AI-Assisted De Novo Design of Small Molecule Candidates Targeting Five Priority Mitochondrial Proteins: A MitoCorex Proof of Concept
收藏资源简介：
Data deposit accompanying the manuscript: AI-Assisted De Novo Design of Small Molecule Candidates Targeting Five Priority Mitochondrial Proteins: A MitoCorex Proof of Concept This dataset contains molecular property profiles, docking score predictions, ADMET filter results, novelty assessments, and pharmacological characterizations for 15 de novo designed candidate therapeutics targeting five priority mitochondrial disease proteins (DNM1L, PINK1, NFE2L2/Keap1, NDUFV1, SDHA). Four rescue mechanism classes are represented: stalk-domain oligomerization inhibitor, mutant-selective kinase activator, non-covalent PPI disruptor, and cofactor-binding stabilizer. Structural data (SMILES) for all 15 novel candidates is restricted pending non-provisional patent filing and is available upon reasonable request to the corresponding author (jyborges@bu.edu). Tanimoto similarity assessments against ChEMBL confirm structural novelty for all candidates. Patent pending: US Provisional Application 64/018,624, filed March 27, 2026. This deposit contains pharmacological property data only. Novel compound structures are withheld pending patent protection. Manuscript series: DrugSynth AI / MitoCorex — Computational Drug Discovery for Mitochondrial Diseases (10 manuscripts).
