was performed using Sentosa SQ HCV Genotyping Assay
(4 × 16) (Vela Diagnostics, Singapore) according to the manufacturer’s instruc-
tions. The workflow started with automated extraction of total nucleic acids from
530 μL of sera or plasma using Sentosa SX Virus Total Nucleic Acid Plus II kit (Vela
Diagnostics) on Sentosa SX101 (Vela Diagnostics). PCR amplification of the HCV
NS3, NS5A and NS5B regions was performed on Veriti 96-Well Thermal Cycler
(Applied Biosystems, CA, USA). In every individual run, a pooled library containing
barcoded amplicons of 15 clinical samples and one system control, was prepared by
Sentosa SX101. The pooled library was subject to sequencing template preparation
and enrichment on Sentosa ST401 (Vela Diagnostics). Sequencing data generated
76
HCV Genotyping with Concurrent Profiling of Resistance-Associated Variants by NGS Analysis
by Sentosa SQ301 (Vela Diagnostics) was automatically channelled for primary

and subsequent secondary analyses using Sentosa SQ Suite (Vela Diagnostics) and
Sentosa SQ Reporter (Vela Diagnostics), respectively. Auto-generated quality con-
trol and pathology reports containing technical information, viral typing, and RAVs
(available only for GTs 1a and 1b) results were manually reviewed, respectively.
2.3 VERSANT HCV Genotype 2.0 Line Probe Assay
Total nucleic acids were extracted from 200 μL sera or plasma using EZ1 Virus
Mini Kit v2.0 (QIAGEN, Hilden, Germany) on Biorobot EZ1 (QIAGEN). Using
VERSANT HCV Genotype 2.0 Line Probe Assay (LiPA) (Siemens Healthineers), a
one-step reverse transcription-polymerase chain reaction (RT-PCR) amplifying the
5’UTR and core regions was performed on GeneAmp PCR System 9700 (Applied
Biosystems). Reverse hybridisation, washing and colour development steps were
performed on Autoblot 3000H (Fujirebio Europe, Gent, Belgium). For GT and ST
determination, band patterns were manually scored by aligning the strips to an
interpretation chart provided by the manufacturer.
2.4 Sanger sequencing
Sanger sequencing was performed on samples previously reported by LiPA as

indeterminate genotype. A primary PCR amplification of a 454 bp fragment of the
NS5B region was initially attempted using primers 5Bo8254 and 5Bo8707 [12]. In
samples with PCR failure using the above-mentioned primers, a secondary PCR
amplifying a 446 bp fragment of the 5’UTR/core regions was subsequently per-
formed using primers UTR45 and Cor490 [12]. PCR products from the amplifiable
gene segments were subjected to direct sequencing with BigDye Terminator v3.1
Cycle Sequencing kit (Applied Biosystems) using the respective PCR primers on a
3130XL Genetic Analyzer (Applied Biosystems).
2.5 Sequence analysis
Sequence analysis was performed by querying the nucleotide sequences obtained

from Sanger sequencing in the Los Alamos hepatitis C sequence database [13]. For Vela
NGS, assembled contigs were downloaded from the Sentosa SQ Reporter software. In
samples with discordant results between LiPA and Vela NGS, NGS contigs were uploaded
to the Los Alamos hepatitis C sequence database [13] to verify Vela NGS results.
3. Results
3.1 Concordance between results generated by the Vela NGS and Versant
platforms at GT and ST levels
The Vela NGS results at both GT and ST levels were tabulated in Table 1 for
170 clinical samples with GT and/or ST results from LiPA. Perfect (100%) con-
cordance at HCV genotype level was achieved in GT 2 (N = 13), GT 3 (N = 55) and
GT 5 (N = 7). For samples reported by LiPA as GT 1 (N = 40), 20% (N = 8) gave
discrepant results when compared to Vela NGS. These samples had been previ-
ously classified by LiPA as either GT 1a with core inconclusive, GT 1b with 96.1%
homology, GT 1b with core inconclusive, or GT 1b with core not available, due to
their unconventional band patterns. There was no discrepancy between samples
firmly reported as GT 1a and GT 1b by LiPA. In samples reported as GT 4 (N = 16)
77
by LiPA, 43.8% (N = 7) were found to be GT 3 by Vela NGS. Two samples (5.1%)

