Hey!

We analyzed some data in primary human macrophages however our upstream response to different cytokines has huge variation in the absolute level of expression of the genes. 

We used DeSeq2 package to calculate the differential gene expression. However we are wondering if we lose on certain genes because of different absolute levels of expression even though the fold changes  between the stimulus and untreated conditions are constant across different biological replicates.

Does DeSeq2 package already normalizes the data to base level and considers the between sample variance? If yes, can anyone please reference me to the paper mentioning the same?

Thanks

Sahil

DESeq2 considers fold changes per gene, the fact that the genes themselves have different absolute level is not a problem. Here is a reference for the statistical model:

https://genomebiology.biomedcentral.com/articles/10.1186/s13059-014-0550-8