MUDI: A Multimodal Biomedical Dataset for Understanding Pharmacodynamic Drug-Drug Interactions

TL;DR

This paper introduces MUDI, a large-scale multimodal biomedical dataset that combines various data types to better understand pharmacodynamic drug-drug interactions, and benchmarks learning methods on it.

Contribution

The paper presents MUDI, a comprehensive multimodal dataset for DDI understanding, and evaluates machine learning models using novel fusion strategies.

Findings

MUDI includes over 310,000 annotated drug pairs with multimodal data.

Benchmark models show varying performance on unseen drug pairs.

Open dataset and evaluation scripts are released for research use.

Abstract

Understanding the interaction between different drugs (drug-drug interaction or DDI) is critical for ensuring patient safety and optimizing therapeutic outcomes. Existing DDI datasets primarily focus on textual information, overlooking multimodal data that reflect complex drug mechanisms. In this paper, we (1) introduce MUDI, a large-scale Multimodal biomedical dataset for Understanding pharmacodynamic Drug-drug Interactions, and (2) benchmark learning methods to study it. In brief, MUDI provides a comprehensive multimodal representation of drugs by combining pharmacological text, chemical formulas, molecular structure graphs, and images across 310,532 annotated drug pairs labeled as Synergism, Antagonism, or New Effect. Crucially, to effectively evaluate machine-learning based generalization, MUDI consists of unseen drug pairs in the test set. We evaluate benchmark models using both late fusion voting and intermediate fusion strategies. All data, annotations, evaluation scripts, and baselines are released under an open research license.