originally reported by LiPA as GT 3 were classified by Vela NGS as GT 6 samples.
At ST level, Vela NGS reclassified 1 sample previously assigned as HCV 1a with
core inconclusive by LiPA as 1c. Two samples each reported as 4a/4c/4d and 4e by
LiPA, respectively, were reclassified as 4n and 4o by Vela NGS. Another 29 GT 6
(ST c-l) samples reported by LiPA were reassigned by Vela NGS as 6e/6u (N = 1),
6j (N = 1), 6m (N = 9), and 6n (N = 18), respectively. One sample with LiPA 6m
(77.9% homology) was reassigned as 6u by Vela NGS.
3.2 Verification of contig sequences generated by the Vela NGS in samples with
discordant results
Of the 170 samples tested, there were 104 agreements at both GT and ST levels,
49 partial agreements at genotype but not the subtype levels, and 117 discordant
results generated by LiPA and Vela NGS (Table 1). At GT level, the calculated
Cohen’s Kappa is 0.869 (95% confidence interval: 0.810–0.928), suggesting good
strength of agreement between the two assays. The 66 NGS contig sequences of
samples with partial agreement or discordant results were submitted to the online
analysis in the Los Alamos hepatitis C sequence database. HCV GT and ST called by
Vela NGS were verified in all 66 contigs.
3.3 Intra-run and inter-run reproducibility on GT and ST calling by Vela NGS
HCV genotyping and subtyping results were found to be reproducible for a panel
of 5 samples with different HCV GT/ST including 1a, 1b, 2a, 3a and 3b tested in
triplicates within a single run on the Vela NGS platform (Figure 1a). For inter-run
reproducibility testing (Figure 1b), GT and ST results were consistently reported in
another panel of 7 samples including 1a, 1b, 2b, 3a, 4d, 5a and 6n, which were repeat-
edly tested in three separate runs on different days. Details of viral load and median
coverage of the targeted NS5B region are depicted in Figure 1a and b, respectively.
3.4 RAV analysis in GT 1 samples reported by the Vela NGS platform
In the current Vela NGS assay, a list of variants differing from the wild-type codons
are detectable for HCV 1a and 1b. The 16 target codons in the NS3 gene are 36, 41, 43, 54,
55, 80, 109, 122, 132 (1a only), 138, 155, 156, 158, 168, 170 (1b only) and 175 (1b only).
For NS5A, variants at nine codons including 28 (1a only), 30 (1a only), 31, 32, 54 (1b
only), 58, 62 (1b only), 92 and 93, are detectable. Eight codons in the NS5B gene includ-
ing 414, 419, 422, 423, 495, 499 (1b only), 554 and 559, are also covered in this assay.
Of 13 GT 1a samples (Table 2), five were found to carry at least one target variant
in the NS3 gene. Notably, two samples carried the Q80K RAV. For NS5A, the M28A
variant was detected in one sample in which NS3 Q80K was also present. None of the
GT 1a samples was found to carry any of target variants in the NS5B gene.
Of 18 HCV 1b samples (Table 2), five were detected with at least one target vari-
ant in the NS3 gene. Twelve samples were identified with at least one target variant
in the NS5A gene. For NS5B, the P495A and V499A variants were detected in one
and eight samples, respectively. Notably, there were four samples detected with at
least one target variant in each of the NS3, NS5A and NS5B genes.
3.5 Intra-run and inter-run reproducibility on variant calling and frequency
In intra-run reproducibility analysis, the Q80K variant was reproducibly

detected in the NS3 gene of the GT 1a samples. Another two variants, namely Q54H
78
Comparison of GT and ST distribution in 170 samples tested by both LiPA and Vela NGS.
Table 1.
79
Figure 1.
Precision studies on the Vela NGS. (a) Intra-run and (b) inter-run reproducibility on median read depth
were tested on 5 and 7 clinical specimens, respectively. For RAV analysis, variants were called with reproducible
frequency (c) within a run (intra-run) and (d) between runs (inter-run).
No ID GT1 RAVs (variant frequency)

