CORAL: Learning Amyloid Fibril Ligand Docking with Cooperative Binding Rewards

2026-07-19

AI Summary

This study addresses the challenge of applying conventional molecular docking methods to amyloid fibrils, which exhibit a distinctive β-sheet groove binding mode and suffer from a scarcity of co-crystal structures. To overcome this limitation, the authors propose a reinforcement learning–based generative docking framework specifically tailored to the longitudinal groove geometry of amyloid fibrils. Notably, this approach introduces ligand–ligand cooperative stacking energy into the reward function for the first time. The model is trained on a newly constructed, expert-validated dataset of amyloid–ligand complexes. Experimental results demonstrate that the proposed method significantly outperforms existing docking tools on both native structures and the evaluation dataset, achieving higher accuracy in binding pose prediction and improved correlation in affinity estimation.

Abstract

A hallmark of neurodegenerative diseases such as Alzheimer's and Parkinson's is the aberrant aggregation of proteins into amyloid fibrils, and small molecules that selectively bind to these fibrils hold promise as diagnostics, imaging probes, and therapeutics. Predicting how such ligands bind to fibril targets, however, presents two fundamental challenges. First, resolved co-crystal structures of amyloid-ligand complexes are exceptionally scarce; even with recent advances in cryo-EM only a handful have been structurally characterized, making supervised training of docking models impractical for this target class. Second, amyloid fibrils present a binding mode fundamentally different from globular proteins: ligands intercalate into longitudinal cross-$β$ grooves and stack cooperatively along the fibril axis, a geometry that existing docking models are not designed to capture. To address these challenges, we present CORAL (COopeRative Amyloid Ligand docking), a reinforcement learning framework that trains a generative docking model to produce ligand pose distributions tailored to the cross-$β$ groove geometry. Our reward explicitly incorporates cooperative ligand-ligand stacking energy alongside protein-ligand docking affinity, directly capturing the distinctive binding geometry of amyloid fibrils. We further introduce a curated evaluation set of amyloid-ligand complexes constructed from model-generated poses validated by domain experts. Experiments on both experimentally resolved structures and this evaluation set demonstrate improved pose quality and binding affinity correlation over existing docking baselines.

Problem

• amyloid fibrils
• ligand docking
• cooperative binding
• cross-β groove
• neurodegenerative diseases

Innovation

• reinforcement learning
• cooperative binding
• amyloid fibrils
• molecular docking
• cross-β groove