Hi all,

I need to reanalyse an RNA-seq experiment with 2 conditions and unfortunately no replicates. I probably don't have big changes in expression, 100-200 genes with logFC max 2-3.

I'm planning to do the the following:

1. Salmon/Kallisto with a recent RefSeq collection to estimate read counts.
2. tximport to import gene level counts with an offset that corrects for changes to the average transcript length across samples.
3. DESeq2 with all the warnings read in the manual.
4. Rank genes based on lfc, and compare it to other stuff

Any thoughts on this? Can I improve it in any way?

hi Endre,

I like the pipeline you've set up here. MA plot will be most revealing. The p-values are not of any use, but the moderated LFCs can help with ranking genes for follow-up.