MitoCorex: A Validated Artificial Intelligence Pipeline for Precision Drug Design Targeting Genetically Defined Mitochondrial Disease — End-to-End Proof of Concept

Data deposit accompanying the manuscript: MitoCorex: A Validated Artificial Intelligence Pipeline for Precision Drug Design Targeting Genetically Defined Mitochondrial Disease — End-to-End Proof of Concept This dataset contains the complete platform validation data for DrugSynth AI v1.0 applied to the MitoCorex mitochondrial disease validation case. Data includes platform performance benchmarks, comparative analysis against published computational drug design tools, ADMET baselines for 201 mitochondrial compounds, pathway network data (25 targets, 31 edges), binding pocket characterizations, fragment and toxicity alert registries, freedom-to-operate landscape data, and end-to-end pipeline assessment metrics. Validation demonstrates 11/15 candidates passing all four in silico filters (73% pass rate) with complete provenance tracing from genetic variant through final candidate. Platform architecture is disease-agnostic; mitochondrial disease is the validation case. Patent pending: US Provisional Application 64/018,624, filed March 27, 2026. This deposit contains validation data and scientific registries only. Platform scoring algorithms, reward function weights, and constraint decision logic are withheld pending non-provisional patent filing. Manuscript series: DrugSynth AI / MitoCorex — Computational Drug Discovery for Mitochondrial Diseases (10 manuscripts).