Running on mouse data

[dvantwisk]

Is genion capable of running on mouse data? I've switched out the references, but it doesn't appear to be working for me so I just want to know if the package was only designed for analysis of human data.

[f0t1h]

Hi, genion is species agnostic (However, only tested on human data).

The current version requires segmental duplication annotation (in the genomicSuperDups format. If it is not possible to find it for the mouse genome, an empty file should work.

If the problem is something else please let me know.

[dvantwisk]

I've gotten the program to work without error, however, it doesn't find any output whereas other long-read fusion transcript programs do. Perhaps it is the case that it is not able to find the fusions in the mouse data that I have, but I suspect something may not be going right. I suspect it may have something to do with generating the sequence similarity file. The command:

cat [cdna.selfalign.paf] | cut -f1,6 | sed 's/_/\t/g' | awk 'BEGIN{OFS=\"\\t\";}{print substr($1,1,15),substr($2,1,15),substr($3,1,15),substr($4,1,15);}' | awk '$1!=$3' | sort | uniq > [cdna.selfalign.tsv]

Does not work on mouse data as mouse transcript ensembl identifiers begin with ENSMUST whereas human ones begin with ENST, therefore the awk substr(row,1,15) command must be substr(row,1,18) for mouse. It's also worth noting that the awk command listed above is not specific for all operating systems and may fail on some. It may be worth making a quick tool for generating it instead of giving the above piping of commands.

[f0t1h]

Thank you for bringing this up. I will patch the tool as soon as possible to avoid this problem on non-human datasets.

[fhach]

@f0t1h any update on this?

[f0t1h]

I finished the fix, testing it at the moment.

[dvantwisk]

Has the fix been implemented?