Book reviews

budding biologists. Here we have a book reconstruction); chapters 6–9 provide

that makes it easy. mathematical background for the
statistical methods used in the first part. It
Heikki Lehväsaiho is based upon courses the author has
E-mail: [email protected] presented to undergraduates at Australian
The author is a Staff Scientist at the European National University.
Bioinformatics Institute and a core developer of This is a hugely ambitious undertaking;
Bioperl a young man reaching for the sky. Entire
volumes have been devoted to each of
these topics, which remain intensely active
areas of research. Professor Isaev has done a
An Introduction to great service to his students by introducing
Mathematical Methods

Downloaded from https://academic.oup.com/bib/article/5/4/393/265912 by guest on 09 September 2022

them to these fields while they are still
in Bioinformatics being explored, while there are still many
Alexander Isaev bright nuggets to be unearthed. The book
Springer Verlag, New York; ISBN is presented within Springer Verlag’s
3 540 21973 0; 294 pp.; A49.95; Universitext series of mathematics, ‘not
2004 necessarily textbooks for classroom use,
concepts that are still being thought about,
When the draft of the human genetic presenting material in a way that
sequence was published in 2001, articles mathematicians can learn from.’ On those
in the popular press excitedly proclaimed terms, it is a great success.
discovery of the ‘Book of Life’. The I found the presentation uneven. Some
authors of one of the academic articles topics (eg parameter estimation for
announcing this feat were more sober, HMMs) were explained at great length,
writing ‘in principle, the string of genetic while others (eg the first serious example,
bits holds long-sought secrets of human of sequence alignment) were flung down
development, physiology, and medicine. in a tangle. On the whole, the text seems
In practice, our ability to transform such to have been rushed into print without
information into understanding remains revision after contact with students. It
woefully inadequate’ [Nature, Vol. 409 would require a great deal of explanation,
(2001), pp. 860–921]. I believe, for all but the most serious of
Transforming data into understanding students, and a great deal of patience for
is the primary goal of informatics, indeed all but the most enthusiastic of instructors.
of all science; informatics is motivated by It is nonetheless a valuable addition to the
the modern requirement that we ‘drink literature, for I know of no text, even at
from the firehose’. The automated the graduate level, which successfully
sequencing machines, the immensity of a achieves so ambitious a mission.
genome, the abundance of species call for
new methods of statistical inference as we David Hart
seek to read that Book. The drive to bring
this information to our students, to young
scientists early in their careers, has inspired
Professor Isaev to write this book, which
Bioinformatics: Sequence and
surveys mathematical methods currently
used to understand the working of the
Genome Analysis, second edn
genome. Edited by David Mount, University of
This book consists of 9 chapters, Arizona, Tucson
divided into two parts. Chapters 1–5 are CSHL Press, Woodbury, NY;
devoted to sequence analysis (alignment, 0 87969 712 1; 600 pp (approx.),
Markov chains and hidden Markov illus., appendices, index; $89;
models, protein folding and phylogenetic August 2004

& HENRY STEWART PUBLICATIONS 1467-5463. B R I E F I N G S I N B I O I N F O R M A T I C S . VOL 5. NO 4. 391–396. DECEMBER 2004 393
Book reviews

This book is a very good introduction to an introduction for the book chapters.
sequence analysis and well suited for Chapter 2 is devoted to the collection and
courses at the master’s level. The first storage of biological sequences, with
edition was aimed mainly at biologists, explanations of laboratory sequencing
although the author recognised its procedures and a very complete review of
usefulness for people in the computational the different formats used to store
sciences interested in biological problems. sequences in biological databases.
In this second edition, the editor has gone Chapters 3 to 5 deal with the
a step further, including chapter guides sequence alignment problem: chapter 3
explaining both the computational/ is concerned with pairwise alignments
statistical and the biological background and chapter 5 with multiple sequence
required for each chapter. alignments. Chapter 4 is new to the

Downloaded from https://academic.oup.com/bib/article/5/4/393/265912 by guest on 09 September 2022

This is a timely and laudable approach. second edition after being part of the
Indeed, any scholar involved in one of the original chapter 3, and provides an
many bioinformatics programmes that are introduction to probability and statistical
being developed around the world is analysis of sequence alignments. This is
aware of one of the main hurdles in an important issue that certainly merits
designing the curriculum for such a its own chapter, as misunderstanding the
programme: the heterogeneity of the statistical significance of sequence
student body. The ever-rising field of alignments can lead to defective
bioinformatics attracts people from many functional genomic analyses, among
different disciplines within the life other problems.
sciences and the computational, statistical Chapters 6, 7 and 8 build on the issues
and engineering areas. Bridging the discussed in the preceding chapters to
complementary knowledge gaps these cover similarity searches on sequence
dissimilar backgrounds suffer from when databases, phylogenetic analyses, and
it comes to dealing with the prediction of RNA secondary structure,
bioinformatics discipline is one of the respectively. Chapter 9 deals with gene
main challenges of graduate prediction and promoter analysis.
bioinformatics instruction. In this sense, Chapter 10 is the longest chapter, and
the book’s design facilitates to some covers structural analysis of proteins,
extent the combined instruction of including protein classification through
students with computational and structural alignments and structure
biological backgrounds. prediction. Chapter 11 introduces the
The book consists of 13 chapters. techniques used in genome analysis,
Chapter elements new to the second including the application of sequence
edition are the above-mentioned chapter analysis methods covered in previous
guides for biologists and computational chapters, as well as new techniques and
scientists, lists of topics that should be problems related to the analysis of
learned, sets of problems and web search genomes, such as single nucleotide
terms. The web references included both polymorphisms (SNPs).
in the main text and the web search terms Chapter 11 sums up the preceding
sections are stellar. Indeed, the web could chapters from an application perspective
be viewed as bioinformatics’ main and closes the sequence analysis theme in
resource, and the book provides many the book. Indeed, this chapter was the
pointers to useful sites and resources, and final one in the first edition. For the
makes good use of web-based tools in the second edition, two new chapters,
accompanying exercises. contributed by other authors, were added
The first chapter provides a good to provide a wider coverage of relevant
historical perspective and an overview of areas of bioinformatics. Chapter 12
the bioinformatics field thoughtfully teaches how to use and generate Perl
arranged in such a way as to serve also as scripts and modules in a variety of

