PheCode-guided multi-modal topic modeling of electronic health records improves disease incidence prediction and GWAS discovery from UK Biobank

Phenome-wide association studies (PheWAS) rely on disease definitions derived from diagnostic codes, often failing to leverage the full richness of electronic health records (EHR). We present MixEHR-SAGE, a PheCode-guided multi-modal topic model that integrates diagnoses, procedures, and medications to enhance phenotyping from large-scale EHRs. By combining expert-informed priors with probabilistic inference, MixEHR-SAGE identifies over 1,000 interpretable phenotype topics from UK Biobank (UKB) data. Applied to 350,000 individuals with high-quality genetic data, MixEHR-SAGE-derived risk scores accurately predict incident type 2 diabetes (T2D) and leukemia diagnoses. Subsequent genome-wide association studies (GWAS) using these continuous risk scores uncovered novel diasease-associated loci, including PPP1R15A for T2D and JMJD6/SRSF2 for leukemia, that were missed by traditional binary case definitions. These results highlight the potential of probabilistic phenotyping from multi-modal EHRs to improve genetic discovery. The MixEHR-SAGE software is publicly available at: https://github.com/li-lab-mcgill/MixEHR-SAGE.