STs
NS3 NS5A NS5B
1 R02-BC02 1a S122G (99.21%), D168E – –
(97.07%)
2 R02-BC03 1a V55A (91.44%) – –
3 R02-BC04 1a Q80K (25.63%) M28V (99.47%) –
4 R02-BC05 1a Q80K (4.84%) – –
5 R13-BC13 1a D168E (51.43%) – –
6 R01-BC02 1b Q80K (55.29%) – V499A (98.15%)

M175L (87.81%)
7 R01-BC03 1b – – V499A (97.03%)
8 R01-BC04 1b – L31M (22.03%), Q54H V499A (33.65%)

(98.82%)
9 R01-BC05 1b – Q54H (99.11%), Y93H –

(99.73%)
10 R01-BC06 1b Q80L (99.52%), S122G Q54H (99.05%) V499A (97.91%)

(9.99%)
11 R01-BC07 1b – Q54H (98.76%), Y93H –

(99.61%)
12 R01-BC08 1b – Q54H (99.22%), Q62E –

(99.04%)
13 R01-BC09 1b S122G (97.69%) Q54H (99.21%), Q62E P495A 8.83%

(51.64%)
14 R01-BC11 1b – Y93H (99.24%) –
80
No ID GT1 RAVs (variant frequency)

STs
NS3 NS5A NS5B
15 R02-BC07 1b – Q54H (99.37%) V499A 98.9%
16 R02-BC-11 1b Q80R (92.29%) Q62E (5.61%) V499A 95.15%
17 R11-BC14 1b M175L (99.97%) Y93H (99.8%) V499A 98.7%
18 R12-BC14 1b – Q54H (80.35%), Y93H –

(8.07%)
19 R12-BC15 1b – Q54H (98.82%), Q62R –

(99.79%)
In this study, RAVs with variant frequency less than 1% are not shown.
Table 2.
List of resistance-associated variants (RAVs) identified in GT 1a and 1b samples by Vela NGS.
No LiPA results (bands) Vela Sanger sequencing Concordance at

NGS GT or ST level
NS5B 5‘UTR/core
1 Indeterminate (3,6,16,24) 6n Not amplified 6n GT & ST
4 Indeterminate (6,7,24) 3b 3b Not done GT & ST
5 Indeterminate (6,7,24) 6m/6u Not amplified 6e/6d GT only
6 Indeterminate (6,7) 6u 6u/6n Not done GT & ST
7 Indeterminate (6,7) 6u 6m/6n Not done GT only
8 Indeterminate (6,7) 6u 6n/6a Not done GT only
9 Indeterminate (6) 6m/6l 6d/6e Not done GT only
10 Indeterminate (17,24) 3b 3b Not done GT & ST
13 Indeterminate (7,8,14,15,24) 3a 3a Not done GT & ST
14 Indeterminate (7,13,17,18) 3b 3b Not done GT & ST
15 Indeterminate (7,13,17,18,24) 3b 3b Not done GT & ST
18 Indeterminate (3,4,13,25) 1a 1a Not done GT & ST
19 Indeterminate (3,4,7,13,25) 1a 1a Not done GT & ST
20 Indeterminate (3,4,7,13,24) 6e 6e Not done GT & ST
21 Indeterminate (3,4,6,7,13,24) 6e Not amplified 6e/6d GT & ST
22 Indeterminate (5,9,21,24) 6a Not amplified 6a GT & ST
23 Indeterminate (5,6,9,17,18) 4a Not amplified 4a GT & ST
24 Indeterminate (5,9,10,13,14,15,24) 2a & Two mixed Not done Likely mixed

3a sequences infections
26 Indeterminate (24) 3b 3b Not done GT & ST
81
No LiPA results (bands) Vela Sanger sequencing Concordance at

NGS GT or ST level
NS5B 5‘UTR/core
27 Indeterminate (24) 3b 3b Not done GT & ST
28 Indeterminate (4,5,9,16,21,24) 6a 6a Not done GT & ST
29 Indeterminate (4,9,21) 6a Not amplified 6a GT & ST
30 Indeterminate 3b 3b Not done GT & ST

