Statistical Analysis of Microarray Data by Hanne Jarmer
Induction of adherence by sub-lethal alcohol concentrations
Microarray experiments on: Induced adherence - Listeria monocytogenes adheres better at sub-lethal alcohol concentrations Microarray experiments on: 1. Wild type +/- alcohol 2. Mutant +/- alcohol 4 biological replicates of each
The microarray data (wild type)
The microarray data (mutant)
Why and what do we test? - We test to extract significant genes Intensity Density Fold change: > 2 Fold change: ~1
The t-test The t statistic is based on the sample mean and variance t
The P-value Definition: The possibility of getting the observed difference by coincidence
Correction for multiple testing Each time we test, there is a certain possibility, that the observed difference is in fact a coincidence when H0 is TRUE Unacceptable many false positives
Correction for multiple testing Bonferroni: P ≤ 0.01 N Confidence level of 99% Benjamini-Hochberg: P ≤ i N 0.01 N = number of genes i = number of accepted genes
Volcano plot P-value log2 fold change (M)
1 The 2 way ANOVA Interaction 2 3 wildtype +alcohol mutant +alcohol
What would be significant? Intensity 1. +/- alcohol wt alcohol both mutant 2. +/- mutation wt alcohol both mutant 3. +/- both wt alcohol both mutant
Acknowledgements Anne Lise Gravesen, KVL Growth experiments Torsten Hain, Institut für Medizinsiche Mikrobiologie Microarray experiments
Correction for multiple testing Bonferroni: P ≤ 0.01 N Confidence level of 99% Benjamini-Hochberg: P ≤ i N 0.01 N = number of genes i = number of accepted genes