Should an additional p-value correction be made for running multiple contrasts? Say for example I have 12 samples and with one factor that has 4 levels, A, B, C, and D.
No code has to be inserted here.If I run DESeq using Wald tests to determine the differentially expressed genes and then compare each level to the others using contrasts like this:
Does running multiple contrasts require an additional correction to the adjusted p-value reported in the results object?
No code has to be inserted here.If I run DESeq using Wald tests to determine the differentially expressed genes and then compare each level to the others using contrasts like this:
Code:
dds <- DESeqDataSetFromMatrix(countData = countData, colData = colData, design = ~ condition) dds <- DESeq(dds) results(dds, contrast=c("condition","A","B")) results(dds, contrast=c("condition","A","C")) results(dds, contrast=c("condition","A","D")) results(dds, contrast=c("condition","B","C")) results(dds, contrast=c("condition","B","D")) results(dds, contrast=c("condition","C","D"))
Comment