Cross-Domain Molecular Relational Learning: Leveraging Chemical Structure-Activity Analysis

TL;DR

This paper introduces DisTrans, a novel method for cross-domain molecular relational learning that leverages structure-activity analysis to improve molecular representation across different domains.

Contribution

It proposes a domain adversarial training network with structural-semantic transfer discrepancy to enhance cross-domain molecular representations.

Findings

DisTrans outperforms 16 baseline methods in experiments.

It effectively learns domain-separable structural representations.

It maintains performance under significant inter-domain discrepancy.

Abstract

Recent advances in molecular representation integrates molecular topological and visual modalities, opening new avenues for precise Molecular Relational Learning (MRL). Existing MRL methods focus on intra-domain modeling, and their inherent domain-closed effect limits applicability to molecular science, particularly in elucidating cross-domain interaction mechanisms. Consequently, the imperative for Cross-Domain Molecular Relational Learning has become increasingly pressing. Benefiting from structure-activity analysis, we propose the Domain Adversarial Training Network with Structural-Semantic Transfer Discrepancy (DisTrans) to optimize cross-domain adaptive representation for molecular structures and visual images. 1) We employ the gradient reversal strategy based on substructure topological discrepancies between domains to learn the domain dependence of molecular structures. This strategy guides the model to adapt to the structural adjacency patterns in the target domain, generating domain-separable structural representations. 2) We apply the cross-domain representation guidance mechanism to align the functional-group semantic information between the source and target domains, learning cross-domain consistency information. The experimental results in two typical cross-domain strategies demonstrate that DisTrans outperforms 16 baseline methods, maintaining satisfactory performance even under pronounced inter-domain discrepancy.