El Agente Fármaco: A Scalable AI Scientist for Drug Discovery
Abstract
Computational drug discovery offers a growing range of tools for proposing and evaluating molecules, but combining them and deciding what to do next still requires substantial expert effort. Here, we introduce El Agente Fármaco, an AI scientist that uses modern large language model (LLM) technology to turn discovery goals expressed in everyday language into complex, scalable computational screening workflows. The agentic harness selects scientific tools, reviews their results, and adjusts subsequent steps to the target and the user's objectives. Our approach centers on a shared scientific memory: Fármaco Atlas, which stores molecules, structures, and results together with their provenance -- a record of how they were obtained. The agent uses targeted database queries to retrieve the evidence needed for each decision. A single request can launch calculations for an entire collection of molecules, without requiring a separate LLM decision for every compound. This design limits the information the model must process as screening collections grow, while keeping the underlying results available for inspection and further analysis. The integrated tools support molecule generation, prediction of binding and molecular properties, and searches for compounds available for purchase. In a retrospective comparison with an expert-led campaign, the agent proposed chemically distinct candidates with competitive docking scores and higher calculated drug-likeness under explicitly imposed selection criteria. In a 22-task benchmark against plain tool calls and Codex, the agent achieved the highest observed mean model-judged score, with lower median cumulative token use and peak context. Fármaco provides a foundation for flexible, scalable drug discovery workflows that connect scientific reasoning with an enduring, traceable record of computational work.
Related
- Deep learning coupled with scalable domain-specific structural validation expands RNA virus discovery from metatranscriptomes.
- Target Discovery and Drug Intervention in Inflammatory Diseases： A Paradigm Shift from Mechanistic Dissection to Precision Therapy
- Fragment-based drug discovery strategy reveals an unprecedented allosteric binding site in Mycobacterium tuberculosis DNA gyrase
- Special Issue “Natural Products in Drug Discovery and Development: 2nd Edition”
- AI-integrated iPSC multi-omics: advancing phenotype discovery and drug target identification in rare diseases
BibTeX
@misc{muller2026agente,
title = {{El Agente Fármaco: A Scalable AI Scientist for Drug Discovery}},
author = {Marcel Müller and Mohammad Ghazi Vakili and Yoshihiro Ogura and Ella Ashoori and Shimeng Wang and Alessandro S. Nascimento and Eddie Calleja and Yutong Zhao and Yimeng Liu and Ella Miray Rajaonson and Aiwei Yin and Jiaru Bai and Yunheng Zou and Abraham C. Stern and Varinia Bernales and Alán Aspuru-Guzik},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.03832},
eprint = {2611.03832},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.03832},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - El Agente Fármaco: A Scalable AI Scientist for Drug Discovery
AU - Marcel Müller
AU - Mohammad Ghazi Vakili
AU - Yoshihiro Ogura
AU - Ella Ashoori
AU - Shimeng Wang
AU - Alessandro S. Nascimento
AU - Eddie Calleja
AU - Yutong Zhao
AU - Yimeng Liu
AU - Ella Miray Rajaonson
AU - Aiwei Yin
AU - Jiaru Bai
AU - Yunheng Zou
AU - Abraham C. Stern
AU - Varinia Bernales
AU - Alán Aspuru-Guzik
PY - 2026
DA - 2026/11/02
PB - arXiv
DO - 10.48550/arxiv.2611.03832
UR - https://arxiv.org/abs/2611.03832
AB - Computational drug discovery offers a growing range of tools for proposing and evaluating molecules, but combining them and deciding what to do next still requires substantial expert effort. Here, we introduce El Agente Fármaco, an AI scientist that uses modern large language model (LLM) technology to turn discovery goals expressed in everyday language into complex, scalable computational screening workflows. The agentic harness selects scientific tools, reviews their results, and adjusts subsequent steps to the target and the user's objectives. Our approach centers on a shared scientific memory: Fármaco Atlas, which stores molecules, structures, and results together with their provenance -- a record of how they were obtained. The agent uses targeted database queries to retrieve the evidence needed for each decision. A single request can launch calculations for an entire collection of molecules, without requiring a separate LLM decision for every compound. This design limits the information the model must process as screening collections grow, while keeping the underlying results available for inspection and further analysis. The integrated tools support molecule generation, prediction of binding and molecular properties, and searches for compounds available for purchase. In a retrospective comparison with an expert-led campaign, the agent proposed chemically distinct candidates with competitive docking scores and higher calculated drug-likeness under explicitly imposed selection criteria. In a 22-task benchmark against plain tool calls and Codex, the agent achieved the highest observed mean model-judged score, with lower median cumulative token use and peak context. Fármaco provides a foundation for flexible, scalable drug discovery workflows that connect scientific reasoning with an enduring, traceable record of computational work.
ER -