Hi,

I am planning to import the transcript abundances from tsv files that were calculated by kallisto using Quantseq 3' mRNA seq data.

But I don't know what is a correct argument for tximport function for Quantseq results, especially for countFromAbundance.

I assume it will be "no", but I don't know well clearly. Could you explain it to me?

Best,

DH

Txi_gene <- tximport(path, 
                     type = "kallisto", 
                     tx2gene = Tx, 
                     txOut = FALSE, #determines whether your data represented at transcript or gene level
                     countsFromAbundance = "no",
                     ignoreTxVersion = TRUE)

Yes, option no is recommended for end-tagged data, see vignette https://bioconductor.org/packages/release/bioc/vignettes/tximport/inst/doc/tximport.html#3%E2%80%99_tagged_RNA-seq and this previous question tximport of 3 tagged RNA-seq counts from Salmon for downstream analysis with DESeq2

You probably don't want to use countsFromAbundance, because the transcript length is probably less relevant than for normal bulk RNA-Seq. But that is just a wild guess. You might consider aligning with STAR or whatever conventional aligner you like, and then look at the count distribution across the transcripts. If it's all just within say the first 100 nt of each transcript, then the transcript length is probably not relevant any longer, and using any transcript length information is likely to introduce bias.