Introduction to Bioinformatics

Microarrays1: Microarray Technology

Course 341
Department of Computing
Imperial College, London
Moustafa Ghanem

Aims for the 2nd part of Course

Microarray Bioinformatics

Appreciate the bigger picture of bioinformatics

Bioinformatics is more than nucleotide sequence analysis

Functional Genomics and Drug Discovery

Understand basic microarray technology and its use in gene

expression analysis.

Learn basic data analysis methods and how to apply them in the
analysis of gene expression data

Data Clustering
Data Classification
Statistical Analysis

Recommended Texts

For this part of the course

General overview of microarray data analysis

Lecture Notes
Handouts

Microarray Gene Expression Data Analysis: A Beginners

Guide (Causton, Quakenbush and Brazma)
Microarray Bioinformatics (Stekel)

Data Mining

Data Mining: Concepts and Techniques (Han)

Microarray Technology
Lecture Overview

Aims, Motivation and Overview of 2nd Part of Course

Biology Background
Basic Idea of Microarrays
Types of Microarray technologies and how they work
Outputs of Microarrays
Image Analysis required to transform output to gene
expression matrices
Generating Gene Expression Matrices

Background
Functional Genomics

Functional Genomics:

Systematic analysis of gene activity in healthy and diseased tissues.

The study of obtaining an overall picture of genome functions, including the
expression profiles at the mRNA level and the protein level.

Functional Genome Analysis:

used to understand the functions of genes and proteins in an organism. This is

typically known as genome annotation.
used in integrative biology and systems biology studies aiming to understand
health and disease states (e.g. cancer, obesity, etc)
Used as an important step in the search for new target molecules in the drug
discovery process.

Background
The Drug Discovery Pipeline

Drug Discovery is a lengthy process that takes years and requires the use
of bioinformatics, chemoinformatics and clinical-informatics tools.

Target
Identification

Target
Validation

Lead
Identification

Lead
Optimization

Preclinical
Trials

clinical
Trials

Functional genomics plays an important role in speeding up the pipeline

and also in allowing us to try new therapeutic methods.

Background
Drug Discovery

Functional genomics plays an important role in identifying functions of

potential therapeutic targets such as encoded proteins. Gene expression
studies plays an important role in most stages:

Target Identification:

Target Validation:

Understand the role of a target and the effects of manipulating a target

candidate (e.g. what if I knock a gene out)

Compound Screening:

Understand disease states, identify genetics changes that cause disease

(genes, proteins, tissues, environmental conditions, etc)

Understand compounds effect on target and its risk profile

Pre-clinical and clinical trials:

Prioritise studies

Cell

Nucleus
Chromosome

Background

Biology, Cells and DNA

Protein

Gene (mRNA),
single strand

Gene (DNA)

All living organisms consist of cells. Humans have trillions of

cells; Yeast - one cell.

Cells are of many different types (blood, skin, nerve), but all
arose from a single cell (the fertilized egg)

Each cell contains a complete copy of the genome (the program

for making the organism), encoded in DNA.

A gene is a segment of DNA that specifies how to make a

protein. Human DNA has about 30-35,000 genes; Rice has
about 50-60,000, but shorter genes.

DNA sequence
(split into genes)

codes for

Amino Acid
Sequence

What is?

folds into

Protein

has

3D
Structure
dictates

Protein
Function

determines

Cell
Activity

Gene Expression:

The process by which the information encoded in a gene is converted into an

observable phenotype (most commonly production of a protein).
The degree to which a gene is active in a certain tissue of the body, measured
by the amount of mRNA in the tissue.

Microarrays:

Tools used to measure the presence and abundance of gene expression in

tissue.
microarray technologies provide a powerful tool by which the expression
patterns of thousands of genes can be monitored simultaneously

Background
Gene Expression

Cells are different because of differential gene expression.

About 40% of human genes are expressed at one time.

Gene is expressed by transcribing DNA into single-stranded

mRNA

mRNA is later translated into a protein

Microarrays measure the level of mRNA expression

A Dynamic View

Gene expression depends on environment!

