Geometric Multi-color Message Passing Graph Neural Networks for Blood-brain Barrier Permeability Prediction

2025-07-24

AI Summary

Existing graph neural networks (GNNs) for blood–brain barrier permeability (BBBP) prediction neglect 3D geometric information and struggle to capture long-range spatial interactions. To address this, we propose GeoMP-GNN, a geometry-aware multi-color message-passing GNN. It constructs weighted color-coded subgraphs based on atomic types, integrates 3D coordinate encoding with scaffold partitioning, and explicitly models molecular spatial conformation and the spatial contributions of functional groups. On multiple benchmark datasets, GeoMP-GNN achieves classification AUC-ROC scores of 0.9704 and 0.9685, and regression performance with RMSE = 0.4609 and Pearson *r* = 0.7759—significantly surpassing state-of-the-art methods. The core innovation lies in unifying atomic-type-driven geometric subgraph modeling with a multi-color message-passing mechanism, thereby enhancing representational capacity and generalization for central nervous system (CNS)-relevant drug properties.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

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

Abstract

Accurate prediction of blood-brain barrier permeability (BBBP) is essential for central nervous system (CNS) drug development. While graph neural networks (GNNs) have advanced molecular property prediction, they often rely on molecular topology and neglect the three-dimensional geometric information crucial for modeling transport mechanisms. This paper introduces the geometric multi-color message-passing graph neural network (GMC-MPNN), a novel framework that enhances standard message-passing architectures by explicitly incorporating atomic-level geometric features and long-range interactions. Our model constructs weighted colored subgraphs based on atom types to capture the spatial relationships and chemical context that govern BBB permeability. We evaluated GMC-MPNN on three benchmark datasets for both classification and regression tasks, using rigorous scaffold-based splitting to ensure a robust assessment of generalization. The results demonstrate that GMC-MPNN consistently outperforms existing state-of-the-art models, achieving superior performance in both classifying compounds as permeable/non-permeable (AUC-ROC of 0.9704 and 0.9685) and in regressing continuous permeability values (RMSE of 0.4609, Pearson correlation of 0.7759). An ablation study further quantified the impact of specific atom-pair interactions, revealing that the model's predictive power derives from its ability to learn from both common and rare, but chemically significant, functional motifs. By integrating spatial geometry into the graph representation, GMC-MPNN sets a new performance benchmark and offers a more accurate and generalizable tool for drug discovery pipelines.

Problem

• Predicts blood-brain barrier permeability for CNS drugs
• Incorporates 3D geometric data into graph neural networks
• Improves accuracy over topology-only molecular property models

Innovation

• Incorporates atomic-level geometric features
• Uses weighted colored subgraphs for spatial relationships
• Enhances message-passing with long-range interactions