Entering edit mode
I have samples from multiple time points. For the results, I get exon usage coefficient for each time point and a fold change for every time point versus 0. That makes perfect sense. However, there is only one pvalue and padj. What exactly do they represent?
Is there a way to extract the statistics for two specific time points without starting from a new sample table? In DESeq2, for example, the results function has a contrast argument specifically for that.
hi,
something related, if you have multiple condition for the same variable, like Treatment: drug1, drug2, drug3 ... to get the p-value for any pairs, i guess you need to construct the design matrix where the column has only two of them for all the pair-wise comparison?
thanks!
Hi Lorena,
If you are interested in pair-wise comparisons, I think the easiest is to subset your DEXSeqDataSet object for each pair-wise comparison and then run the analysis separately for each of the subsets. If you use all your conditions in the same DEXSeqDataSet object, DEXseq will test for differences in exon usage in at least one condition compared to the other conditions, then you could use the effect sizes to explore the differences.
Alejandro