search

jyhehir / mobster

For more details about Mobster please see
https://jyhehir.github.io/mobster/index.html
GNU General Public License v3.0
9 stars 9 forks source link

Feature request on `resources/MobsterVCF/MobsterVCF-0.0.1-SNAPSHOT.jar` #19

Open SHuang-Broad opened 6 years ago

SHuang-Broad commented 6 years ago

I've found (actually IGV did) two issues with the output from this tool:

The second problem is easier to fix, while the first one needs more work (it needs access to reference fasta at the least).

It would be nice if the source code for that could be made public and I am happy to issue PRs for these two.

Thanks!

jyhehir commented 6 years ago

Sure will upload, was actually delayed because ultimately I would like to have this functionality integrated into the main of Mobster. Regarding the first issue, this may be difficult, need to take into account TSD and aswell as non-preicise predictions. Anyway, first things, first will upload the code. Thanks for contributing!

SHuang-Broad commented 6 years ago

Thanks, Jayne!

If I understand correctly, the REF allele problem is not too difficult to solve, because per the VCF spec, the REF allele must be at least one base long, and start with the base pointed to by POS. And since we are dealing with insertions here, the REF allele should always be 1 base long, right?

jyhehir commented 6 years ago

Yes this is true, was thinking about TSD's and events with no supporting split reads. But most I think you idea is the most practical solution. (Have you seen the new code in devlop?)

SHuang-Broad commented 6 years ago

Yes, I just saw that after you mentioned. Let me know if there's anything I can help with that (test run, code review etc.)

kaanokay commented 4 years ago

Hi,

No REF allele issue was fixed or not? I could not visualize my mobster vcf files in IGV.

Thanks.