An open agentic AI co-scientist for clinical and biomedical discovery

Abstract Biomedical research is constrained by the time and domain expertise required to analyze increasingly large and complex datasets. Here we present OpenScientist, an open-source agentic AI co-scientist designed to address this gap by investigating scientist-defined queries and generating verifiable, data-grounded analyses. OpenScientist was evaluated by domain experts across five biomedical case studies, completing analyses in minutes that otherwise would take months of human time and expertise. It accurately quantified Alzheimer’s blood biomarker performance, generated a plasma proteomic survival model that predicted survival in a hidden-evaluation cohort, proposed a reproducible mechanism linking tau pathology to altered lysosomal acidification, generated multiple myeloma hypotheses validated in an external cohort with attenuated signal in randomized controls, and demonstrated concordance with an independent human analysis in a blinded, prespecified proteomic comparison. These results support the use of OpenScientist to assist biomedical data analysis in concert with human experts. To support research involving sensitive clinical datasets, OpenScientist can operate in a fully air-gapped environment using open-source models for inference. OpenScientist demonstrates that open, auditable, agentic AI can support real-world clinical research by researching the literature, generating hypotheses, executing analyses, and generating insights from complex datasets.

Authors

Institutions

Publication Details

Journal
Nature Medicine
Published
2026-11-16
DOI
https://doi.org/10.1038/s41591-026-04808-6
Citations
3
Primary Topic
Artificial Intelligence in Healthcare and Education
Type
article
Field-Weighted Citation Impact
3.04

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

An open agentic AI co-scientist for clinical and biomedical discovery

Tammie L.S. Benzinger, Chris Mungall, Luca Foschini, Kaleigh F. Roberts et al.
3 citations
Nature Medicine
Artificial Intelligence in Healthcare and Education
3.04
article

An open agentic AI co-scientist for clinical and biomedical discovery

Tammie L.S. Benzinger, Chris Mungall, Luca Foschini, Kaleigh F. Roberts, Zachary B. Abrams, Adegoke A. Ojewole, Mukta Phatak, Sarah Biber, Randall John Bateman, Mahdi Moqri, Jinsheng Yu, Erik S. Musiek, Nomi L. Harris, Vitaliy Ovod, Luca Cappelletti, John Harry Caufield, Nicholas Heugel, Jineta Banerjee, Donald Leo Elbert, Brendan P. Lucey, Sarah A. Naylor, Nicolas R. Barthélemy, Eric C. Landsness, Justin Reese, Mathieu Bourdenx, Shaney E. Flores, Diego Galeano, Kejun Ying, Justin Melendez, Brian Andrew Gordon, Kejun Ying, James G. Bollinger, Yingxin He, Mahdi Moqri, J. Harry Caufield, Yan Li, Shaney Flores, Donald L. Elbert, Brian Gordon, Tammie L.S. Benzinger, James G. Bollinger, Melody Li, Sarah Biber, Taylor Pedersen
article en
3 citations

Abstract

Abstract Biomedical research is constrained by the time and domain expertise required to analyze increasingly large and complex datasets. Here we present OpenScientist, an open-source agentic AI co-scientist designed to address this gap by investigating scientist-defined queries and generating verifiable, data-grounded analyses. OpenScientist was evaluated by domain experts across five biomedical case studies, completing analyses in minutes that otherwise would take months of human time and expertise. It accurately quantified Alzheimer’s blood biomarker performance, generated a plasma proteomic survival model that predicted survival in a hidden-evaluation cohort, proposed a reproducible mechanism linking tau pathology to altered lysosomal acidification, generated multiple myeloma hypotheses validated in an external cohort with attenuated signal in randomized controls, and demonstrated concordance with an independent human analysis in a blinded, prespecified proteomic comparison. These results support the use of OpenScientist to assist biomedical data analysis in concert with human experts. To support research involving sensitive clinical datasets, OpenScientist can operate in a fully air-gapped environment using open-source models for inference. OpenScientist demonstrates that open, auditable, agentic AI can support real-world clinical research by researching the literature, generating hypotheses, executing analyses, and generating insights from complex datasets.

Nature Medicine
Broad Institute (US), James S. McDonnell Foundation (US), University of Minnesota (US), Allen Institute for Brain Science (US), Sage Bionetworks (US), Harvard University (US), Lawrence Berkeley National Laboratory (US), University of Fribourg (CH), University of Washington (US), Washington University in St. Louis (US), Biomedical Research Networking Center on Neurodegenerative Diseases (ES), UK Dementia Research Institute (GB), Washington University Medical Center (US), Stanford Medicine (US), Stanley Center for Psychiatric Research (US), University College London (GB), Universidad Nacional de Asunción (PY), Stanford University (US)
U.S. Department of Energy, Foundation for Barnes-Jewish Hospital, GHR Foundation, Good Ventures Foundation, National Institutes of Health, Office of Science, National Institute on Aging, Basic Energy Sciences
Openalex Percentile: Top 9%
Artificial Intelligence in Healthcare and Education
3.04
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon