Question about REDItools v1 and v3 #28
Description
I was wondering how you process the GENCODE GTF file (e.g. v41 human) for use with REDItools AnnotateTable.py. It seems that newer GTF versions omit the transcript_id field, which REDItools (being an older tool) appears to depend on.
When I filter out lines missing transcript_id in the GENCODE GTF file, AnnotateTable.py runs but I'm not sure if the outputs are correct.
In a simulated dataset, I have found that v1 is more sensitive and so I'd like to use v1. REDItools v1 also sets the strand information correctly whereas v3 infers * for the strand. Are there major differences between v1 and v3 in the underlying algorithms?
Thank you for your help!