populationgenomics / talos

Rare Disease variant reanalysis tool
MIT License
6 stars 4 forks source link

HPO tree traversal should be unbounded #433

Closed MattWellie closed 1 month ago

MattWellie commented 1 month ago

When 'fuzzy matching' between the HPO terms in a family, and HPO terms annotated on a family, we allow for a tolerance of 3 tiers separation in the HPO hierarchy between the two terms. This is arbitrary, and has not been extensively evaluated.

We should consider this becoming an unlimited scan to the root of the HPO tree, and see how that affects results.

Value here

Used here