LINKER: Learning Interactions Between Functional Groups and Residues With Chemical Knowledge-Enhanced Reasoning and Explainability

2025-09-03

AI Summary

Accurately identifying residue–functional group interactions between proteins and ligands is critical for elucidating molecular recognition mechanisms and enabling sequence-level rational drug design. This paper introduces the first deep learning framework that predicts biologically defined residue–functional group contacts—categorized by interaction type (e.g., hydrogen bonding, hydrophobic, π-interactions)—using only protein sequences and ligand SMILES strings, without requiring 3D structural inputs. Key methodological innovations include a structure-supervised attention mechanism to guide sequence modeling, chemically informed functional group abstraction and motif extraction, and joint SMILES–sequence encoding. Evaluated on the LP-PDBBind benchmark, our model substantially outperforms existing approaches, achieving high accuracy and strong agreement with biochemical experimental annotations. The framework delivers not only superior predictive performance but also enhanced interpretability and scalability for large-scale applications in early-stage drug discovery.

Technology Category

• Machine Learning: Bio-inspired Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate identification of interactions between protein residues and ligand functional groups is essential to understand molecular recognition and guide rational drug design. Existing deep learning approaches for protein-ligand interpretability often rely on 3D structural input or use distance-based contact labels, limiting both their applicability and biological relevance. We introduce LINKER, the first sequence-based model to predict residue-functional group interactions in terms of biologically defined interaction types, using only protein sequences and the ligand SMILES as input. LINKER is trained with structure-supervised attention, where interaction labels are derived from 3D protein-ligand complexes via functional group-based motif extraction. By abstracting ligand structures into functional groups, the model focuses on chemically meaningful substructures while predicting interaction types rather than mere spatial proximity. Crucially, LINKER requires only sequence-level input at inference time, enabling large-scale application in settings where structural data is unavailable. Experiments on the LP-PDBBind benchmark demonstrate that structure-informed supervision over functional group abstractions yields interaction predictions closely aligned with ground-truth biochemical annotations.

Problem

• Predicts protein-ligand residue-functional group interaction types
• Uses only protein sequences and ligand SMILES as input
• Eliminates dependency on 3D structural data requirements

Innovation

• Sequence-based model predicting residue-functional group interactions
• Structure-supervised attention with functional group motif extraction
• Chemical abstraction enabling inference without structural data