VarSome Help Center

Compound heterozygous and homozygous

Q: I have a question: the algorithmic filter for compound heterozygous and homozygous variants seems very strict to me only allowing variants with a strong pathogenic ACMG criterion. Is there a way to run it less strict, not to miss all the missense variants?

A: It isn't possible at the moment but it is trivial to make it possible.However, the problem is that the ACMG classification criteria results in thousands of variants of unknown function (VAFs), so if we don't restrict these to "strong" pathogenic only, the results are so many that the filter is next to useless.

Also, I don't see why missense variants are relevant. The filter will return:

  • Variants classified as pathogenic, likely pathogenic or of unknown significance (but only if one of the strong pathogenic ACMG rules has fired for this variant) which are either homozygous or for which all of the following apply:
    • they are heterozygous variants in genes that carry at least one other heterozygous variant and no homozygous pathogenic variants.
    • are not in mitochondria

So allowing all variants classified as uncertain significance instead of only those for which at least one of the strong pathogenic ACMG rules has fired will make no change to what missense variants are returned.