Molecular Odor Prediction Based on Multi-Feature Graph Attention Networks

2025-02-03

AI Summary

This study addresses the challenge of mapping molecular structures to multidimensional odor perceptions in Quantitative Structure–Odor Relationship (QSOR) modeling. We propose MF-GAT, the first end-to-end multi-feature graph attention network that eliminates hand-crafted molecular descriptors and instead jointly learns local and global structural features by adaptively fusing atom-, bond-, and substructure-level molecular representations. Evaluated on multiple QSOR benchmark datasets—including Dravnieks and FlavorDB—MF-GAT achieves significant improvements in prediction accuracy across 12 odor attributes and sets a new state-of-the-art performance. To our knowledge, this is the first work to systematically integrate deep graph learning into olfactory computation. By enabling interpretable odor prediction through learned molecular representations, it establishes a novel paradigm at the intersection of computational olfaction and cheminformatics.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Qualitative Reasoning
• Search and Optimization: Learning to Search

Application Category

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

Abstract

Olfactory perception plays a critical role in both human and organismal interactions, yet understanding of its underlying mechanisms and influencing factors remain insufficient. Molecular structures influence odor perception through intricate biochemical interactions, and accurately quantifying structure-odor relationships presents significant challenges. The Quantitative Structure-Odor Relationship (QSOR) task, which involves predicting the associations between molecular structures and their corresponding odors, seeks to address these challenges. To this end, we propose a method for QSOR, utilizing Graph Attention Networks to model molecular structures and capture both local and global features. Unlike conventional QSOR approaches reliant on predefined descriptors, our method leverages diverse molecular feature extraction techniques to automatically learn comprehensive representations. This integration enhances the model's capacity to handle complex molecular information, improves prediction accuracy. Our approach demonstrates clear advantages in QSOR prediction tasks, offering valuable insights into the application of deep learning in cheminformatics.

Problem

• Molecular Structure
• Odor Prediction
• Quantitative Structure-Odor Relationship (QSOR)

Innovation

• Graph Attention Networks
• Molecular Scent Prediction
• Deep Learning in Cheminformatics