DiffDock-PP: Rigid Protein-Protein Docking with Diffusion Models

Mohamed Amine Ketata, Cedrik Laue, Ruslan Mammadov, Hannes Stark, Menghua Wu, Gabriele Corso, Céline Marquet, Regina Barzilay, Tommi S. Jaakkola

Keywords: Machine Learning for Science, Protein-Protein Docking, Geometric Deep Learning, Diffusion Generative Models, Graph Neural Networks

TL;DR: We propose to treat rigid protein-protein docking as a generative modelling problem and develop a diffusion generative model that performs on par with search-based docking methods at a fraction of their computational cost.

Abstract: Understanding how proteins structurally interact is crucial to modern biology, with applications in drug discovery and protein design. Recent machine learning methods have formulated protein-small molecule docking as a generative problem with significant performance boosts over both traditional and deep learning baselines. In this work, we propose a similar approach for rigid protein-protein docking: DiffDock-PP is a diffusion generative model that learns to translate and rotate unbound protein structures into their bound conformations. We achieve state-of-the-art performance on DIPS with a median C-RMSD of 4.85, outperforming all considered baselines. Additionally, DiffDock-PP is faster than all search-based methods and generates reliable confidence estimates for its predictions.

Community Implementations: [ 3 code implementations](https://www.catalyzex.com/paper/diffdock-pp-rigid-protein-protein-docking/code)