Warfarin Pharmacogenomics Dosing Cohort 500-patient

The Warfarin Pharmacogenomics Dosing Cohort (Sample) is a synthetic, physiologically anchored benchmark dataset designed for the development and validation of machine-learning algorithms in pharmacogenomic dosing. This sample comprises 500 patients (WARF_00000–WARF_00499), each observed under three counterfactual warfarin dosing strategies, yielding 1,500 unique regimens and 361,500 longitudinal profile rows. The sample is a verified, cell-identical subset of the full 10,000-patient release, and is intended for preliminary evaluation of algorithms prior to accessing the complete dataset. Full (download) dataset access: https://sentineldata.com.ua/dataset/warfarin-pharmacogenomics-dosing-cohort-data-full-generator-code MECHANISTIC MODEL & MATHEMATICAL FRAMEWORK The dataset was generated using an explicit mechanistic causal chain of warfarin pharmacology. Full mathematical specifications are documented in the associated technical guide (spec/SPEC.md). 1.1 Pharmacokinetic Model A two-compartment oral model with first-order absorption was utilized. Individual S-warfarin clearance is scaled by CYP2C9 diplotype (f_CYP2C9) and body weight: CL_i = θ_CL · f_CYP2C9 · (wt/70)^0.75 · exp(η_CL) 1.2 Pharmacodynamics (Vitamin K Cycle) VKORC1 inhibition is modeled via a saturable binding equation: I = C_c / (EC50/s_VKORC1 + C_c) Hepatic vitamin K stores follow first-order turnover kinetics: ... Full description with all tables is on the Wiki page of this project. Full dataset and purchase: https://sentineldata.com.ua/dataset/warfarin-pharmacogenomics-dosing-cohort-data-full-generator-code . Zenodo copy: https://doi.org/10.5281/zenodo.22178657