(3,4,6,13,17,18,24,26)
31 Indeterminate (6) 6n/6a 6d/6u Not done GT only
33 Indeterminate (6) 6u 6u Not done GT & ST
35 Indeterminate (7) 6u Not amplified 6v/6l/6d/6k GT only
37 Indeterminate (3,4,5,16,25) 1a Not amplified 1a GT & ST
38 Indeterminate (5,6,18,24) 3k 3k Not done GT & ST
40 Indeterminate (3,4,6,16,24) 6q 6q Not done GT & ST

PCR amplification for NS5B was first attempted in all 40 specimens. A secondary PCR amplifying 5'UTR/core
regions were performed in samples with unsuccessful amplification of NS5B. Sanger sequencing was performed on
PCR amplicon obtained.
Table 3.
Comparison of genotyping results produced by the Vela NGS and Sanger sequencing methods in 40 specimens
with indeterminate genotypes by LiPA.
and V499A were also repeatedly identified in the NS5A and NS5B genes of the GT 1b
sample, respectively. Variant frequencies of the three variants were highly repro-
ducible within run (Figure 1c).
In the inter-run reproducibility study, NS3 S122G and NS5B V499A variants were
tested. Variant frequencies of the two variants were found to be highly reproducible
among the three separate runs (Figure 1d).
3.6 Vela NGS assigned HCV GT and ST to samples with indeterminate

LiPA results
Forty specimens, which were previously reported as HCV indeterminate GT by

