Microarray Technique Some background M. Nath
Microarray Technique
Some background
M. Nath
Outline
Introduction
Spotting Array Technique
GeneChip Technique
Data analysis
Applications
Conclusion
Now
Blind Guess?
Functional Pathway
MicroarrayTechnique
Principle
Comprehensive functional analysis of genome
Simultaneous analysis of patterns of gene expression
Genome > Transcriptome > Proteome
Types of Microarray
cDNA Array (Brown et. al., 1995)
Genomic DNA Array (DeRisi et. al., 1997)
Oligonucleotide Array (Morton et. al., 1998)
Spotted Array Technology
LibrarySpotted Array Technology
An array of slides is printedSlides can be glass or nylon
Printing SlidesSpotted Array Technology
Spotted Array Technology
Hybridisation of Slides
Slide developer
Spotted Array Technology
Up to 48 slides are developed under uniform conditions
ScanningConfocal laser scanner is used
Two different lasers to read Red and Green dye intensities
Imaging software reads Red & Green intensity for each dot applied
A Graphic image is savedLaser Scanner
Spotted Array Technology
Results
Green = Active in Sample 1Red = Active in Sample 2Yellow = Active in both samplesBlack = Active in neither
Spotted Array Technology
GeneChip TechnologyAffymetrix chip
Oligos of 25 nt long
40 oligos for detection of each gene
11-20 oligos as Perfect Match (PM)
11-20 oligos as Mismatch (MM) at position 13
GeneChip Technology
GeneChip Technology
Spotted Array Technology
10000No. of genes / array
1Probes pair per gene
10-20 µg total RNAStarting material
RoutineFeatures
Spotted Array Technology
Laborious
Inexpensive
Moderate specificity
Moderate representation
Low Density
Cannot detect polymorphism
GeneChip Technology
4000012000No. of genes / array
420Probes pair per gene
93% identity
70-80% identity
Discrimination of related genes
1:1061:105Detection specificity
2 ng total RNA
5 µg total RNAStarting material
LimitRoutineFeatures
GeneChip Technology
Easy
Expensive
High specificity
High representation
High density
Can detect polymorphism
Data Analysis
ScalingLinearity
Data Analysis
Gene intensity Chip 1
Gen
e in
tens
ity C
hip
2
Scaling
Linear and non-linear models
Constitutively and constantly expressed Maintenance gene
More genes on chip
Data Analysis
Gene intensity Chip 1G
ene
inte
nsity
Chi
p 2
Outlier
Two chips may differ in expression for same gene
If one replicate deviates several standard deviation from mean, remove it
Data Analysis
Data Analysis
Absolute measurements
AvgDiffΣ ( PMn – MMn ) / N
Weighted AvgDiffΣ ( PMn – MMn ) φn / N
Fold Change
Log2 of ratio of intensities after being corrected for background
E.g. Log2 (Sample / Control) = Log2 (Red / Green)
=1 : unchanged; >1 : upregulated; <1 : downregulated
Affymetrix chip (AffyFold)(Sample - Control) / Min (Sample, Control)
Data Analysis
Significance Test
Test of significancet-test with unequal varianceANOVA and F testREML
Non-parametric testsWilcoxon testMann-Whitney rank sum test
Correction for multiple testingBonferroni correction
Data Analysis
Cluster AnalysisSingle array not suitable
Functional analysisCo-regulationNew gene discovery
Samples collected temporally, spatially …
Multiple array & Cluster analysis
Clustering of similarly behaving genes
Genes with similar functions generally cluster together
Data Analysis
Cluster Analysis
Cluster analysisHierarchical clusteringK-means clusteringSelf Organising MapsDistance measures
Data Analysis
Beyond Clustering
Discovery of regulatory elements in promoter regionIdentifying regulatory networks
Time series approachSteady-state approachNeural network technique
Selection of genesGene findingSelection of regions within the genesSelection of PCR primersSelection of unique oligomer probes
Data Analysis
Software Package
Affymetrix Data Mining ToolAffymetrix NetAffxBiomax Gene Expression Analysis SuiteGeneData ExpressionistInformax XpressionInvitrogen Corp. ResGen PathwaysRosetta Resolver Gene Expression SystemSilicon Genetics GeneSpringSpotfire
Data Analysis
Applications
Analysis of patterns of gene expressionFunctional relationship between genes
Expression in coregulatory gene group
Monitoring changes in genomic DNACellular pathways affected by mutation
Changes in expression profiles of mutants
Applications
Simultaneous detection of many genes
Gene discovery
Pathway analysis
Molecular basis of disease progression
Applications
Molecular signatures of pathogens
Comparative genomic studies of pathogensVirulence difference
Pathogen genetics and manifestationLife cycleReplication, translational control
Applications
Host-parasite interaction
Pathogen establishmentHost cell recognitionHost cell responseParasite response to host immune response
Applications
Constraints
Complex system of eukaryotes & multi-cellular organisms
Transcriptome analysis
Developing technology
Many stages
Design of experiment
Constraints
Array quality
Highly variable data
Analysis of data
Published experiments
Cost
From Here to Tomorrow
Recent & Powerful
More improvementProtocolHardwareExperimental designComputational technique
Integrate with other data
Reproducible, fast, sensitive & economic