ModelGenotype

For human data, we are using the GA4GH Variation Representation Specification (VRS). However, mouse data often use symbolic representations of variants that are not immediately computable, for instance.

{
  "count": 1,
  "komp_genotypes": [
    {
      "genesymbol": "Slc1a3",
      "genotype_id": 1855703,
      "genotypedescription": "Slc1a3<tm1.1(KOMP)Mbp> -/+",
      "genotypesymbol": "-/+",
      "jr": 18557,
      "latest_pheno": "2013-04-05",
      "strainstatus": "KOMP Complete"
    }
  ]
}

Question -- should we also support HGVS-style variants with the VRS framework?