LiPA, were subject to Vela NGS analysis. Sanger sequencing were successfully per-
formed on NS5b (N = 30) or 5’UTR/core (N = 10) regions in 40 samples (Table 3).
Of the 40 samples with Sanger sequencing results, Vela NGS results were confirmed
at GT level in 39 samples (97.5%). In a sample with LiPA complex band patterns (5,
9, 10, 13, 14, 15 & 24), a mixed genotypes of GT 2a and GT 3a were assigned by Vela
NGS. Sanger sequencing on NS5B showed overlapping nucleotide base calls in the
overall sequences, in which putative mixed infection with two different HCV GTs
was likely inferred.
4. Discussion
The application of NGS assays to analyse quasispecies HCV genomes has been
increasing in recent years. Several laboratory-developed NGS assays had been
82
previously described in the literature for phylogenetic studies [14], outbreak inves-
tigation [15, 16], characterisation of HCV full genome [17, 18] and identification
of HCV GT and ST in clinical samples [19, 20]. However, there are fewer reports
of adoption of NGS assays in routine HCV genotyping. In 2016, Vela NGS became
available as a CE-IVD certified commercial kit for diagnostic use in the clinical
laboratories. In this study, we report the performance characteristics of Vela NGS in
comparison to the widely used LiPA assay for HCV genotyping.
The performance of Vela NGS in determining the HCV GT and ST in the clinical
specimens had been discussed in several previous studies [21–23]. Perfect agree-
ment at GT level was observed between Vela NGS and LiPA in a study by Manee
et al. [21]. Samples with unclear ST results in GTs 2, 3, 4 and 6 reported by LiPA
were each assigned with a specific subtype after subject to Vela NGS analysis. Dirani
et al. [22] also performed a direct comparison of GT and ST calling between Vela
NGS and LiPA for samples from patients infected with HCV GTs including GT 1, 2,
3 and 4, and found a high concordance (>99%) at GT level between the two tests.
Vela NGS was also found to have better performance in assigning HCV STs among
the four GTs when compared to LiPA [22]. In another study by Rodriguez et al. [23],
Vela NGS achieved high concordance rates with Sanger sequencing in assigning GTs
1 to 6, 1a and 1b STs, and other STs for GTs 4, 5 and 6. Discrepant calls at ST level
was mainly found among HCV GTs 1 and 2 between Vela NGS and Sanger sequenc-
ing; the latter was used as the reference method to sequence the 286 bp segment of
NS5B for which phylogenetic analysis was performed.
In the present study, discrepancy in results was mainly observed in samples with
LiPA GT 1b with incomplete or missing bands at the core region. In this particular
result group, GT 6 with different STs were assigned by Vela NGS. This observation
was not unexpected as it has been specified in the LiPA interpretation chart that GT
6 (STs c-1) cannot be differentiated from ST 1a and 1b without additional informa-
tion from the core region sequence. Among LiPA GT 4 samples, all ST 4h were
reassigned as GT 3 by Vela NGS. Some geographical regions, for example, Southeast
Asia, where GT 6 is highly prevalent [24], could thus be impacted more by this mis-
classification with the use of LiPA method.
In contrast to LiPA which utilises primarily the 5’UTR in GTs 1-6 and core regions
for the discrimination of GT 6 STs c-l from 1a and 1b, Vela NGS targets the non-
structural genes implicated in both accurate genotyping/subtyping and resistance to
DAAs. The LiPA is known to be poor at detecting and identifying recombinant forms
of HCV [25]. Due to the assay design of Vela NGS, this may also pose a problem for
this platform, despite the application of NGS technology. The HCV recombinant
forms can be accurately detected via sequencing of recombination breakpoint
junctions or the whole HCV genome [26]. For example, in our study, one previously
LiPA-indeterminate sample was reported by the Vela NGS to have mixed HCV infec-
tions with HCV 2a and 3a. This NGS finding was confirmed by Sanger sequencing in
which overlapping Sanger electropherograms were observed for NS5B.
The Vela NGS offers information on RAVs in HCV 1a or 1b positive samples,
where such profiling will be useful when prescribing DAA regimes, and detecting
of baseline or emerging RAVs. Targeted assays had been previously developed to
identify a specific RAV [27, 28]. RAVs which are found at levels with at least 15%
variant frequency, at baseline, are known to confer resistance to certain DAAs [29],
and therefore may impact on the effectiveness of DAA treatment [30]. Vela NGS tar-
gets relevant RAVs in three non-structural gene segments (NS3, NS5A and NS5B) of
HCV 1a and 1b, and although the RAV profiling is comprehensive but not exhaustive
due to the assay design, any baseline RAVs present in any of these DAA target genes,
can affect the therapeutic effectiveness [31]. In our study, four HCV 1b samples were
found to harbour variants in all three NS3, NS5A and NS5B genes concurrently.
83
5. Conclusions
In conclusion, the genotyping results of the Vela NGS were found to be highly
concordant with those of the LiPA method. Vela NGS refined the ST assignment in
GT 6 and resolved previously indeterminate GTs reported by LiPA. Technically, the
HCV Vela NGS was found to have consistent intra- and inter-run reproducibility in
terms of GT and ST calling and RAVs identification. Detection of infections with
multiple HCV GTs or STs is feasible by Vela NGS. Due to the assay design which
relies on investigating the HCV sub-genomic regions, HCV recombinant strains
may still be potentially missed. Deep sequencing allows sensitive identification of
RAVs in the GT1a and 1b NS3, NS5A and NS5B regions, but the list of target RAVs
is not exhaustive. We would also suggest the RAVs detection spectrum should be
extended to cover GTs other than HCV 1a and 1b, namely GTs 2-6.
Acknowledgements
We thank Cui-Wen Chua, Mui-Joo Khoo and Lily Chiu of the Department of
Laboratory Medicine at the National University Hospital, Singapore, for their
technical assistance in performing the NGS and LiPA analysis. We also thank Vela
Diagnostics Singapore for funding the NGS reagents in this study.
The authors have no conflict of interest to declare.
Author details
Kok-Siong Poon1, Julian Wei-Tze Tang2 and Evelyn Siew-Chuan Koay1,3*
1 Department of Laboratory Medicine, National University Hospital, Singapore
2 University Hospitals of Leicester NHS, Leicester, United Kingdom
3 Department of Pathology, Yong Loo Lin School of Medicine, National University

