PHENOCAUZ: Decoding the Molecular Basis of Human Symptoms from Mendelian Disease Biology

PHENOCAUZ maps 7,152 human symptoms to molecular determinants across 4,222 Mendelian proteins, achieving ~0.60 precision among the top 20 predictions for well-characterized symptoms.

• Why it matters: A general framework linking observable clinical phenotypes to their underlying molecular mechanisms is missing, limiting mechanistic interpretation and therapeutic prioritization.
• What they did: The proteome-scale framework learns symptom–protein relationships from Mendelian disease biology and predicts candidate symptom drivers, then validates enrichment and pathway context using held-out and curated comparisons.
• The result: It more than doubles candidate symptom-associated proteins and enables pathway-level mechanistic interpretation, including identifying proteins linked to severe adverse effects and improving target-based prediction of serious drug reactions.

AI summary of the abstract · may contain errors · not medical advice · read the paper.