Bioinformatics: Applications

1
Bioinformatics Applications

• ZOO 4903
• Fall 2006, MW 1030-1145
• Sutton Hall, Room 312
• The Study of Gene Expression

2
Lecture overview

• What weve talked about so far
• Finding genes within genomes
• The different forms genes can take (alternative
  splicing)
• Overview
• Measuring levels of transcription
• Microarray technology

3
Why Measure Gene Expression?

• Gene expression levels correspond to protein
  levels
• More abundant genes/transcripts are more
  important
• Normal cells have a standard expression
  profile/signature
• Changes to expression profiles indicate something
  is happening
• Gene expression is a proxy measure for what
  sort of environment or control the cell is under.

4
Problems and potentials for high-throughput
analysis
Less
Easy
DNA
RNA
Biological relevance
Ease of measurement
protein
metabolite
More
Hard
phenotype
5
Transcription is the most commonly reported form
of regulation
6
mRNA level Protein level?

• There is a correlation, but
• Gygi et al. (1999) Mol. Cell. Biol. compared
  protein levels (MS, 2D gels) and RNA levels
  (SAGE) for 156 genes in yeast
• In some genes, mRNA levels were essentially
  unchanged, but protein levels varied by up to 20X
• In other genes, protein levels were essentially
  unchanged, but mRNA levels varied by up to 30X
• Highly expressed mRNAs correlate well with
  protein levels

7
mRNA level Protein level?
R 0.35
R 0.95
Gygi et al. (1999) Mol. Cell. Biol
8
Measuring Gene Expression

• Northern/Southern Blotting
• Serial Analysis of Gene Expression (SAGE)
• RT-PCR (real-time PCR)
• DNA Microarrays or Gene Chips
• Others
• Differential display
• Ribonuclease protection assays
• In-situ hybridization

9
Northern Blots

• Method of measuring RNA abundance
• Name makes fun of Southern blots (which measure
  DNA abundance)
• mRNA is first separated on an agarose gel, then
  transferred to a nitrocellulose filter, then
  denatured and finally hybridized with 32P
  labelled complementary DNA
• Intensity of band indicates abundance

10
Northern Blotting
11
The Blot Block
12
Northern Blots
13
Advantages of Northerns

• Inexpensive, quantitative method of measuring
  transcript abundance
• Well tested and understood technology
• Use of radioactive probes makes it very sensitive
  with high dynamic range

14
Disadvantages of Northerns

• Relies on radioactive labeling dirty
  technology
• Old fashioned and low-throughput technology,
  now largely replaced by microarrays and other
  technologies

15
Serial Analysis of Gene Expression (SAGE)

• Convert mRNA to cDNA
• Split each sequence into short (12-15 base),
  unique tags with a restriction enzyme (NlaIII)
• After creating the tags, these are concatenated
  into a longer sequence or, essentially, a list
  of shorter tags
• Each tag is separated by a SAGE tag
• The list can be read using a DNA sequencer and
  rapidly compared against known sequence databases
  to estimate the frequency of genes

16
SAGE Tools
17
SAGE of Yeast Chromosome
18
Advantages of SAGE

• Very direct and quantitative method of measuring
  transcript abundance
• Open-ended technology
• Sensitive high dynamic range
• Built-in quality control
• e.g. spacing of tags 4-cutter restriction sites
  

19
Disadvantages of SAGE

• Expensive, time consuming technology - must
  sequence gt50,000 tags per sample
• Most useful with fully sequenced genomes
  (otherwise difficult to associate 15 bp tags with
  their genes)
• 3 ends of some genes can be very polymorphic and
  throw off the uniqueness assumption

20
Real Time PCR
21
Principles of PCR
Polymerase Chain Reaction
22
RT-PCR

• RT-PCR is a method to quantify mRNA and cDNA in
  real time
• Generates quantitative fluorescence data at
  earliest phases of PCR cycle when replication
  fidelity is highest
• Measures the build up of fluorescence with each
  PCR cycle

23
RT-PCR
An oligo probe with 2 flurophores is used (a
quencher reporter)
24
RT-PCR vs. Microarray
25
Advantages of RT-PCR

• Sensitive assay, highly quantitative, highly
  reproducible
• Considered gold standard for mRNA quantitation
• Can detect as few as 5 molecules
• Excellent dynamic range, linear over several
  orders of magnitude

26
Disadvantages of RT-PCR

• Expensive (instruments are gt150K, materials are
  also expensive)
• Not a high throughput system (10s to 100s of
  genes not 1000s)
• Can pick up RNA carryover or contaminating RNA
  leading to false positives

27
Microarrays
28
Microarrays

• Basic idea
• Reverse Northern blot on a huge scale
• The clever tricks
• Miniaturize the technique, so that many assay can
  be carried out in parallel
• Hybridize control and experimental samples
  simultaneously use distinct fluorescent dyes to
  distinguish them

29
First, mRNA is made into cDNA clones
30
Robots lay down cDNA probes on the microarrays
31
Target genes are labeled with Cy3 and Cy5 Dyes
Cy3
Cy5
32
Samples are labeled and hybridized to the array
33
Hybridized arrays are put in a scanner

