Phenotype Association
Rank genes and variants using patient phenotypes (HPO) plus upstream evidence (pathogenicity + clinical mapping).
What this plugin does
This step links patient phenotypes to candidate genes/variants so the clinical workflow can prioritize
the most plausible explanations and drive downstream literature evidence retrieval.
- Phenotypes: HPO terms (OmniObject, upload, or paste)
- Candidates: normalized variant table from upstream steps (OmniObject, upload, or paste)
- Scoring: phenotype similarity at gene-level + optional upstream scores
- Outputs: ranked genes TSV, ranked variants TSV, agent-ready JSON summary
Expected input formats
Phenotypes (HPO)
One per line (or TSV first column).
HP:0001250 HP:0002373 HP:0004322
Candidate variants (TSV)
Minimum recommended columns:
variant_id, gene,
pathogenicity_score, clinical_score.
variant_id gene pathogenicity_score clinical_score 1:100000:A:G SCN1A 0.85 0.40 17:43071000:A:T BRCA1 0.55 0.60
Next step in the pipeline
After ranking, send top genes/variants to Literature RAG to collect evidence, then generate a structured clinical insight report.