Geo Pertea

There is no such limit (and error message) in gffcompare, though it was as you described it in cuffcompare -- there is indeed a line in gtf_tracking.h (if you look...

Apologies, I actually forgot about that github copy of cuffcompare, that was abandoned a while ago and I realize now that it no longer compiles with the current version of...

There is a complication here I can see: when multiple sources do not agree on the CDS, which CDS should be used in the output, for otherwise structurally-compatible transcripts ?...

How long did you wait? It actually finished for me after over 20 hours. But that BAM seems to be problematic. It seems there are multiple samples merged in that...

@ttgump, that is incorrect (I hope!), prepDE.py should work with version 2.0 just the same, I tested it on the original RNA-Seq protocol data (chrX) and it worked as expected....

Then that's a problem we'd surely like to fix. Stringtie v2 did improve _assembly_ accuracy in our tests (to answer @andy3404's question), but there were sections of the code which...

@Gin-Wang did you run the new prepDE.py with the `-v` option ? That should show the name/path of each sample as it is processed. I'd be interested in the BAM...

Thank you for your help with this problem. From that log it looks like the files that I would need should be shown in the first line of the file:...

Thank you so much @Gin-Wang I took a look at the data you uploaded and I can confirm now that the problem is real and correctly reported by users in...

The main author fixed this issue in v2.0.4 which was just released (see https://github.com/gpertea/stringtie/releases/tag/v2.0.4). Now stringtie v2.0.4 should behave as expected with the `-e` option (just like v1.3.x) so the...