I could/should have asked this question at bioC forum, but the answers there are usually (just) over my head.
I wondered how to use a table of "Expected counts" from rsem to obtain DEGs using DeSeq2/EdgeR? The expected counts are from UCSC Xena- processed GTEX data. There seems to be more than one (proposed) workflows present for this kind of study. Before delving deeper into calculations, I wanted to know which approach is more often used:
- Rounding the expected count: Question about how to transform RSEM expected_count of TCGA TARGET GTEX to integers?
As the tximport vignette at bioC explains https://bioconductor.org/packages/release/bioc/vignettes/tximport/inst/doc/tximport.html#rsem . However, in the vignette, the starting txi object is built differently from my data (I still have not checked the structure of the resulting R object, so that I might create it manually; waiting for the approval of this method over the simpler method 1 above ) .
txi <- tximport(files, type = "rsem", txIn = FALSE, txOut = FALSE) dds <- DESeqDataSetFromTximport(txi, sampleTable, ~condition)