Saved in:
Bibliographic Details
Main Authors: Jiang, Tianyi, Wang, Zeyu, Yu, Shanqing, Xuan, Qi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.05844
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908306882166784
author Jiang, Tianyi
Wang, Zeyu
Yu, Shanqing
Xuan, Qi
author_facet Jiang, Tianyi
Wang, Zeyu
Yu, Shanqing
Xuan, Qi
contents Molecular property prediction is essential for applications such as drug discovery and toxicity assessment. While Graph Neural Networks (GNNs) have shown promising results by modeling molecules as molecular graphs, their reliance on data-driven learning limits their ability to generalize, particularly in the presence of data imbalance and diverse molecular substructures. Existing methods often overlook the varying contributions of different substructures to molecular properties, treating them uniformly. To address these challenges, we propose ASE-Mol, a novel GNN-based framework that leverages a Mixture-of-Experts (MoE) approach for molecular property prediction. ASE-Mol incorporates BRICS decomposition and significant substructure awareness to dynamically identify positive and negative substructures. By integrating a MoE architecture, it reduces the adverse impact of negative motifs while improving adaptability to positive motifs. Experimental results on eight benchmark datasets demonstrate that ASE-Mol achieves state-of-the-art performance, with significant improvements in both accuracy and interpretability.
format Preprint
id arxiv_https___arxiv_org_abs_2504_05844
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Adaptive Substructure-Aware Expert Model for Molecular Property Prediction
Jiang, Tianyi
Wang, Zeyu
Yu, Shanqing
Xuan, Qi
Machine Learning
Molecular property prediction is essential for applications such as drug discovery and toxicity assessment. While Graph Neural Networks (GNNs) have shown promising results by modeling molecules as molecular graphs, their reliance on data-driven learning limits their ability to generalize, particularly in the presence of data imbalance and diverse molecular substructures. Existing methods often overlook the varying contributions of different substructures to molecular properties, treating them uniformly. To address these challenges, we propose ASE-Mol, a novel GNN-based framework that leverages a Mixture-of-Experts (MoE) approach for molecular property prediction. ASE-Mol incorporates BRICS decomposition and significant substructure awareness to dynamically identify positive and negative substructures. By integrating a MoE architecture, it reduces the adverse impact of negative motifs while improving adaptability to positive motifs. Experimental results on eight benchmark datasets demonstrate that ASE-Mol achieves state-of-the-art performance, with significant improvements in both accuracy and interpretability.
title Adaptive Substructure-Aware Expert Model for Molecular Property Prediction
topic Machine Learning
url https://arxiv.org/abs/2504.05844