GFlowNets for Learning Better Drug-Drug Interaction Representations

TL;DR

This paper introduces a novel framework combining GFlowNets and VGAEs to generate synthetic data for rare drug-drug interactions, improving prediction accuracy and clinical reliability in imbalanced datasets.

Contribution

It presents a new method that leverages generative models to address class imbalance in DDI prediction, focusing on rare but critical interactions.

Findings

Improved prediction performance across interaction types.

Enhanced model balance with synthetic data generation.

Better detection of rare drug-drug interactions.

Abstract

Drug-drug interactions pose a significant challenge in clinical pharmacology, with severe class imbalance among interaction types limiting the effectiveness of predictive models. Common interactions dominate datasets, while rare but critical interactions remain underrepresented, leading to poor model performance on infrequent cases. Existing methods often treat DDI prediction as a binary problem, ignoring class-specific nuances and exacerbating bias toward frequent interactions. To address this, we propose a framework combining Generative Flow Networks (GFlowNet) with Variational Graph Autoencoders (VGAE) to generate synthetic samples for rare classes, improving model balance and generate effective and novel DDI pairs. Our approach enhances predictive performance across interaction types, ensuring better clinical reliability.