preprint · ChemRxiv
We present a reproducible, reactivity-weighted covalent-screening workflow that separates molecular recognition, intrinsic electrophile reactivity, and matched host-protease covalent capture, and we validate it before applying it. Screening 1,012 South African Natural Compounds Database entries against falcipain-2 and falcipain-3, a warhead/β-accessibility cascade retained 185 compounds, and a consensus of recognition docking, uniform-mode covalent docking, MM-GBSA, DFT reactivity descriptors, and predicted developability prioritized SANC00867. A retrospective benchmark (256 falcipain-2 actives, 5,016 propertymatched decoys) shows recognition docking alone does not enrich covalent falcipain-2 actives (ROC-AUC 0.53), and that the covalent signal is carried by warhead reactivity, not recognition: adding warhead reactivity significantly improves retrieval over docking on both propertymatched and stringent warhead-matched decoy benchmarks (ROC-AUC 0.62 vs 0.56, DeLong p = 0.015 in the stringent test), while docking alone remains near-random throughout; covalent pose recovery is validated against a crystallographic vinyl-sulfone complex, and the dual-target ranking is robust to food-vacuole (pH 5.5) preparation (Spearman ρ = 0.77 and 0.97 for FP-2 and FP-3). Across capped para-substituted cinnamate models, global electrophilicity and the site-local β-carbon Fukui index diverge under strong electron withdrawal (4-NO2 raises global ω yet collapses the β-carbon f+ through LUMO localization), so warhead tuning is governed by local, not global, electrophilicity. A matched host-cathepsin covalent counter-screen shows SANC00867’s apparent selectivity does not survive receptor-normalized analysis, reported as a cross-target scoring result rather than a selectivity claim. The workflow thus delivers a validated, transparent prioritization framework and a chemically interpretable design rule, with all inputs, scripts, and settings deposited for reuse.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.26434/chemrxiv.15008280/v1
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.