Update the VCI gnomAD table from gnomAD v2.1.1 to v4.0.0 data

[cgpreston]

Sub-ticket for gnomAD v4 update #47

The goal of this ticket is to do the "like for like" update of the VCI population tab from gnomAD v2 to v4 data, along with deprecating a few sections of the population tab and expanding the variant search parameters in gnomAD.

Items to be completed:

• [ ] Confirm the LDH has the appropriate data in the agreed upon GA4GH format for gnomAD v4, specifically the data needed is:

  1. The gnomAD Genetic Ancestry Group names, Allele Counts, Allele Numbers, Number of homozygotes, and Allele Frequency.
  2. The exome/genomes filters
  3. The number of hemizygotes for the Genetic Ancestry Groups (not required for this ticket, but needed for other development).
  4. The GroupMax filtering AF, including allele counts, allele numbers, allele frequency, the filtering allele frequency, and the fraction of individuals with >20X coverage for Exomes, Genomes and the combined total.

• [ ] Add gnomAD v4.1.0 data, ensure the following changes (see screenshots below).

  1. Update the left column table name from “Population” to “Genetic Ancestry Group”
  2. Ensure the new Genetic Ancestry Groups will be appropriately represented in the table
  3. Update header with correct coordinates, version, and genome build

• [ ] Update the VCI logic to look for all variants in gnomAD v4 (right now we limit the variants we search for to very small ones)

  • For example we don't pull in gnomAD v2 data for the variant CA914148882, we should pull in any variant that's available in gnomAD v4.0.0.

• [ ] Remove the EXAC data table, ESP, and 1000 Genomes data from the VCI population tab (confirmed and finalized on proxy call 6/24/2024). This means we will display gnomAD v2, gnomAD v4, and PAGE on the pop tab (with v4 at the top)

• [ ] Remove the HMAF table from the VCI population tab

SP ticket: https://broadinstitute.atlassian.net/browse/CGSP-524

[cgpreston]

We may want to retire all of the non-gnomAD resources in the population tab, (specifically ExAC, PAGE, 1000 genomes and ESP) as all are listed as contributory projects, I've requested feedback on whether they are completely subsumed by gnomAD v4 or not here: https://discuss.gnomad.broadinstitute.org/t/contributing-projects-esp-1000-genomes-page/205?u=christine_preston

[wrightmw]

@cgpreston is there a ticket for adding in constraint scores from gnomAD?