Hi! I have performed the transcript abundance quantification with RSEM and then I created the gene-level count matrices for use with DESeq2 by importing the quantification data using tximport following the vignette (http://bioconductor.org/packages/release/bioc/vignettes/tximport/inst/doc/tximport.html#Importtranscript-levelestimates):

txi.rsem <- tximport(files, type = "rsem", txIn = FALSE, txOut = FALSE)
txi.rsem$length[txi.rsem$length == 0] <- 1
ddsTxi <- DESeqDataSetFromTximport(txi.rsem, colData = samples, design = ~ condition)

I was wondering if this approach corrects by default for potential changes in gene length across samples (e.g. from differential isoform usage).

Thank you, Concetta

Yes it should.

When you run DESeq() it should even print out a message, along the lines of, "using average transcript length when estimating size factors..."