FuseDiff: Symmetry-Preserving Joint Diffusion for Dual-Target Structure-Based Drug Design

2026-03-05

AI Summary

Existing methods struggle to jointly generate a single ligand tailored to dual-target binding pockets along with its target-specific binding conformations, often failing due to decoupled assumptions or excessive constraints. This work proposes FuseDiff, the first end-to-end diffusion model that integrates local contextual information from both pockets via a Dual-target Local Context Fusion module, enabling geometric adaptability while preserving symmetry. FuseDiff jointly models a shared molecular graph and two target-specific binding conformations, incorporating a symmetry-preserving mechanism and explicit bond generation to ensure topological consistency and geometric compatibility. Experiments demonstrate that FuseDiff achieves state-of-the-art docking performance on both benchmark and real-world dual-target systems and, for the first time, enables systematic pre-docking evaluation of dual-target conformation quality.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Mixed Discrete/Continuous Search
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Dual-target structure-based drug design aims to generate a single ligand together with two pocket-specific binding poses, each compatible with a corresponding target pocket, enabling polypharmacological therapies with improved efficacy and reduced resistance. Existing approaches typically rely on staged pipelines, which either decouple the two poses via conditional-independence assumptions or enforce overly rigid correlations, and therefore fail to jointly generate two target-specific binding modes. To address this, we propose FuseDiff, an end-to-end diffusion model that jointly generates a ligand molecular graph and two pocket-specific binding poses conditioned on both pockets. FuseDiff features a message-passing backbone with Dual-target Local Context Fusion (DLCF), which fuses each ligand atom's local context from both pockets to enable expressive joint modeling while preserving the desired symmetries. Together with explicit bond generation, FuseDiff enforces topological consistency across the two poses under a shared graph while allowing target-specific geometric adaptation in each pocket. To support principled training and evaluation, we derive a dual-target training set and use an independent held-out test set for evaluation. Experiments on the benchmark and a real-world dual-target system show that FuseDiff achieves state-of-the-art docking performance and enables the first systematic assessment of dual-target pose quality prior to docking-based pose search.

Problem

• dual-target drug design
• binding pose generation
• structure-based drug design
• polypharmacology
• joint modeling

Innovation

• FuseDiff
• dual-target drug design
• symmetry-preserving diffusion
• joint pose generation
• local context fusion