394 & HENRY STEWART PUBLICATIONS 1467-5463. B R I E F I N G S I N B I O I N F O R M A T I C S . VOL 5. NO 4. 391–396. DECEMBER 2004
 Book reviews

bioinformatics applications, and also book a useful instrument for the

provides an introduction to data formats classroom. On the other hand, the book’s
and relational databases. Chapter 13 comprehensive coverage of problems,
presents techniques for gene expression tools and techniques, the handy glossaries,
analysis from microarray data, including and the extensive references to available
experimental design, analysis methods web resources and literature, correspond
and programming resources. Both to a very handy reference work.
chapters 12 and 13 are presented at a Despite the book being originally
more advanced level than the rest of the conceived with the biologist in mind,
book and assume some background on the computational scientist also stands to
the reader’s part. benefit greatly from it. Indeed, the
The methods presented – such as book does an excellent job of

Downloaded from https://academic.oup.com/bib/article/5/4/393/265912 by guest on 09 September 2022

hidden Markov models (HMMs), presenting the biology and the relevant
dynamic programming and neural underlying sequence analysis problems
networks – as well as the tools described and their application to genome
– such as BLAST, CLUSTALW, analysis, and provides ample, though
PILEUP and PHD – are too numerous to not very mathematical, explanations of
list here. Suffice to say that the book the algorithms employed by the diverse
includes all the most commonly used tools covered. This approach offers the
techniques and tools, plus several less computational scientist a good base to
well-known ones. The most important start developing tools for bioinformatics
sequence, structure and microarray problems, but the effect on the biologist
databases are also described throughout is more attuned to the idea of
the different chapters. becoming a knowledgeable ‘tool user’,
The book is comprehensive in terms well versed on the uses and limitations
of problems and tools covered, with very of available resources. Both types of
few omissions. For example, no mention student would be able to understand, to
is made of terminators or operons in the a reasonable extent, computational
regulation part of chapter 9, nor of the biology papers dealing with interesting
computational problem of operon problems, but the computational
prediction, but, in general, the most scientists, thanks to their backgrounds,
relevant aspects of sequence analysis are will be better prepared to create and
certainly covered. Some unevenness in build software tools to tackle those
detail is apparent in the descriptions of problems. In this sense, some references
different computational techniques, as, and pointers to available general
for example, HMMs and dynamic statistics and machine learning/data
programming are explained in greater mining software, such as HMM or
detail than neural networks. In all cases, neural network builders, could provide
however, the explanations are such that a better head start into tool building for
the reader will have a reasonable students with little computational
understanding of how the methods background, and make this book even
work. more useful than it already is. As
The book, as its author states, is explained above, there is indeed a
intended both as a reference and a chapter devoted to the Perl scripting
textbook, and it does a very good job on language, but, logically for the scope of
both counts. On the textbook side, clear the book, it is devoted more towards
objectives, glossaries and chapter guides at interfacing with known tools and with
the beginning of each chapter, databases than to the implementation of
informative text boxes, handy references the computational techniques described
to literature and web resources, and throughout.
problem sets involving hands-on In conclusion, the second edition of
experience with common tools make this ‘Bioinformatics: Sequence and Genome

& HENRY STEWART PUBLICATIONS 1467-5463. B R I E F I N G S I N B I O I N F O R M A T I C S . VOL 5. NO 4. 391–396. DECEMBER 2004 395
Book reviews

Analysis’ is an excellent textbook for Pedro Romero

bioinformatics introductory courses for Assistant Professor of Bioinformatics,
both life sciences and computer science School of Informatics,
students, and a good reference for Center for Computational Biology and
current problems in the field and the Bioinformatics,
tools and methods employed in their Indiana University,
solution. Indianapolis, USA

Downloaded from https://academic.oup.com/bib/article/5/4/393/265912 by guest on 09 September 2022

396 & HENRY STEWART PUBLICATIONS 1467-5463. B R I E F I N G S I N B I O I N F O R M A T I C S . VOL 5. NO 4. 391–396. DECEMBER 2004