Representational Alignment with Chemical Induced Fit for Molecular Relational Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Peiliang, Yuan, Jingling, Xie, Qing, Zhu, Yongjun, Li, Lin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910246940704768
author Zhang, Peiliang
Yuan, Jingling
Xie, Qing
Zhu, Yongjun
Li, Lin
author_facet Zhang, Peiliang
Yuan, Jingling
Xie, Qing
Zhu, Yongjun
Li, Lin
contents Molecular Relational Learning (MRL) is widely applied in natural sciences to predict relationships between molecular pairs by extracting structural features. The representational similarity between substructure pairs determines the functional compatibility of molecular binding sites. Nevertheless, aligning substructure representations by attention mechanisms lacks guidance from chemical knowledge, resulting in unstable model performance in chemical space (\textit{e.g.}, functional group, scaffold) shifted data. With theoretical justification, we propose the \textbf{Re}presentational \textbf{Align}ment with Chemical Induced \textbf{Fit} (ReAlignFit) to enhance the stability of MRL. ReAlignFit dynamically aligns substructure representation in MRL by introducing chemical Induced Fit-based inductive bias. In the induction process, we design the Bias Correction Function based on substructure edge reconstruction to align representations between substructure pairs by simulating chemical conformational changes (dynamic combination of substructures). ReAlignFit further integrates the Subgraph Information Bottleneck during fit process to refine and optimize substructure pairs exhibiting high chemical functional compatibility, leveraging them to generate molecular embeddings. Experimental results on nine datasets demonstrate that ReAlignFit outperforms state-of-the-art models in two tasks and significantly enhances model's stability in both rule-shifted and scaffold-shifted data distributions.
format Preprint
id arxiv_https___arxiv_org_abs_2502_07027
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Representational Alignment with Chemical Induced Fit for Molecular Relational Learning
Zhang, Peiliang
Yuan, Jingling
Xie, Qing
Zhu, Yongjun
Li, Lin
Machine Learning
Artificial Intelligence
Molecular Relational Learning (MRL) is widely applied in natural sciences to predict relationships between molecular pairs by extracting structural features. The representational similarity between substructure pairs determines the functional compatibility of molecular binding sites. Nevertheless, aligning substructure representations by attention mechanisms lacks guidance from chemical knowledge, resulting in unstable model performance in chemical space (\textit{e.g.}, functional group, scaffold) shifted data. With theoretical justification, we propose the \textbf{Re}presentational \textbf{Align}ment with Chemical Induced \textbf{Fit} (ReAlignFit) to enhance the stability of MRL. ReAlignFit dynamically aligns substructure representation in MRL by introducing chemical Induced Fit-based inductive bias. In the induction process, we design the Bias Correction Function based on substructure edge reconstruction to align representations between substructure pairs by simulating chemical conformational changes (dynamic combination of substructures). ReAlignFit further integrates the Subgraph Information Bottleneck during fit process to refine and optimize substructure pairs exhibiting high chemical functional compatibility, leveraging them to generate molecular embeddings. Experimental results on nine datasets demonstrate that ReAlignFit outperforms state-of-the-art models in two tasks and significantly enhances model's stability in both rule-shifted and scaffold-shifted data distributions.
title Representational Alignment with Chemical Induced Fit for Molecular Relational Learning
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2502.07027