Interactions

Environment
Metabolites

DNA

Growth rate

RNA

Protein

Expression

A Dynamic View

Gene expression varies with time !

forwards-propagated
correlations

metabolites
protein
mRNA
time
event

Microarray Technology
Quantitative Measurement of Gene Expression

Also known as DNA microarrays, DNA arrays, DNA chips, gene

chips, Whatever the name, their use is effectively transforming
a living from a black box into a transparent box.

Applications of Microarray
Technology

10

9
0.8

8
7

0.6

6
5

0.4

4
0.2

3
2

1
0

-0.2
0

100

200

300

400

500

600

700

Data Analysis over microarray data

What type of data analysis is required to:

Identify Genes expressed in different cell types (e.g. Liver vs finger)

Learn how expression levels change in different developmental
stages (embryo vs. adult)
Learn how expression levels change in different developmental
stages (cancerous vs non-cancerous)
Learn how groups of genes inter-relate (gene-gene interactions)
Identify cellular processes that genes participate in (structure,
repair, metabolism, replication, etc)

Applications covered only as example contexts, emphasis is on

analysis methods

Affymetrix Inc. is the leading

provider of Microarray

Microarrays
Basic Idea

technology (GeneChip )
http://www.affymetrix.com/

A Microarray is a device that detects the presence and abundance

of labelled nucleic acids in a biological sample.

In the majority of experiments, the labelled nucleic acids are derived

from the mRNA of a sample or tissue.

The Microarray consists of a solid surface onto which known DNA

molecules have been chemically bonded at special locations.

Each array location is typically known as a probe and contains many

replicates of the same molecule.
The molecules in each array location are carefully chosen so as to
hybridise only with mRNA molecules corresponding to a single gene.

Several companies sell equipment to make DNA chips, including

spotters to deposit the DNA on the surface and scanners to detect
the fluorescent or radioactive signals.

Basic Idea

A Microarray works by exploiting the ability of a given mRNA

molecule to bind specifically to, or hybridize to, the DNA template
from which it originated.

By using an array containing many DNA samples, scientists can

determine, in a single experiment, the expression levels of
hundreds or thousands of genes within a cell by measuring the
amount of mRNA bound to each site on the array.

With the aid of a computer, the amount of mRNA bound to the

spots on the Microarray is precisely measured, generating a
profile of gene expression in the cell.

Background

DNA/RNA Hybridization

DNA molecules:

DNA molecules are long doublestranded chains; 4 types of bases are

attached to the backbone: adenine (A),
guanine (G), cytosine (C), and thymine
(T). A pairs with T, C with G.

DNA-RNA hybridization:
When a mixture of DNA and RNA
is heated to denaturation
temperatures to form single
strands and then cooled, RNA can
hybridize (form a double helix) with
DNA that has a complementary
nucleotide sequence.

The Array

The technology for making DNA chips has become so well-defined

that it is even possible to construct all of the equipment for under
$50,000 using directions on the Internet from Professor Pat
Browns laboratory at Stanford. http://cmgm.stanford.edu/pbrown/

Applying a Labelled
Sample

The molecules in the target biological sample are labelled using a

fluorescent dye before sample is applied to array

If a gene is expressed in the sample, the corresponding mRNA hybridises

with the molecules on a given probe (array location).
If a gene is not expressed, no hybridisation occurs on the corresponding
probe.

Reading the array output

After the sample is applied, a laser light source is applied to the array.
The fluorescent label enables the detection of which probes have hybridised
(presence) via the light emitted from the probe.
If gene is highly expressed, more mRNA exists and thus more mRNA
hybridises to the probe molecules (abundance) via the intensity of the light
emitted.

Chemistry Basics:
Surface Chemistry is used to attach the probe molecules
to the glass substrate.

The Process

Chemical reactions are used to attach the florescent

dyes to the target molecules
Probe and Target hybridise to form a double helix

Labelled targets
in solution
Heteroduplexes
Probes on array
Hybridisation

The array

Steps of a Microarray Experiment

1.

Prepare DNA chip(s) by choosing probes and attaching them to

glass substrate. Note location and properties of each probe.

2.

Generate a hybridization solution containing a mixture of

fluorescently labelled targets.

3.

Incubate hybridization mixture.

4.

Detect probe hybridization using laser technology

a)
b)
c)
d)
e)

5.

Scan the arrays and store output as images

Quantify each spot
Subtract background
Normalize
Export a table of fluorescent intensities for each gene in the array

Analyze data using computational methods.