of Singapore, Singapore
84
References
[1] Murphy DG, Sablon E, Chamberland J, cobas® HCV GT is a new tool that
Fournier E, Dandavino R, et al. accurately identifies Hepatitis C virus
Hepatitis C virus genotype 7, a new genotypes for clinical practice. PLoS
genotype originating from central One. 2017;12:e0175564
Africa. Journal of Clinical Microbiology.
2015;53:967-972 [9] Benedet M, Adachi D, Wong A,
Wong S, Pabbaraju K, Tellier R, et al.
[2] Ghany MG, Strader DB, Thomas DL, The need for a sequencing-based assay
Seeff LB. Diagnosis, management, and to supplement the Abbott m2000
treatment of hepatitis C: An update. RealTime HCV Genotype II assay:
Hepatology. 2009;49:1335-1374 A 1 year analysis. Journal of Clinical
Virology. 2014;60:301-304
[3] ASLD-IDSA HCV Guidance Panel.
Hepatitis C guidance 2018 update: [10] Larrat S, Poveda JD, Coudret C,
AASLD-IDSA recommendations Fusillier K, Magnat N, et al. Sequencing
for testing, managing, and treating assays for failed genotyping with the
hepatitis C virus infection. Clinical versant hepatitis C virus genotype assay
Infectious Diseases. 2018 (LiPA), version 2.0. Journal of Clinical
Microbiology. 2013;51:2815-2821
[4] Germer JJ, Majewski DW, Rosser M,
Thompson A, Mitchell PS, et al. [11] Chueca N, Rivadulla I, Lovatti R,
Evaluation of the TRUGENE HCV Reina G, Blanco A, et al. Using NS5B
5'NC genotyping kit with the new sequencing for hepatitis C virus
GeneLibrarian module 312 for genotyping reveals discordances with
genotyping of hepatitis C virus from commercial platforms. PLoS One.
clinical specimens. Journal of Clinical 2016;11:e0153754
Microbiology. 2003;41:4855-4857
[12] Quer J, Gregori J, Rodríguez-
[5] Verbeeck J, Stanley MJ, Shieh J, Celis L, Frias F, Buti M, Madejon A, et al.
Huyck E, et al. Evaluation of Versant High-resolution hepatitis C virus
hepatitis C virus genotype assay (LiPA) subtyping using NS5B deep sequencing
2.0. Journal of Clinical Microbiology. and phylogeny, an alternative to
2008;46:1901-1906 current methods. Journal of Clinical
Microbiology. 2015;53:219-226
[6] González V, Gomes-Fernandes M,
Bascuñana E, Casanovas S, Saludes V, [13] Kuiken C, Yusim K, Boykin L,
et al. Accuracy of a commercially Richardson R. The Los Alamos hepatitis
available assay for HCV genotyping C sequence database. Bioinformatics.
and subtyping in the clinical 2005;1:379-384
practice. Journal of Clinical Virology.
2013;58:249-253 [14] Gonçalves Rossi LM, Escobar-
Gutierrez A, Rahal P. Multiregion
[7] Némoz B, Roger L, Leroy V, Poveda JD, deep sequencing of hepatitis C virus:
Morand P, et al. Evaluation of the An improved approach for genetic
cobas® GT hepatitis C virus genotyping relatedness studies. Infection, Genetics
assay in G1-6 viruses including low and Evolution. 2016;38:138-145
viral loads and LiPA failures. PLoS One.
2018;13:e0194396 [15] Escobar-Gutiérrez A, Vazquez-
Pichardo M, Cruz-Rivera M, Rivera-
[8] Fernández-Caballero JA, Alvarez M, Osorio P, Carpio-Pedroza JC, et al.
Chueca N, Pérez AB, García F. The Identification of hepatitis C virus
85
transmission using a next-generation [22] Dirani G, Paesini E, Mascetra

