RareAgent: Self-Evolving Reasoning for Drug Repurposing in Rare Diseases

2025-10-07

AI Summary

Knowledge graph completion and GNN-based methods for rare-disease drug repurposing suffer from limited performance due to the scarcity of prior drug–disease associations. Method: This paper proposes a self-evolving multi-agent reasoning framework that shifts from passive prediction to active evidence seeking. It orchestrates collaborative agents to construct dynamic evidence graphs, incorporates task-specific adversarial debate to validate hypotheses, and refines agent decisions via post-hoc policy analysis and textual feedback. Furthermore, it distills transferable heuristic rules from successful reasoning paths to enhance transparency and generalizability. Contribution/Results: On instruction-driven tasks, the framework achieves an 18.1% improvement in AUPRC over state-of-the-art baselines. It generates clinically coherent, interpretable reasoning chains, significantly improving both accuracy and trustworthiness in rare-disease drug repurposing.

Technology Category

• Multiagent Systems: Adversarial Agents
• Knowledge Representation and Reasoning: Reasoning with Beliefs
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Computational drug repurposing for rare diseases is especially challenging when no prior associations exist between drugs and target diseases. Therefore, knowledge graph completion and message-passing GNNs have little reliable signal to learn and propagate, resulting in poor performance. We present RareAgent, a self-evolving multi-agent system that reframes this task from passive pattern recognition to active evidence-seeking reasoning. RareAgent organizes task-specific adversarial debates in which agents dynamically construct evidence graphs from diverse perspectives to support, refute, or entail hypotheses. The reasoning strategies are analyzed post hoc in a self-evolutionary loop, producing textual feedback that refines agent policies, while successful reasoning paths are distilled into transferable heuristics to accelerate future investigations. Comprehensive evaluations reveal that RareAgent improves the indication AUPRC by 18.1% over reasoning baselines and provides a transparent reasoning chain consistent with clinical evidence.

Problem

• Addresses computational drug repurposing for rare diseases
• Overcomes lack of reliable drug-disease association signals
• Transforms passive pattern recognition into active evidence-seeking reasoning

Innovation

• Self-evolving multi-agent system for drug repurposing
• Dynamic evidence graph construction through adversarial debates
• Transferable heuristics distilled from successful reasoning paths