Human-supervised Agentic AI for Hypothesis Generation and Experimental Assistance in Drug Repurposing

Computational drug repurposing has focused on rapid candidate suggestion, yet real-world campaigns span a far broader lifecycle: generating hypotheses, designing experiments, analyzing assay data, and iteratively refining the hypotheses. Here, we demonstrate that agentic AI can operate throughout this lifecycle. To this end, we developed RepurAgent, a hierarchical multi-agent AI system comprising a supervisor agent and a planning agent that coordinate four specialized sub-agents (research, prediction, data, and report), with a human-in-the-loop design, with episodic memory and retrieval-augmented generation. The system is grounded in data, tools, and standard operating procedures specific to drug repurposing, developed within the REMEDi4ALL consortium. We demonstrated the agentic system across three scenarios spanning the various stages within the repurposing lifecycle: in Acute Myeloid Leukemia, a blinded expert evaluation indicated that RepurAgent produced substantially more novel and mechanistically credible candidates compared to a vanilla LLM baseline; in a retrospective COVID-19 antiviral screen, RepurAgent acted as an adaptive experimental collaborator, prioritizing compounds with AUC-ROC up to 0.99 and flagging confounders by comparing readouts across multiple experiments; and for Multiple Sulfatase Deficiency, it prioritized 81 high-confidence candidates from 5159 compounds, which were further corroborated by domain experts. These results demonstrate that agentic AI can support across the drug repurposing lifecycle, from hypothesis generation to experimental analysis.