AANet: Virtual Screening under Structural Uncertainty via Alignment and Aggregation

TL;DR

AANet introduces an alignment-and-aggregation framework that improves virtual screening accuracy under structural uncertainty, especially for apo and predicted protein structures, by aligning ligand and pocket representations and aggregating binding site information.

Contribution

The paper presents a novel contrastive learning and cross-attention based approach that enhances virtual screening robustness on apo and predicted structures, addressing a key challenge in early-stage drug discovery.

Findings

Significantly outperforms state-of-the-art methods on apo structures, increasing EF1% from 11.75 to 37.19.

Maintains strong performance on holo structures, demonstrating versatility.

Provides a publicly available implementation for broader use.

Abstract

Virtual screening (VS) is a critical component of modern drug discovery, yet most existing methods--whether physics-based or deep learning-based--are developed around holo protein structures with known ligand-bound pockets. Consequently, their performance degrades significantly on apo or predicted structures such as those from AlphaFold2, which are more representative of real-world early-stage drug discovery, where pocket information is often missing. In this paper, we introduce an alignment-and-aggregation framework to enable accurate virtual screening under structural uncertainty. Our method comprises two core components: (1) a tri-modal contrastive learning module that aligns representations of the ligand, the holo pocket, and cavities detected from structures, thereby enhancing robustness to pocket localization error; and (2) a cross-attention based adapter for dynamically aggregating candidate binding sites, enabling the model to learn from activity data even without precise pocket annotations. We evaluated our method on a newly curated benchmark of apo structures, where it significantly outperforms state-of-the-art methods in blind apo setting, improving the early enrichment factor (EF1%) from 11.75 to 37.19. Notably, it also maintains strong performance on holo structures. These results demonstrate the promise of our approach in advancing first-in-class drug discovery, particularly in scenarios lacking experimentally resolved protein-ligand complexes. Our implementation is publicly available at https://github.com/Wiley-Z/AANet.