Uncertainty-Aware Metabolic Stability Prediction with Dual-View Contrastive Learning

TL;DR

TrustworthyMS introduces a contrastive learning framework that improves molecular metabolic stability prediction by capturing bond-level features and providing reliable uncertainty estimates, advancing drug development research.

Contribution

It presents a novel dual-view contrastive learning approach with topology remapping and uncertainty quantification for more accurate and trustworthy MS predictions.

Findings

Outperforms state-of-the-art methods in predictive accuracy

Effectively captures bond-level topological features

Provides reliable uncertainty calibration

Abstract

Accurate prediction of molecular metabolic stability (MS) is critical for drug research and development but remains challenging due to the complex interplay of molecular interactions. Despite recent advances in graph neural networks (GNNs) for MS prediction, current approaches face two critical limitations: (1) incomplete molecular modeling due to atom-centric message-passing mechanisms that disregard bond-level topological features, and (2) prediction frameworks that lack reliable uncertainty quantification. To address these challenges, we propose TrustworthyMS, a novel contrastive learning framework designed for uncertainty-aware metabolic stability prediction. First, a molecular graph topology remapping mechanism synchronizes atom-bond interactions through edge-induced feature propagation, capturing both localized electronic effects and global conformational constraints. Second, contrastive topology-bond alignment enforces consistency between molecular topology views and bond patterns via feature alignment, enhancing representation robustness. Third, uncertainty modeling through Beta-Binomial uncertainty quantification enables simultaneous prediction and confidence calibration under epistemic uncertainty. Through extensive experiments, our results demonstrate that TrustworthyMS outperforms current state-of-the-art methods in terms of predictive performance.