minimum sequence length 54bp?

[bwprice]

Hi Nathan, we are working with museum samples, so lots of very short fragments. I've noticed that exonerate only uses sequences that are 54bp or longer from our data. We are using exonorate via MGE and the MGE author doesn't think the issue is with MGE so wondering if you know how we could use shorter sequences?

Thanks,
Ben

[nathanweeks]

@bwprice Could provide an example exonerate command & input files from the MGE workflow? That would help pinpoint the issue.

[bwprice]

Hi Nathan! Thanks so much for the quick response :)

Input files are an AA sequence in a fasta file and the PE reads in fastq format. MGE looks for a vulgar file and if there isn't one it calls exonorate. The MGE app is here:

https://github.com/cmayer/MitoGeneExtractor/blob/main/MitoGeneExtractor.cpp

[nathanweeks]

It appears MGE supports a debug mode:

https://github.com/cmayer/MitoGeneExtractor/blob/552c04440de36d2036d10412dbaa5e7d84b59c04/MitoGeneExtractor.cpp#L1096-L1100

Could you try enabling this, and provide an example of an exonerate command string that "fails", and attach the query & target files to this issue (assuming they fit within GitHub's current 25 MB maximum file size limit)?