Types of Microarrays

How are Microarrays are made?

What molecules make the probes?

How are the probes added to the chip?

Spotting vs. In-situ synthesis

Output type

cDNA (PCR products) vs Oligos

Single label vs. Dual label

Why ? Appreciation of some of the concepts of the technology.

Helps us understand and choose between available technology.

Helps us design our experiments.
Helps understand sources of errors in array outputs and compensate
for them.

Each probe represents the measurement for a single gene

An array represents measurements for many genes

Designing the Probes

The probes need to be of high specificity to avoid hybridization with

wrong target molecules.

The probes need to generate an output that is easy to read (spots lie in
defined positions and be of regular size and shape and even spacing).

The probes have to have high sensitivity to detect the mRNA and the
intensity of the spot light must be differentiable from background noise.

The intensity of a spot light also needs to correlate with the abundance
of the target molecule in the sample.

Results must be reproducible across multiple experiments.

Different chip manufacturers use different technologies

Probe Types

As an end user you will use the probe types

recommended for the chips, but would have to select
the sequences for the probes to be used in your
experiments
Affymetrix technology is based on oligos (20 bases per
probe)

The DNA probes used on a an array can either be polymerase chain

reaction (PCR) products (cDNAs) or Oligonucleotides.

In the first case (cDNA), highly parallel PCR is used to amplify DNA
from a clone library, and the amplified DNA is purified, the clones are
typically long sequences (Complete genes or ESTs).

In the second case, DNA oligonucleotides are presynthesised for use on

the array --- An oligonucleotide, or oligo as it is commonly called, is a
short fragment of a single-stranded DNA that is typically 5 to 50
nucleotides long. This can achieve a higher density of probes per chip.

In both cases the probes are attached (fixed or immobilized) to a

glass (or nylon) surface using special surface chemical techniques
(Beyond this course).

Spotting vs. In-situ Synthesis

Spotting

Spotting works for both cDNA probes and oligo probes

The Spotting Process

1.
2.
3.

The DNA probes are produced and stored in wells.

A Spotting robot is used to deposit them onto individual
locations on the glass slide
The glass slide is post-processed so no further DNA can attach
to it.

Spotting is easy to automate but may generate poor quality

spots (irregular spots of different shapes and sizes)

The Spotting Robot

The Operation of the Spotting Robot

1.
2.

3.

4.
5.

The pins are dipped into the wells to collect the

first batch of DNA.
This DNA is spotted onto a number of different
arrays, depending on the number of arrays
being made and the amount of liquid the pins
can hold.
The pins are washed to remove any residual
solution and ensure no contamination of the
next sample.
The pins are dipped into the next set of wells.
Return to step 2 and repeat until the array is
complete.

Spotting Process

Affymetrix technology is based on in-situ synthesis in a

series of addition steps separated by mask addition and
then photo-deprotection.

Spotting vs. In-situ Synthesis

In-situ Synthesis

Since oligos are synthesized short sequences,

their bases can be added to the glass surface
one at a time.

Using high tech processes this can generate

best quality (regular even spots).

Different patented technologies are used to

enable this to happen while not allowing more
than one base to be added at a time, including

Photodeprotection technology (Affymetrix)

Inkjet Array Synthesis

In-situ Synthesis
Affymetrix

Many other variations of the technology exist, such as

the use of longer oligos, the use of fibre optics, etc.

Comparison of Probe Types

In-situ Synthesis / Oligos

PCR Products / cDNA Probes

Advantages

Advantages

No need to isolate and purify cDNAs

because oligonucleotides can be
synthesized.
Short oligonucleotides are less likely to have
cross-reactivity with other sequences in the
target DNA.
Density of chips is higher than with cDNAs.

Flexibility to study cDNAs from any source.

cDNAs do not require any a priori information
about the corresponding genes.
Longer sequences increase hybridization
specificity, which reduces false positives.

Limitations

Limitations

The sequence has to be known.

Synthesis can be expensive and timeconsuming.
The short sequences are not as specific for
target DNA, so appropriate controls must be
added.

Isolation of individual cDNAs to immobilize

on each spot can be cumbersome.
Density is lower than synthesizing
oligonucleotides on the surface of the chip.
cDNAs are longer sequences and are more
likely to randomly contain sequences found
in target DNA, which results in crossreactivity.

