I was following the protocol for analysis of Time course experiments using DESeq2 (http://www.bioconductor.org/help/workflows/rnaseqGene/#time-course-experiments) and came up with a doubt regarding the criteria used to calculate the fold-changes and padj.
In this example the fission data package was used, which contains RNA-seq data for a time-course of fission yeast from two different strains.
The following code performs a likelihood ratio test. Genes with small p value from this test are those which at one or more time points after time 0 showed a strain-specific effect.
library("fission") data("fission") ddsTC <- DESeqDataSet(fission, ~ strain + minute + strain:minute)
ddsTC <- DESeq(ddsTC, test="LRT", reduced = ~ strain + minute) resTC <- results(ddsTC)
My question is: When you look to the calculated fold-changes, are they the average of the fold-changes for all the time-points, or if a particular gene is affected only in one time-point, DESeq calculates the fold-change for that particular time-point?
Also, regarding the padj, how does DESeq rank the genes? If a gene X has a 1.3 fold-change across all the time-points and another gene Y a 2 fold-change difference in only one or two time-points, which one will be more significant?
Finally, is there a way of knowing in how many and which time-points the differences are significant?