search

broadinstitute / variant-curation-portal

Web application for curating loss of function variants
https://lof.curation.broadinstitute.org
MIT License
4 stars 1 forks source link

no gene page #232

Open sannagudmundsson opened 3 years ago

sannagudmundsson commented 3 years ago

Quite often I get this

image

it does not seem to be gene-specific (and is always for both v3 and v2). I just go to gnomAD and look it up there when it happens but it is pretty frequent so would be great if it is an easy fix.

Page source if that helps

Line wrap

  | <!doctype html>   |   |   |     |   |   | Variant curation   |   |   |   |

  |     |   |   |   |  

nawatts commented 3 years ago

This looks like an issue with importing data into the curation portal. Not sure how this variant is in the curation portal without any VEP annotations. Maybe it got included because of annotations in RefSeq transcripts (#227) and we later deleted those?

sannagudmundsson commented 3 years ago

Yes, that's right. I assume the gene page can't be imported based on the gene name or something and not the variant? We have quite a lot of these annoying RefSeq variants.

--- Sanna Gudmundsson, Ph.D., Wallenberg Postdoctoral fellow, O'Donnell-Luria Lab Program in Medical and Population Genetics, Broad Institute of Massachusetts Institute of Technology and Harvard, Cambridge, MA, USA Division of Genetics and Genomics, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA Analytic and Translational Genetics Unit, Massachusetts General Hospital, Boston, MA, USA @. @.

On Tue, Aug 17, 2021 at 10:36 AM Nick Watts @.***> wrote:

This looks like an issue with importing data into the curation portal. Not sure how this variant is in the curation portal without any VEP annotations. Maybe it got included because of annotations in RefSeq transcripts (#227 https://github.com/macarthur-lab/variant-curation-portal/issues/227) and we later deleted those?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/macarthur-lab/variant-curation-portal/issues/232#issuecomment-900354688, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKXDZQLQ2IUVRJVJMN4H4GLT5JXW7ANCNFSM5CKAEDSA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email .

nawatts commented 3 years ago

I assume the gene page can't be imported based on the gene name or something and not the variant?

Which gene page to show is based on gene IDs in the variant's annotations, and the gnomAD browser currently only supports Ensembl IDs. Looking up the gene page by symbol doesn't work as well since multiple genes may have the same symbol, symbols change over time, etc.

nawatts commented 3 years ago

The script in this repository filters transcript consequences based on the provided gene IDs. So as long as Ensembl gene IDs are given, it would not include consequences for RefSeq transcripts.

https://github.com/macarthur-lab/variant-curation-portal/blob/2a466c9437b3eb45e42422fb72e6577c4131bbc8/scripts/get_gnomad_lof_variants.py#L162-L170

Based on Slack messages, my best guess is that variants for this projects were pulled using a different version of the script that filtered by gene symbol instead of ID. That would include both Ensembl and RefSeq consequences, since they both have the same gene symbols.