Assessing interaction recovery of predicted protein-ligand poses

2024/09/30 by David Errington, Errington, David, Constantin Schneider +5 · 4 citations
Biochemistry, Genetics and Molecular Biology · #Cell Image Analysis Techniques

paper · pdf · doi:10.48550/arxiv.2409.20227

Abstract

The field of protein-ligand pose prediction has seen significant advances in recent years, with machine learning-based methods now being commonly used in lieu of classical docking methods or even to predict all-atom protein-ligand complex structures. Most contemporary studies focus on the accuracy and physical plausibility of ligand placement to determine pose quality, often neglecting a direct assessment of the interactions observed with the protein. In this work, we demonstrate that ignoring protein-ligand interaction fingerprints can lead to overestimation of model performance, most notably in recent protein-ligand cofolding models which often fail to recapitulate key interactions.

Cited by

• AlphaFold3 for Structure-guided Ligand Discovery
• Generative molecule evolution using 3D pharmacophore for efficient Structure-Based Drug Design
• A fully automated benchmarking suite to compare macromolecular complexes
• FLOWR: Flow Matching for Structure-Aware De Novo, Interaction- and Fragment-Based Ligand Generation

Related

• Accurate Prediction of Ligand-Protein Interaction Affinities with Fine-Tuned Small Language Models
• Assessing the potential of deep learning for protein-ligand docking
• DeepRLI: A Multi-objective Framework for Universal Protein--Ligand Interaction Prediction
• FlexVDW: A machine learning approach to account for protein flexibility in ligand docking
• Protein-ligand binding representation learning from fine-grained interactions