• GenePix 4000

34
Microarrays Spot Color
35
Typical cDNA microarray
36
Spots are then gridded with software and
normalized vs. background
37
Microarray technical problems
Anti-probe Locally high background
Spot overlap
Precipitate Locally low signal
Comet-tails (donut hole)
38
Two Types of glass-slide microarrays

• Spotted glass slide cDNA (500-1000 bp) arrays
• Photolithographically prepared short
  oligonucleotide (25-70 bp) arrays

39
Affymetrix GeneChips
40
Maskless photolitography
41
Affymetrix SNP chips (resequencing arrays)

• Each probe 25 bp long
• 22-40 probes per gene
• Perfect Match (PM) as well as MisMatch (MM) probes

42
(No Transcript)
43
Microarray mania

• Expression arrays
• Exon arrays
• Genomic arrays!
• Methylation arrays
• Protein arrays
• Antibody arrays
• Tissue arrays

44
Exon microarrays
45
Comparative Genomic Hybridization Microarray
Reference DNA Labeled with Cy5 (Detected with Red
Laser)
Patient DNA Labeled with Cy3 (Detected with Green
Laser)
Cot-1 DNA Unlabeled


Glass slide spotted with DNA from known locations
in the Genome
46
CGH Microarrays

• Reveals DNA copy number and allele-specific
  information
• Shows chromosomal deletions and duplications

47
Methylation-Specific Oligonucleotide Microarrays
Reference DNA
Test DNA
mCG
mCG
CG
CG
Bisulfite treatment and PCR
CG
CG
TG
TG
3 end-labeling with Cy3 or Cy5 and co-hybridized
on the chip
48
Methylation-specific hybridization
0 Methylation
100 Methylation
--C--C--
--T--T--
--C--C--
--T--T--
--T--T--
--C--C--
--C--C--
--T--T--
49
Protein microarrays

• True protein microarrays are evolving very slow
  and only a few exist.
• Technology is not straight forward due to
  inherent characteristic of proteins (e.g.
  available ligands, folding, drying)
• Mostly limited to binding interactions (e.g.,
  antibodies, protein-protein)
• Some detect protein-protein interaction by
  surface plasmon resonance other use a
  fluorescence based approach

50
Protein microarrays
51
Antibody microarrays
Antibody or ligand is on the microarray, proteins
are tagged with different dyes
52
Chromatin Immunoprecipitation
53
ChIP-chip
Chromatin immunoprecipitation on DNA chip
54
Tissue microarrays
Tissue samples are spotted for histological or
immunochemical staining, in-situ hybridization,
or just visualization of morphology
55
Advantages to Microarrays

• High throughput, quantitative method of measuring
  transcript abundance
• Avoids radioactivity (fluorescence)
• Kit systems and commercial suppliers make
  microarrays very easy to use
• Uses many high-tech techniques and devices
  cutting edge
• Good dynamic range

56
Disadvantages to Microarrays

• Relatively expensive (gt1000 per array for Affy
  chips, 300 per array for home made systems)
• Quality and quality-control is highly variable
• Quantity of data often overwhelms most users
• Analysis and interpretation is difficult
• Not as sensitive as other methods low abundance
  transcripts are harder to reliably detect

57
Nature prefers a low-energy solution to the
transcriptional response
58
Nature prefers a low-energy solution to the
transcriptional response
59
Microarrays provide quantity at the cost of some
quality
60
Summary

• There are a number of methods to measure
  transcriptional levels, but most are being
  replaced by microarray technology
• What is lost in sensitivity is made up for by a
  more global survey and the ability to do QC via
  replicates
• Miniaturization and automation are creating a new
  high-throughput biology where data analysis and
  management is the biggest challenge
• What all these microarrays have in common is the
  gathering of a massive amount of comparative data.

61
For next time

• Homework 4 due
• Read Mount chapter 13, pages 628-58

62
Different Kinds of Omes
Genome Transcriptome Proteome
63
-Omics Mania
biome, cellomics, chronomics, clinomics,
complexome, crystallomics, cytomics,
cytoskeleton, degradomics, diagnomicsTM,
enzymome, epigenome, expressome, fluxome,
foldome, secretome, functome, functomics,
genomics, glycomics, immunome, transcriptomics,
integromics, interactome, kinome, ligandomics,
lipoproteomics, localizome, phenomics,
metabolome, pharmacometabonomics, methylome,
microbiome, morphome, neurogenomics, nucleome,
secretome, oncogenomics, operome,
transcriptomics, ORFeome, parasitome, pathome,
peptidome, pharmacogenome, pharmacomethylomics,
phenomics, phylome, physiogenomics, postgenomics,
predictome, promoterome, proteomics,
pseudogenome, secretome, regulome, resistome,
ribonome, ribonomics, riboproteomics,
saccharomics, secretome, somatonome, systeome,
toxicomics, transcriptome, translatome,
secretome, unknome, vaccinome, variomics...