BRIDGE: A Four-Tier, AOP-Anchored In Silico IATA for Hepatic Steatosis — Code, Data, and Results

Résumé fourni par la source

Reproducible deposit for the BRIDGE pipeline, a four-tier in silico Integrated Approach to Testing and Assessment (IATA) for hepatic steatosis hazard identification, anchored to the Adverse Outcome Pathway (AOP) framework. The deposit contains: (1) all Python pipeline scripts (35 scripts across three sequential pipelines: chemical, target, and Bridge); (2) pre-computed Bridge Pipeline Results and Tier 3 AOP scoring JSONs for the pharmaceutical external validation set (n=40), environmental external validation set (n=40), and internal calibration set (n=40); (3) all required reference data files including AOP-Wiki event tables (AOPs 34, 36, 57, 58, 59, 60, 61, 62, 232, 318, 517, 518, 529), GO-constrained STRING PPI network files, Human Protein Atlas liver expression data (hpa_normal_tissue.tsv), pre-prepared receptor PDBQT files for 26 holo structures representing 12 MIE targets (PPARα, PPARγ, PPARδ, LXRα, LXRβ, AhR, CAR, PXR, FXR, GR, HNF4A, NRF2), and the Cotterill et al. (2020) LDA model; (4) Electronic Supplementary Material files ESM_1 through ESM_11 as reported in the associated manuscript. Tier 3 scores are not fully reproducible due to live CTD database dependency (documented limitation — see README); all other tiers are fully deterministic. The deposited per-compound results reflect CTD data accessed in June 2026. Companion manuscript: Chrysochoou et al. (under review), Archives of Toxicology. Related dataset: Gomatam et al. (2026), doi:10.5281/zenodo.20236679.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.