search

ClinGen / gene-and-variant-curation-tools

ClinGen's gene and variant curation interfaces (GCI & VCI). Developed by Stanford ClinGen team.
https://curation.clinicalgenome.org/
MIT License
3 stars 1 forks source link

Adding predictors to Variant type tab #2

Open wrightmw opened 2 years ago

wrightmw commented 2 years ago

Several predictors have been recommended/requested to be added to the VCI Variant Type predictors table, additionally the predictor table should be updated to be in line with Pejaver et al 2022 and Stenton et al (in preparation).

1) VEST4, which is available in myvariant.info (Source display name (name shown in MyVariant.info output) VEST4 (vest4)) 2) BayesDel, which is available in myvariant.info

We took a survey of users on tools in July 2021 and presented the results on a ClinGen tools topic. (see: https://forms.gle/qEnYX3BfGqEvidaUA). The feedback from the tools topic call included the information that the SVI is working on a tools recommendation.

For each predictor need to know: Score Range Impact Threshold Prediction values URL to link out to for further information Info for mouseover "i" table in the Prediction column header

SP ticket: https://broadinstitute.atlassian.net/browse/CGSP-652 Ticket transferred from old platform repo: https://github.com/ClinGen/clincoded/issues/2310

cgpreston commented 11 months ago

Based on the current missense variants decision tree for V4 curation there are some tools highlighted that we do not currently bring into the VCI,

Specifically:

  1. VEST
  2. BayesDel
  3. MutPred2/MutPred-InDel

Will revisit this ticket with the proxy group in September 2023.

wrightmw commented 11 months ago

@cgpreston These 3 are all available in myvariant.info... they bring in data from dbNSFP, see here for list: https://sites.google.com/site/jpopgen/dbNSFP

cgpreston commented 8 months ago

https://docs.google.com/presentation/d/1S5NJIlbVmuPNMWzaaqI3m808Ma_o2nlg9tEyXpjk5gM/edit#slide=id.g295a07e3824_0_92

cgpreston commented 8 months ago

Feedback from VCI proxies on 11/17:

  1. Add in the predictors called out in the ACMG v4 guidelines
  2. Remove the "prediction" row/letters
  3. Display column "benign impact threshold" - display the benign supporting threshold from Pejaver et al
  4. Display column "pathogenic impact threshold" - display the pathogenic supporting threshold from Pejaver et al
  5. Add in some text providing additional curator guidance and linking to the paper
cgpreston commented 6 months ago

Feedback from VCI proxies on 1/29/24:

  1. Consider a two-tiered system, one with the Pejaver et al scores and another the current style (to accommodate the predictor tools not in Pejaver et al)
  2. Try to get some stats from the CSpec and /or ERepo on the usage of the various tools
  3. Add a 5th column for "Prediction", to maintain the current experience for the Pejaver et al tools
  4. Double check the values listed in the mockups (REVEL threshold needs to be updated for example)