DESeq2 and edgeR are complicated programs for identifying differential gene expression from high-throughput sequencing data. This is the first in a long series of videos that explains how these programs work.
DESeq2 and edgeR are complicated programs for identifying differential gene expression from high-throughput sequencing data. This is the first in a long series of videos that explains how these programs work.
Hi Josh, thank you for your clear explanation. I could understand that DESeq2 performs this library normalization method for facilitation of DE analysis between groups of samples. But I have a question here: if DESeq2’s library normalization methodology does not take into account the gene size (or length), then after normalization, can we compare the expression level of, say Gene A vs. Gene B “WITHIN” the same sample? And if not, what normalization method should we use for this purpose? Thanks in advance!
This is a really great question. The bummer is that I don’t have a great answer. That said, I’d be willing to guess that there is an “off the shelf” solution to this – so look around. If you can’t find anything, let me know – I’ve got some ideas that could make it work in DESeq2 by borrowing the normalization factors to calculate TPM and then borrowing the dispersion factors to do a paired sample design.