DDIAgents: Mechanism-Conditioned Context Flow for Drug-Drug Interaction Prediction

2026-06-29

AI Summary

This work addresses the challenges in drug–drug interaction (DDI) prediction posed by diverse interaction mechanisms and the integration of heterogeneous evidence, which conventional methods struggle to adapt to dynamically. The authors propose DDIAgents, a novel multi-agent framework that introduces mechanism-conditioned contextual flows for the first time: a mechanism-aware planner instantiates expert agents and dynamically routes and allocates mechanism-specific knowledge sources. Through collaborative multi-agent reasoning and conclusion aggregation, the framework achieves both high adaptability and interpretability in DDI prediction. Experimental results demonstrate that DDIAgents significantly outperforms existing approaches—including feature engineering, graph neural networks, large language models, and prior agent-based methods—across multiple real-world DDI benchmarks, while providing interpretable, agent-level rationales for its predictions.

Abstract

Drug-drug interaction (DDI) prediction is essential for medication safety, yet it requires reasoning over heterogeneous biomedical evidence whose relevance changes across interaction mechanisms. We propose DDIAgents, a mechanism-conditioned multi-agent framework that performs DDI prediction through dynamic knowledge orchestration. Given a drug pair, a planner agent instantiates specialized expert agents, routes mechanism-relevant knowledge sources to each agent, and aggregates their analyses through a conclusion agent. By adapting context flow to the inferred interaction mechanism, DDIAgents reduces irrelevant information, supports complementary expert reasoning, and produces interpretable agent-level rationales. Extensive experiments on realistic DDI prediction benchmarks show that DDIAgents consistently outperforms existing feature-based, graph-based, LLM-based, and agent-based baselines. Beyond prediction performance, DDIAgents demonstrates how multi-agent systems can organize heterogeneous scientific knowledge for adaptive and interpretable AI4Science reasoning.

Problem

• Drug-Drug Interaction
• Interaction Mechanism
• Heterogeneous Biomedical Evidence
• Context Relevance
• Interpretable Prediction

Innovation

• mechanism-conditioned
• multi-agent framework
• dynamic knowledge orchestration
• interpretable reasoning
• drug-drug interaction prediction