Any experience using SomaticSeq for RNAseq SNV?

Question

[umn_bist]

I just wanted to discuss the reliability of the tool. I understand that it is a high performing variant caller but it seems that the tool was made for DNA-seq? Reason being, it uses BWA aligner as shown in this paper. However, it also uses mutation callers that work "well" for RNAseq variant calling (JointSNVMix2, Mutect, SomaticSniper).

I am also curious if using Mutect2 vcf files (which are in v4.2 format) rather than Mutect1 vcf files will cause any issues?

Again, just wanted to open up discussion to speak with any one with experience using SomaticSeq for RNAseq variant calling.

Thanks a lot for your time and help!

[Benn]

@umn_bist

Can you stop deleting your post? Your questions might be relevant for others that search this websites for answers.

[Pierre Lindenbaum]

b.nota is right: I've suspended your account for now.

[ivivek_ngs]

Seems like the guys is running a custom script to delete his/her posts, is he fidgeting with the biostars code base?

[Pierre Lindenbaum]

@istvan : can we restore those posts ?

[venu]

There are at least 7-8 posts on the home page like this. Its really annoying. :(

[John]

Well every time he/she edits, their post will get bumped. If it's only 8 so far, I dont think they're scripting it. And honestly, although I think the way they are going about it is wrong, if they post under the assumption that the edit button would always allow them to redact or delete things, it's unethical to prevent it now. We should ask them not to (done), suggest another more community-friendly way (??), and just live with whatever they decide to do :(

[Michael]

If the question is interesting and has been answered, it should be undeleted or restored. The content is provided on a CC-BY-SA license, and therefore we can decide if we want to keep it or not in my understanding. Without a traceable history it is hard to make this decision though. I am guessing that the user wants to avoid un-deletion of posts but this is just vandalism imo.

[venu]

May be user is thinking that people will copy his work.

[Istvan Albert]

Yes looks like the account has been compromised. We can restore these posts but it will take a bit.

[Michael]

Istvan, could you restore the posts from history, please?

[Istvan Albert]

The posts have been restored.

The author of the posts contacted me stating that their account was compromised.

[John]

I can't even begin to wrap my head around why anyone in their right mind would do such a thing... but I'm glad it's all fixed :)

Answer 1

[morovatunc]

I had been trying to operate SomaticSeq for 3 months before I switched to Bcbio. It s meant to be a great approach to mutation calling but it was so hard coded that I couldnt fix it to get it run for my own computer.

Li Tai Fang is a really nice guy who gets back pretty quick. However, the situation of somaticseq is that way because he was alone and couldnt make it modular to run in any machines.

[lethalfang]

It's been a long time coming, but we've finally found the time to have SomaticSeq dockerized:

• https://hub.docker.com/r/lethalfang/somaticseq/ (explains on https://github.com/bioinform/somaticseq)

And individual mutation callers as well, that you may invoke directly using SomaticSeq software:

• https://github.com/bioinform/somaticseq/tree/master/somaticseq/utilities/dockered_pipelines