This white paper from Schrödinger outlines a modern virtual screening (VS) workflow that leverages ultra-large libraries, machine learning-guided docking, and Absolute Binding Free Energy (ABFEP+) calculations to dramatically improve hit discovery. Traditional VS methods achieved only ~1–2% hit rates due to small library sizes and inaccurate scoring, but Schrödinger’s approach screens billions of compounds efficiently and accurately. The workflow uses active learning for docking, WScore for water-mediated binding accuracy, and ABFEP+ for rigorous affinity predictions. Applied across multiple campaigns, it consistently delivered double-digit hit rates, reduced false positives, and lowered wet-lab costs and timelines.

Join for free to read