sequencing approach. Journal of Clinical E, Farabegoli P, Dalmo B, et al. A
Microbiology. 2012;50:1461-1463 novel next generation sequencing
assay as an alternative to currently
[16] Caraballo Cortes K, Rosińska M, available methods for hepatitis C virus
Janiak M, Stępień M, Zagordi O, et al. genotyping. Journal of Virological
Next-generation sequencing analysis of Methods. 2018;251:88-91
a cluster of hepatitis C virus infections
in a haematology and oncology center. [23] Rodriguez C, Soulier A, Demontant V,
PLoS One. 2018;13:e0194816 Poiteau L, Mercier-Darty M, et al. A
novel standardized deep sequencing-
[17] Salmona M, Caporossi A, Simmonds based assay for hepatitis C virus
P, Thélu MA, Fusillier K, et al. First genotype determination. Scientific
next-generation sequencing full- Reports. 2018;8:4180
genome characterization of a hepatitis
C virus genotype 7 divergent subtype. [24] Bunchorntavakul C,
Clinical Microbiology and Infection. Chavalitdhamrong D, Tanwandee T.
2016;22:e1-e947 Hepatitis C genotype 6: A concise
review and response-guided therapy
[18] Thomson E, Ip CL, Badhan A, proposal. World Journal of Hepatology.
Christiansen MT, Adamson W, et al. 2013;5:496-504
Comparison of next-generation
sequencing technologies for [25] Schuermans W, Orlent H,
comprehensive assessment of full-length Desombere I, Descheemaeker P, Van
hepatitis C viral genomes. Journal of Vlierberghe H, et al. Heads or tails:
Clinical Microbiology. 2016;54:2470-2484 Genotyping of hepatitis C virus
concerning the 2k/1b circulating
[19] Pedersen MS, Fahnøe U, Hansen TA, recombinant form. International Journal
Pedersen AG, Jenssen H, et al. A of Molecular Sciences. 2016;17
near full-length open reading frame
next generation sequencing assay [26] Iles JC, Njouom R,
for genotyping and identification Foupouapouognigni Y, Bonsall D,
of resistance-associated variants in Bowden R, et al. Characterization
hepatitis C virus. Journal of Clinical of hepatitis c virus recombination
Virology. 2018;195:49-56 in Cameroon by use of nonspecific
next-generation sequencing.
[20] Del Campo JA, Parra-Sánchez M, Journal of Clinical Microbiology.
Figueruela B, García-Rey S, Quer J, et al. 2015;53:3155-3164
Hepatitis C virus deep sequencing for
sub-genotype identification in mixed [27] Chui CK, Dong WW, Joy JB, Poon
infections: A real-life experience. AF, Dong WY, et al. Development
International Journal of Infectious and validation of two screening assays
Diseases. 2018;67:114-117 for the hepatitis C virus NS3 Q80K
polymorphism associated with reduced
[21] Manee N, Thongbaiphet N, response to combination treatment
Pasomsub E, Chantratita W. Clinical regimens containing simeprevir.
evaluation of a newly developed Journal of Clinical Microbiology.
automated massively parallel 2015;53:2942-2950
sequencing assay for hepatitis C virus
genotyping and detection of resistance- [28] Vicenti I, Falasca F, Sticchi L,
association variants Comparison with a Bruzzone B, Turriziani O, et al.
line probe assay. Journal of Virological Evaluation of a commercial real-time
Methods. 2017;249:31-37 PCR kit for the detection of the Q80K
86
polymorphism in plasma from HCV

genotype 1a infected patients. Journal of
Clinical Virology. 2016;76:20-23
[29] Pawlotsky JM. Hepatitis C virus

resistance to direct-acting antiviral
drugs in interferon-free regimens.
Gastroenterology. 2016;151:70-86
[30] Yoshida K, Hai TA, Teranishi Y,

Kozuka R, et al. Long-term follow-up
of resistance-associated substitutions
in hepatitis C virus in patients in
which direct acting antiviral-based
therapy failed. International Journal of
Molecular Sciences. 2017;18
[31] Mawatari S, Oda K, Tabu K,

Ijuin S, Kumagai K. New resistance-
associated substitutions and failure
of dual oral therapy with daclatasvir
and asunaprevir. Journal of
Gastroenterology. 2017;52:855-867
87
Genomic variations and phenotypic diversity are closely linked and form the
underlying mechanism for development of many human diseases. This book addresses
the methods of detection, analysis, and interpretation of genomic variations in
clinically relevant scenarios. If your research or clinical practice involves handling
of genomic sequencing data, this book is for you. Topics covered include: methods
for identifying genetic diversity, the workflow for analyzing whole exome and whole
genome sequencing data, local ancestry deconvolution models, the value of molecular
patterns and pattern biomarkers in cancer diagnosis and prognosis, and genotyping
and profiling resistance-associated variants of hepatitis C. If your research or clinical
practice involves handling of genomic sequencing data, this book is for you.
ISBN
ISBN 978-1-83881-844-9
978-1-78923-799-3
Published in London, UK
© 2019 IntechOpen
© Bilge Yurtsever / iStock

