Introduction to microarray technology and data analysis

Transcription

1 Introduction to microarray technology and data analysis Aron C. Eklund Cancer Systems Biology group Center for Biological Sequence Analysis Technical University of Denmark Introduction to Systems Biology February 17, 2014

2 Tools we can use to study the cellular system Mass Spectrometry Total protein levels and post transla7onal modificia7ons metabolomics Sequencing Genomic and exosome sequence Muta7ons in DNA, RNA Arrays Levels of RNA Gene manipula7on studies Knockdown RNA Overexpress RNA IE: Generate a large amount of data

3 System biology work flow DNA RNA Protein Metabolites Modeling of all informa7on (determine regulators of certain func7on) Computer driven hypothesis of molecular regulators of func7on Experimental manipula7on DNA RNA Protein Metabolites

4 Learning objectives 1. Describe a gene expression microarray and what it measures 2. Explain the steps needed to generate gene expression microarray data 4. Describe at least one application of a gene expression microarray 5. Use R at a basic level 6. Explain and calculate a log2 ratio 7. Explain and calculate a P value using the t test

5 Expression of a gene Transcription Translation DNA mrna Protein gene expression = how much of a certain mrna is present protein expression = how much of a certain protein is present

6 Why do we want to measure gene expression? 1. Biological / medical research - inference of the function of a gene - understanding gene regulation - biomarker discovery 2. Clinical diagnostics - predictive biomarkers for cancer - identifying cancers of unknown primary

7 The DNA microarray Thousands to millions of DNA spots in a few square centimeters Cartridge Glass slide a. b.

8 The gene expression microarray a. Labeled target RNA made from a biological sample Specific hybridization between a complementary probe and target detected by fluorescence DNA probe attached to the microarray substrate expression profile

9 Sequence-specific hybridization 5 -CTTGACACCCAGCTGGTCCTGTGGATGGGAGCGGT-3 3 -GTGGGTCGACCAGGACACCTACCCT-5 Probe Target mrna Substrate 5 -TAGCACACCCCGCTGCTCCTGTGGTCCCAAATGCA-3 X 3 -GTGGGTCGACCAGGACACCTACCCT-5 other mrna

10 Gene expression matrix Samples Genes 100.CEL.gz 103.CEL.gz 104.CEL.gz 105.CEL.gz 106.CEL.gz DDR RFC HSPA PAX GUCA1A UBA THRA PTPN CCL CYP2E EPHB ESRRA CYP2A GAS MMP TRADD FNTB PLD which values can be compared here?

11 The process Biological specimen Expression data

12 The process 1. Isolate RNA from a biological specimen 2. Create labeled target from the RNA 3. Let the labeled target hybridize (incubate) on the microarray 4. Wash off the unbound target 5. Scan the microarray 6. Analyze the data

13 SAMPLE Two color labeled cdna (mostly outdated simple approach!) CONTROL mrna mrna cdna cdna Cy5-cDNA Cy3-cDNA

14 The Eberwine Protocol SAMPLE 42 C 2 h + Reverse Transcriptase RNA ssdna T7 70 C 10 min 16 C 2 h + RNase H + Polymerase clean up dsdna T7 pol 37 C 6 h dsdna arna + Biotin-labeled nucleotides

15 Affymetrix fluidics station and scanner Hybridize, incubate, rinse Scan

16 Image processing and normalization Normalization adjusts data to correct for systematic bias in intensity levels

17 Normalized expression matrix Samples Genes 100.CEL.gz 103.CEL.gz 104.CEL.gz 105.CEL.gz 106.CEL.gz DDR RFC HSPA PAX GUCA1A UBA THRA PTPN CCL CYP2E EPHB ESRRA CYP2A GAS MMP TRADD FNTB PLD now what?

18 What is the question? Typically, we use microarrays to compare between groups: Disease vs. normal Drug treatment vs. control Disease A vs. Disease B Good prognosis vs. bad prognosis

19 Comparing a gene between two groups MBNL fold change log2 ratio P value (Student s t test) HD N

20 Visualization of microarray data 0.00 scanner origin PCC BPE4 BPLE3 BPLE2 BPLER2 BPE3 BPE2 BPLE4 BPLER3 BPLER4 OCE1 FNE1 OCE2 FNE2 OCLER1 FNLER1 OCLE1 FNLE1 OCLE2 OCLER2 FNLE2 FNLER2 HMLE3 HMLER3 HME3 HME2 HME4 HMLER4 HMLE2 HMLE4 HMLER2

21 Individual probe response on microarrays Log scale Natural scale Log intensity Intensity Log concentration (pm) Concentration (pm) D. Skvortsov et al., (2007) NAR

22 Microarray data: what do the numbers mean 1. No calibration: Expression values cannot be readily translated into concentration 2. Response curves are not parallel: Comparison between one gene and another gene is almost meaningless 3. Noisy at low levels (on the log scale): Fold changes can be misleading. 4. Saturation at high levels: Cannot detect changes

23 Evolution of gene expression (mrna) measurements s - Northern blot - one gene at a time, not very accurate s - Quantitative reverse transcriptase polymerase chain reaction (RT-PCR) - one gene at a time, very accurate 3. mid-1990s - DNA microarray - many genes at a time, medium accuracy 4. Late 2000s - RNA-seq - all genes, high accuracy?

24 Other applications of microarrays (in general) DNA microarray - Gene expression - Genotyping (SNP profile) - DNA profiling (CGH, copy number) - ChIP-Chip Protein microarrays Tissue microarrays Cellular microarrays Compound microarrays

25 About the exercises

26 The data set in the exercise Genome-wide expression profiling of human blood reveals biomarkers for Huntington's disease. Borovecki F, Lovrecic L, Zhou J, Jeong H, Then F, Rosas HD, Hersch SM, Hogarth P, Bouzou B, Jensen RV, Krainc D. Proc Natl Acad Sci U S A Aug 2;102(31): samples: - 14 normal - 5 presymptomatic - 12 symptomatic Affymetrix HG-U133A arrays Huntingon s disease: - neurological disorder - polyglutamine expansion in the huntingtin gene Why search for marker of disease progression (not diagnosis)? - assess treatment efficacy - surrogate endpoint in drug trials

27 Practicalities

28 Lunch break now Exercises at 13:00