Affymetrix technology is based on the use of single

labels

Single Label vs. Dual Label

Single Channel vs Dual Channel

Most laboratories use fluorescent labelling, with the two dyes Cy3 (excited by a
green laser) and Cy5 (excited by a red laser).

In Dual label experiments, two samples are hybridised to the arrays, one
labelled with each dye; this allows the simultaneous measurement of two
samples (e.g. for differential analysis)

In Single label experiments, only one sample is hybridised to the arrays labelled
with one dye. (in which case control needs to be measured using a separate
chip).

Choice between single and dual label is governed by array technology and
underlying chemistry.

Dual Label Experiments

+ Red label

+ Green label

RNA sample 2

RNA sample 1
e
Slid

Typically used in custom made cDNA chips

Typically used to study one sample (e.g. diseased tissue) vs. a
control sample (e.g. normal tissue)
Separate images are obtained for each channel, and then combined

Qualitative Interpretation of Double

Label Experiments

GREEN represents High Control hybridization

RED represents High Sample hybridization
YELLOW represents a combination of Control and Sample
where both hybridized equally.
BLACK represents areas where neither the Control nor Sample
hybridized.

Main issue is to quantify the results:

How green is green?

What is the ratio of the signal to background noise?
How to compare multiple experiments using different chips?
How to quantify cross hybridization (if any)?

Affymetrix GeneChip

Example of Single Label Chips

Hundreds of thousands of oligonucleotide probes packed at extremely high

densities. The probes designed to maximize sensitivity, specificity, and
reproducibility, allowing consistent discrimination between specific and
background signals, and between closely related target sequences.

RNA labeled and scanned in a single color one sample per chip

Interpreting Affymetrix Output

Perfect Match/Mismatch Strategy

GeneChips use a Perfect Match/Mismatch probe strategy

Each probe designed to be perfectly complementary to a target sequence,

a partner probe is generated that is identical except for a single base
mismatch in its centre.

These probe pairs, called the Perfect Match probe (PM) and the Mismatch
probe (MM), allow the quantitation and subtraction of signals caused by
non-specific cross-hybridization.

The difference in hybridization signals between the partners, as well as

their intensity ratios, serve as indicators of specific target abundance.

PM to maximizehybridization
MM toascertainthedegreeofcrosshybridization

Affymetrix GeneChips

Perfect Matches and Mismatches

Other Image Processing Problems

Spot Quality Problems

Various Image processing techniques may be applied to read and interpret the
outputs of Microarrays
Commercial Microarray (e.g. Affymetrix) systems use proprietary software
Image Analysis software packages exist for the analysis of the output of custom made
chips (e.g. GenePix Pro, Array Vision, TIGR Spot Finder, etc)
Typical Problems of Raw Output
Uneven grid positions
Curves within a grid
Variable Spot size or shape
Variable Distance between spots

From Microarray images to

Gene Expression Matrices
Final data
Gene Expression Matrix

Intermediate data

Array scans

Images

Samples

Spots

Genes

Raw data

Spot/Image
quantiations

Gene
expression
levels

From Microarray images to

Gene Expression Matrices

In spot quantitation matrices, rows typically represent all the measurements made from
individual spots on the array. These can include mean and median pixel intensities of the spot
and local background, etc.

An experiment typically consists of one or more spot quantitation matrices representing all
arrays used in the study.

In the gene expression matrix, rows represent genes (as opposed to features/spots on the array)
and columns represent measurements from different experimental conditions measured on
individual arrays.

An example is each column representing measurements at different time points (to, t1, t2, ) in time
course experiments
A second example is each column representing different tissue type
A third is each column representing a different individual
A fourth is having groups of columns representing measurements from diseased cells, and other groups
representing measurements from health cells,
etc,

Each of the above matrices requires the application of data normalisation technuiques as
discussed in the next lecture.

Summary
Microarrays

Basic Concept

Different Microarray technologies exist.

Based on Crick-Watson Hybridization

Probe type (cDNA vs oligo)

Spotting vs in-situ synthesis
Single vs. dual channel

Output is a typically an image

Sources of errors
Image processing is required
Images are converted into gene expression matrices for further analysis