FABind: Fast and Accurate Protein-Ligand Binding

TL;DR

FABind addresses the challenge of fast and accurate protein–ligand docking by unifying pocket prediction and docking into a single end-to-end framework. It introduces ligand-informed pocket prediction, a three-step FABind layer with E(3)-equivariant updates, and a training pipeline that couples pocket and docking tasks through scheduled sampling and distance-map refinement. The method demonstrates strong generalization to unseen proteins and achieves substantial efficiency gains, including around 170× speedups over DiffDock while maintaining competitive docking accuracy. These contributions offer a practical, scalable approach for drug discovery that reduces reliance on external pocket-detection modules and improves inference speed without sacrificing pose quality.

Abstract

Modeling the interaction between proteins and ligands and accurately predicting their binding structures is a critical yet challenging task in drug discovery. Recent advancements in deep learning have shown promise in addressing this challenge, with sampling-based and regression-based methods emerging as two prominent approaches. However, these methods have notable limitations. Sampling-based methods often suffer from low efficiency due to the need for generating multiple candidate structures for selection. On the other hand, regression-based methods offer fast predictions but may experience decreased accuracy. Additionally, the variation in protein sizes often requires external modules for selecting suitable binding pockets, further impacting efficiency. In this work, we propose $\mathbf{FABind}$, an end-to-end model that combines pocket prediction and docking to achieve accurate and fast protein-ligand binding. $\mathbf{FABind}$ incorporates a unique ligand-informed pocket prediction module, which is also leveraged for docking pose estimation. The model further enhances the docking process by incrementally integrating the predicted pocket to optimize protein-ligand binding, reducing discrepancies between training and inference. Through extensive experiments on benchmark datasets, our proposed $\mathbf{FABind}$ demonstrates strong advantages in terms of effectiveness and efficiency compared to existing methods. Our code is available at https://github.com/QizhiPei/FABind

Figures (3)

• Figure 1: An overview of FABind. Left: The pocket prediction module takes the whole protein and the ligand as input and predicts the coordinates of the pocket center, where the ligand is randomly placed at the center of the protein. After determining the pocket center, a pocket is defined as a set of amino acids within a fixed radius around the center. Subsequently, the docking module moves the ligand to the pocket center and the ligand-pocket pair iteratively goes through the FABind layers to obtain the final pose prediction. $M$ and $N$ are the number of layers in pocket prediction and docking. Right: Architecture of FABind layers. Each layer contains three modules: independent message passing takes place within each component to update node embeddings and coordinates; cross-attention captures correlations between residues and ligands and updates embeddings only; and interfacial message passing focuses on the interface, attentively updating coordinates and representations.
• Figure 2: Case studies. Pose prediction by FABind (green), DiffDock (wheat), E3Bind (magenta), TankBind (cyan), and EquiBind (orange) are placed together with protein target, and RMSDs to ground truth (red) are reported. (a) For large unseen protein (PDB 6NPI), FABind successfully identifies the pocket, while the others are all off-site. (b) For the other protein (PDB 6G3C), all models find the right pocket, among which FABind predicts the most precise and valid binding pose as the DiffDock but with faster speed.
• Figure 3: Additional case studies. Pose prediction by FABind (green), DiffDock (wheat), E3Bind (magenta), TankBind (cyan), and EquiBind (orange) are placed together with protein target structure, and RMSD to ground truth (red) are reported. (a) For unseen protein (PDB 6N93), FABind, E3Bind, and TankBind successfully identify the pocket, among which FABind predicts the most precise binding pose with the lowest RMSD $2.7$Å, while the other methods are all off-site. (b) For PDB 6JB4, all deep learning models find the right pocket, among which FABind predicts the most precise binding pose with the lowest RMSD $1.9$Å.