Contrastive Geometric Learning Unlocks Unified Structure- and Ligand-Based Drug Design

TL;DR

ConGLUDe is a unified contrastive geometric model that integrates structure- and ligand-based drug design, enabling improved virtual screening, pocket prediction, and target fishing through joint training on diverse data.

Contribution

It introduces a novel contrastive geometric learning framework that unifies structure- and ligand-based drug design in a single model, removing the need for pre-defined pockets.

Findings

Achieves competitive zero-shot virtual screening performance.

Outperforms existing methods on target fishing tasks.

Demonstrates state-of-the-art ligand-conditioned pocket prediction.

Abstract

Structure-based and ligand-based computational drug design have traditionally relied on disjoint data sources and modeling assumptions, limiting their joint use at scale. In this work, we introduce Contrastive Geometric Learning for Unified Computational Drug Design (ConGLUDe), a single contrastive geometric model that unifies structure- and ligand-based training. ConGLUDe couples a geometric protein encoder that produces whole-protein representations and implicit embeddings of predicted binding sites with a fast ligand encoder, removing the need for pre-defined pockets. By aligning ligands with both global protein representations and multiple candidate binding sites through contrastive learning, ConGLUDe supports ligand-conditioned pocket prediction in addition to virtual screening and target fishing, while being trained jointly on protein-ligand complexes and large-scale bioactivity data. Across diverse benchmarks, ConGLUDe achieves competitive zero-shot virtual screening performance, substantially outperforms existing methods on a challenging target fishing task, and demonstrates state-of-the-art ligand-conditioned pocket selection. These results highlight the advantages of unified structure-ligand training and position ConGLUDe as a step toward general-purpose foundation models for drug discovery.