Variant Interpretation is rarely a matter of data alone. It’s about connecting molecular evidence with the clinical picture. It’s a process that depends on experience, context, and pattern recognition. VarSome Picks brings that reasoning into the analysis pipeline.
VarSome Picks is an algorithmic filter that prioritizes variants based on their likelihood of explaining your chosen phenotype. It combines AI-driven logic with curated genetic and phenotypic knowledge, drawing on VarSome’s MolecularDB to identify the most plausible disease-causing variants for each case.
When phenotypes are provided, the algorithm identifies top candidate genes linked to them and applies the correct mode of inheritance. It then ranks variants accordingly, filtering out artefacts and highlighting those classified as Pathogenic, Likely Pathogenic, or strong VUS under ACMG guidelines.
Built on data from multiple sources, including HPO, ClinGen, OMIM, GenCC, and PanelApp, VarSome Picks integrates complementary views of gene-disease relationships for a more complete prioritization.
It supports single-sample germline, gene lists, and trio analyses from FASTQ or VCF, can run automatically when phenotypes are added or updated, and re-ranks results to reflect the latest information.
While VarSome Picks accelerates interpretation, expert review remains central. The precision of the output and clinical context help guide clinical judgement.
Watch our KnowledgeHub session with Laura Macias (Product Management Team Lead) and Charles Chapple (Chief Scientific Officer), to hear more about VarSome Picks.
For more information on VarSome Picks, visit our Help Center.
If you'd like to learn more about VarSome Clinical, visit our website.