Bio in For Matic Stools

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

Bio in For Matic Stools

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Bio in For Matic Stools

Uploaded by

Copyright:

Available Formats

Bioinformatics Tools for

Detection and Clinical

Published in London, United Kingdom

© The Editor(s) and the Author(s) 2019

First published in London, United Kingdom, 2019 by IntechOpen

British Library Cataloguing-in-Publication Data

Additional hard and PDF copies can be obtained from [email protected]

Bioinformatics Tools for Detection and Clinical Interpretation of Genomic Variations

4,200+ 116,000+ 125M+

Our authors are among the

Selection of our books indexed in the Book Citation Index

Interested in publishing with us?

Ali Samadikuchaksaraei is a Professor of Medical Biotechnology,

Dr. Morteza Seifi completed his PhD and postdoctoral training

Bioinformatics Tools for Detection and Clinical Interpretation of Genomic Variations is

Chapter 1 discusses the application of non-decimated wavelet transform (NDWT)

Ali Samadikuchaksaraei, MD, PhD

Morteza Seifi, PhD

The Bioinformatics Tools for

The genome era allowed us to evaluate different aspects on genetic varia-

Keywords: wavelet, genome, NDWT, elastic net, Hurst exponent

The genome era allowed us to evaluate different aspects on genetic variation,

obtaining different characteristics in each level. The analysis using technique of

Wavelet analysis, nowadays, is used widely in subjects such as signal processing,

Non-decimal wavelet transform (NDWT) possesses ample spectra of applica-

which is a behavior function of two real parameters, a and b. If we define ψ a,b(t) as

An important parameter in genetics is the GC content. They are referred as

The Daubechies wavelets, established on the study done by Ingrid Daubechies,

The employment of Daubechies wavelets though software such as Mathematica

7. Non-decimal wavelet transform

Non-decimal wavelet transform (NDWT) has the benefits of period invariance

Spectrogram is an ample prevalent tool in signal analysis because it provides

is broadly known like scalogram [25]. Comprehend a distribution of energy in

Cluster analysis also known as unsupervised classification is a grouping of items

objects, etc. In general, whenever a “mountain” of unknown data is required to be

10. Elastic net

In statistics and specifically in the suitable of linear or logistic regression models,

11. Hurst exponent

Hurst exponent is applied as a degree of long-standing memory of time series.

The authors do not have conflict of interests.

Leila Maria Ferreira1, Thelma Sáfadi1, Tesfahun Alemu Setotaw2 and

1 Universidade Federal de Lavras, Lavras, Minas Gerais, Brazil

2 Ethiopian Institute of Agricultural Research, Addis Ababa, Ethiopia

3 Empresa Brasileira de Pesquisa Agropecuária, Bagé, Rio Grande do Sul, Brazil

*Address all correspondence to: [email protected]

[3] Dodin G, Vandergheynst P, Levoir [12] Karimi K, Wuitchik D, Oldach M,

[8] Vannucci M, Liò P. Non-decimated [16] Kang M. Non-decimated wavelet

[18] Shensa MJ. The discrete wavelet and understanding continuous

[26] Grossmann A, Kronland- [35] Ogutu JO, Schulz-Streeck T, Piepho

linear regression models: Ridge [44] Simonsen I, Hansen A, Nes

[37] Mandelbrot BB, Wallis JR. Noah,

[38] Mandelbrot BB, Wallis JR.

[39] Hurst HE. Long-term storage

[40] Geweke J, Porter-Hudak S. The

[41] Beran J, Terrin N. Estimation of

[42] Robinson PM. Gaussian

[43] Riedi RH. Multifractal processes.