search

Clinical-Genomics / scout

VCF visualization interface
https://clinical-genomics.github.io/scout
BSD 3-Clause "New" or "Revised" License
147 stars 43 forks source link

view Tomte VCF in Scout #4700

Open Lucpen opened 3 days ago

Lucpen commented 3 days ago

Is your feature request related to a problem in the current program to new available techology or software? Please describe and add links/citations if appropriate. Would it be possible to view the vcf output of tomte in Scout, there is information there about SNPs and VAF that could potentially be interesting

Describe the solution you'd like To see vcf in Scout

Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered.

Additional context Add any other context or screenshots about the feature request here.

dnil commented 3 days ago

Are there any particular keys missing that you don't get if you do that today?

Overall, this is probably more of a policy question. In principle, it is already be possible to load a WTS SNV VCF on the case, assuming there is no WGS SNV VCF, which we would typically wish there is. The original intent for Tomte or a follow up to it afaik was to annotate the WGS with info from the WTS calls, e.g. with AF imbalance and potential allele specific expression calls. I even understood there is a drop module for this, if we can just provide the SNV vcf as input?

Lucpen commented 2 days ago

Yes, I was more thinking on the Allelic imbalance part (there is a field regarding that in the vcf), but they should be able to asses that (at least partially) from the igv view

dnil commented 2 days ago

Good, what is the key? I had a quick look through the VCF header, but must have missed it?

Lucpen commented 2 days ago

I think this one will have to wait until you guys are back from vacation as they were added in the latest release, which is not what has been run in production. It will probably be added next week, I will add a comment with the header as soon as that happens

dnil commented 2 days ago

I see. Have you solved the issue with pulling in wgs as well so that we get imbalance calls also for completely silent alleles?

Lucpen commented 2 days ago

Not yet, that will happen in the future. At the moment we only detect allelic imbalance rather than monoallelic expression. In other words, we just look at how many calls come from each allele if more than one allele is found in RNAseq. Then we use Bootstrapann to calculate significance.

dnil commented 2 days ago

It’s a start! 👍🏻

dnil commented 2 days ago

Maybe you already have it in the newer version, but I noticed the old one was lacking genmod annotation? As you probably know, there is an nf-core module for it you could drop in, and get inheritance model support for that VCF as well as a basic scoring. Maybe you would like to incorporate the imbalance prediction in your score model as well?