Evidence that was never looked up must abstain

The rank is a weighted mean of four auditable components. Which ones may score is decided per job, from what the annotation run actually produced.

phenotype — 0.35

Share of the patient's HPO terms annotated to the gene, each weighted by its information content.

always scores

rarity — 0.25

Allele frequency in gnomAD, as a step function a reviewer can read.

only if the run looked up frequencies

consequence — 0.20

Predicted impact severity: HIGH, MODERATE, LOW, MODIFIER.

always scores

model — 0.20

LightGBM P(pathogenic) from variant effect.

only with CADD or AlphaMissense

Before — the bug

rarity_score(None) → 1.00

"No frequency in this run" was read as "absent from gnomAD, therefore maximally rare", and every variant collected a free 0.25. The model, meanwhile, scored 0.887 on features it had never seen. Two of four components were fiction, and the score looked fully informed.

After — abstention and renormalisation

component → null, weights renormalised

The job records has_frequencies and has_effect_scores. Absent components drop out and the remaining weights renormalise over 0.55, so the score stays on a 0–1 scale and still means the same thing. The UI prints "not looked up", never a bar.

Published Loeys–Dietz case, run without a VEP cache	score	phenotype (0.64)	rarity	consequence (0.36)	model
TGFBR2 3:30672252 G>T missense — the published diagnosis	0.855	1.00 (30/30)	not looked up	0.60	not looked up
OSBPL10 3:31748090 missense — incidental	0.218	0.00	not looked up	0.60	not looked up

Both are rare missense variants, identical on every piece of evidence this run holds except one. The phenotype is what separates a published diagnosis from an incidental variant in a lipid-transport gene — which is the argument for phenotype-driven triage, in one table.