Meaning of the 'variant origin' qualifier

[mbrush]

We initially defined the variantOriginQualifier as refining the statement/annotated knowledge to indicate that it hold for variants of a particular genetic origin (e.g. germline vs. somatic).

It has recently been questioned whether this is an accurate reflection of what is meant when indication of 'germline' or 'somatic' origin accompanies a variant annotation.

An alternative interpretation is that this reflects how the variant was detected, but should not impact/refine the meaning of the statement itself. e.g. for VPI, a 'somatic' qualifier would not mean that the statement asserts specifically that somatic versions of the variant to be pathogenic.

Apologies if am re-treading settled ground here, but the question was raised by a DP rep and I wanted to be sure we are clear here.

The comment leading to this ticket verbatim is below:

not sure i follow the description. something might be detected somatically, later proven to be germline but is still pathogenic for risk and relevant for selected tumour tx. i would just say how was the variant detected and including :somatically detected as a cover for variants detected by tu seq but insuff info to tell whether trully germline or only somatic. see below

[larrybabb]

We should verify this with clinical geneticists. The genomic source of the variant is a very important attribute in regards to the statement being made. I’ve never heard the scenarios where it might be discovered later that the author of the statement discovered it was wrong and therefore needed to revise the original version of the statement. I suppose anything is possible but this type of significant change to the meaning of a statement is something we will need to handle with special consideration. We should also involve ClinVar and VICC to see if they’ve ever encountered this and if so how it impacts those that rely on these statements as expert decisions.