MTPNet: Multi-Grained Target Perception for Unified Activity Cliff Prediction

2025-06-05

AI Summary

Current active cliff prediction methods are limited to single-target scenarios, hindering multi-target drug design. To address this, we propose MTPNet, a multi-target-aware neural network that— for the first time—incorporates 3D structural knowledge of receptor proteins as conditional guidance signals to jointly model macro-level target semantics and micro-level binding-pocket semantics, enabling dynamic, multi-granularity molecular representation optimization. MTPNet integrates graph neural networks, protein sequence and structure encoding, conditional molecular representation learning, and multi-granularity attention mechanisms. Evaluated on 30 benchmark datasets, MTPNet achieves an average 18.95% reduction in RMSE over state-of-the-art GNN-based methods. It establishes a unified, scalable predictive framework for multi-target-oriented compound optimization and rational drug design.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Metareasoning and Metaheuristics
• Multiagent Systems: Other Foundations of Multi Agent Systems

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Activity cliff prediction is a critical task in drug discovery and material design. Existing computational methods are limited to handling single binding targets, which restricts the applicability of these prediction models. In this paper, we present the Multi-Grained Target Perception network (MTPNet) to incorporate the prior knowledge of interactions between the molecules and their target proteins. Specifically, MTPNet is a unified framework for activity cliff prediction, which consists of two components: Macro-level Target Semantic (MTS) guidance and Micro-level Pocket Semantic (MPS) guidance. By this way, MTPNet dynamically optimizes molecular representations through multi-grained protein semantic conditions. To our knowledge, it is the first time to employ the receptor proteins as guiding information to effectively capture critical interaction details. Extensive experiments on 30 representative activity cliff datasets demonstrate that MTPNet significantly outperforms previous approaches, achieving an average RMSE improvement of 18.95% on top of several mainstream GNN architectures. Overall, MTPNet internalizes interaction patterns through conditional deep learning to achieve unified predictions of activity cliffs, helping to accelerate compound optimization and design. Codes are available at: https://github.com/ZishanShu/MTPNet.

Problem

• Predicts activity cliffs in drug discovery
• Handles multi-target interactions for molecules
• Optimizes molecular representations using protein semantics

Innovation

• Multi-grained protein semantic conditions optimize molecular representations
• Macro-level Target Semantic guidance enhances prediction accuracy
• Micro-level Pocket Semantic guidance captures critical interaction details