Reducing fold change cut off in cuffdiff

[nitin.mantri]

Can anyone confirm if default fold change in cuffdiff is 2? how can I reduce this to 1.5?

[jeales]

As far as I know there is no fold change cutoff applied by cuffdiff
Nothing here about it
Why do you think there is a cutoff applied?

http://cufflinks.cbcb.umd.edu/manual.html#cuffdiff

[nitin.mantri]

Hi Jeales

thanks for your reply. I thought so as well. however, some genes that were 1.5-1.8 fold expressed are missing in the final list of significantly differentially expressed genes. also, I could only see -1.5 and 1.5 as lowest log2 values for DE.

[jeales]

Which output file are you looking at?
Also how many lines does it contain and is it the correct number? (i.e the number of genes/transcripts in the gtf)
All log 2 fold changes should be reported whether they are significant or not

Also if you're just looking at a list of significant results, then broadly speaking, results with smaller fold changes are less likely to be significant, and this might be why you aren't seeing any fold changes nearer to zero