Predicting Activity Cliffs for Autonomous Medicinal Chemistry

TL;DR

This study develops a model to predict activity cliffs in medicinal chemistry, reducing experimental efforts by identifying key positions where small changes cause large potency shifts, using a large dataset and open-source tools.

Contribution

Introduces an 11-feature model with 3D pharmacophore context for activity cliff prediction, generalizing across targets and scaffolds, and provides open-source code and webapp.

Findings

Model achieves high NDCG@3 scores (0.910) for activity cliff prediction.

Reduces chemist exploration from 3.1 to 2.1 positions, halving initial experiments.

Predicting modification outcomes from structure alone is limited (Spearman 0.268).

Abstract

Activity cliff prediction - identifying positions where small structural changes cause large potency shifts - has been a persistent challenge in computational medicinal chemistry. This work focuses on a parsimonious definition: which small modifications, at which positions, confer the highest probability of an outcome change. Position-level sensitivity is calculated using 25 million matched molecular pairs from 50 ChEMBL targets across six protein families, revealing that two questions have fundamentally different answers. "Which positions vary most?" is answered by scaffold size alone (NDCG@3 = 0.966), requiring no machine learning. "Which are true activity cliffs?" - where small modifications cause disproportionately large effects, as captured by SALI normalization - requires an 11-feature model with 3D pharmacophore context (NDCG@3 = 0.910 vs. 0.839 random), generalizing across all six protein families, novel scaffolds (0.913), and temporal splits (0.878). The model identifies the cliff-prone position first 53% of the time (vs. 27% random - 2x lift), reducing positions a chemist must explore from 3.1 to 2.1 - a 31% reduction in first-round experiments. Predicting which modification to make is not tractable from structure alone (Spearman 0.268, collapsing to -0.31 on novel scaffolds). The system is released as open-source code and an interactive webapp.