What problem does it solve?
Medchem helps you triage large compound libraries by applying medicinal-chemistry drug-likeness rules, structural alert filters, and complexity/constraint checks to quickly prioritize molecules and avoid problematic chemotypes.
Core Features & Use Cases
- Drug-likeness & lead-likeness rules: Apply RO5, Veber, Oprea, CNS, leadlike, REOS, Golden Triangle, and related heuristics to estimate oral/druggable property space.
- Structural alert filters: Screen for PAINS and common/reactive/toxicophores-like substructures using Common Alerts, NIBR filters, and Lilly demerits scoring.
- Medicinal chemistry property screens: Filter by molecular complexity metrics and optional property constraints (e.g., MW/logP/TPSA/rotatable bonds).
- Group and pattern detection: Detect chemical groups (e.g., hinge binders, phosphate binders, Michael acceptors, reactive groups) and optionally extend with custom SMARTS.
- Batch + parallel workflows: Process large lists of molecules efficiently using parallel execution and structured results for downstream selection.
Quick Start
Use medchem to filter an input set of SMILES by applying Rule of Five and Veber, then removing compounds flagged by PAINS and common structural alerts, to produce a prioritized “drug-like” shortlist.
