search

williamritchie / IRFinder

Detecting intron retention from RNA-Seq experiments
53 stars 25 forks source link

About the warning message #145

Open Codezy99 opened 3 years ago

Codezy99 commented 3 years ago

Dear Dadi,

Thank you for the cool tool. Recently I met a problem with the IRFinder, when I run the IRFinder agaist my pair-end library. There is a warning message come out all the time. 

WARN: This sample appears to have been prepared with a strand-specific protocol (directional RNA-Seq), however that process was substantially non-successful. A large number of reads appear opposed to the expected direction.

I think our library was sequenced with the True-seq protocol that means the library are mapped to the reverse strand. I was wondering whether this warning message is due to this reason. If so, I was wondering whether we can solve it by any specific parameter.

Thank you and best regards, You Zhou

dg520 commented 3 years ago

@Codezy99 Sorry for a late reply. Did you quantify IR from FASTQ directly? If so, Illumina TruSeq would end up with this message. I would suggest you to align the FASTQ first using STAR and then quantify IR from BAM. In that way, STAR will take care of the directionality of splice junction mapping. If you had already quantified IR from BAM and still seen this warning, something might funky happen. We can investigate more if so. Let me know.