{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,10,30]],"date-time":"2026-10-30T14:07:08Z","timestamp":1793369228448,"version":"4.5.0"},"posted":{"date-parts":[[2025,4,4]]},"group-title":"Genomics","reference-count":48,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T00:00:00Z","timestamp":1743724800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,4,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>\n                  Spatial transcriptomics has revolutionized our understanding of tissue organization by simultaneously capturing gene expression and spatial localization within intact tissues. However, analyzing these increasingly complex datasets requires specialized expertise across computational biology, statistics, and biological context. To address this challenge, we introduce the Spatial Transcriptomics AI Agent (STAgent), an autonomous multimodal agentic AI that integrates multimodal large language models (LLMs) with specialized computational tools to transform weeks-long analysis tasks into minutes of automated processing. Unlike conventional machine learning approaches that are limited to narrow, predefined tasks, STAgent leverages the emergent capabilities of multimodal LLMs \u2013 such as flexible reasoning, contextual understanding, and cross-modal integration \u2013 which allow it to adapt to novel data, execute multi-step analyses, and generate biologically meaningful insights with minimal human input. STAgent enables autonomous deep research through integrated capabilities, including dynamic code generation for complex analytical workflows, visual reasoning for interpreting spatial patterns, real-time retrieval of relevant peer-reviewd scientific literature, and synthesis of comprehensive, actionable reports. We applied STAgent to investigate the\n                  <jats:italic>in vivo<\/jats:italic>\n                  maturation of human stem cell-derived pancreatic cells (SC-pancreas) transplanted into immunodeficient mice. We generated single-cell spatial transcriptomics data spanning multiple developmental timepoints. STAgent autonomously (1) identified the maturation of initially scattered endocrine cells into well-defined islet-like structures, with predominantly peripheral \u03b1-cells surrounding \u03b2-cell cores supported by an expanding mesenchymal network; (2) revealed strengthening endocrine-endocrine cell interactions over time and, through context-aware gene set analysis, uncovered spatially resolved biological processes driving maturation; (3) unlike traditional analytical approaches, STAgent offers mechanistic explanations of spatial patterns, contextualizing findings with relevant literatures and developing cohesive insights into human pancreatic development. This agentic approach establishes a new paradigm in spatial transcriptomics analysis by substantially lowering the expertise barrier and reducing analysis time, accelerating biological and biomedical discovery.\n                <\/jats:p>","DOI":"10.1101\/2025.04.01.646731","type":"posted-content","created":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T17:45:25Z","timestamp":1743788725000},"source":"Crossref","is-referenced-by-count":17,"title":["Spatial transcriptomics AI agent charts hPSC-pancreas maturation\n                  <i>in vivo<\/i>"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-5786-9768","authenticated-orcid":false,"given":"Zuwan","family":"Lin","sequence":"first","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"},{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4282-2542","authenticated-orcid":false,"given":"Wenbo","family":"Wang","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"},{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8128-4119","authenticated-orcid":false,"given":"Arnau","family":"Marin-Llobet","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qiang","family":"Li","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Samuel D.","family":"Pollock","sequence":"additional","affiliation":[{"name":"Department of Cell and Developmental Biology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7286-9448","authenticated-orcid":false,"given":"Xin","family":"Sui","sequence":"additional","affiliation":[{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"},{"name":"Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Almir","family":"Aljovic","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jaeyong","family":"Lee","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jongmin","family":"Baek","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ningyue","family":"Liang","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinhe","family":"Zhang","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"},{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Connie Kangni","family":"Wang","sequence":"additional","affiliation":[{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"},{"name":"Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiahao","family":"Huang","sequence":"additional","affiliation":[{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"},{"name":"Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mai","family":"Liu","sequence":"additional","affiliation":[{"name":"Department of Cell and Developmental Biology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zihan","family":"Gao","sequence":"additional","affiliation":[{"name":"Department of Cell and Developmental Biology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hao","family":"Sheng","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jin","family":"Du","sequence":"additional","affiliation":[{"name":"School of Statistics, University of Minnesota Twin Cities, Minneapolis, MN, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stephen J.","family":"Lee","sequence":"additional","affiliation":[{"name":"Department of Cell and Developmental Biology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Brandon","family":"Wang","sequence":"additional","affiliation":[{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yichun","family":"He","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"},{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jie","family":"Ding","sequence":"additional","affiliation":[{"name":"School of Statistics, University of Minnesota Twin Cities, Minneapolis, MN, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiao","family":"Wang","sequence":"additional","affiliation":[{"name":"Broad Institute of MIT and Harvard, Cambridge, MA, USA"},{"name":"Department of Chemistry, Massachusetts Institute of Technology, Cambridge, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Juan R.","family":"Alvarez-Dominguez","sequence":"additional","affiliation":[{"name":"Department of Cell and Developmental Biology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jia","family":"Liu","sequence":"additional","affiliation":[{"name":"John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026041202193870000_2025.04.01.646731v1.1","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-022-01448-2"},{"key":"2026041202193870000_2025.04.01.646731v1.2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2024.07.040"},{"key":"2026041202193870000_2025.04.01.646731v1.3","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-022-01409-2"},{"key":"2026041202193870000_2025.04.01.646731v1.4","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2210.03629"},{"key":"2026041202193870000_2025.04.01.646731v1.5","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2402.15116"},{"key":"2026041202193870000_2025.04.01.646731v1.6","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.13601"},{"key":"2026041202193870000_2025.04.01.646731v1.7","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2206.07682"},{"key":"2026041202193870000_2025.04.01.646731v1.8","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2303.12712"},{"key":"2026041202193870000_2025.04.01.646731v1.9","unstructured":"Brown, T. et al. Language models are few-shot learners. Proceedings of the 34th International Conference on Neural Information Processing Systems 1877-1901 (NeurIPS, 2020). https:\/\/dl.acm.org\/doi\/abs\/10.5555\/3495724.3495883"},{"key":"2026041202193870000_2025.04.01.646731v1.10","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2302.00923"},{"key":"2026041202193870000_2025.04.01.646731v1.11","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2412.16720"},{"key":"2026041202193870000_2025.04.01.646731v1.12","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2205.11916"},{"key":"2026041202193870000_2025.04.01.646731v1.13","unstructured":"Wang, X. et al. Executable code actions elicit better LLM agents. Proceedings of the 41st International Conference on Machine Learning 1877-1901 (ICLR, 2024). https:\/\/dl.acm.org\/doi\/10.5555\/3692070.3694124"},{"key":"2026041202193870000_2025.04.01.646731v1.14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2406.00515"},{"key":"2026041202193870000_2025.04.01.646731v1.15","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i2.27888"},{"key":"2026041202193870000_2025.04.01.646731v1.16","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-08675-y"},{"key":"2026041202193870000_2025.04.01.646731v1.17","doi-asserted-by":"crossref","unstructured":"Chen, L. et al. Large language models are visual reasoning coordinators. Proceedings of the 37th International Conference on Neural Information Processing System 70115-70140 (NeurIPS, 2023). https:\/\/dl.acm.org\/doi\/10.5555\/3666122.3669194","DOI":"10.52202\/075280-3072"},{"key":"2026041202193870000_2025.04.01.646731v1.18","unstructured":"18. Claude 3.7 Sonnet System Card. (2025). https:\/\/anthropic.com\/claude-3-7-sonnet-system-card"},{"key":"2026041202193870000_2025.04.01.646731v1.19","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2410.21276"},{"key":"2026041202193870000_2025.04.01.646731v1.20","doi-asserted-by":"publisher","DOI":"10.1101\/2025.02.11.637754"},{"key":"2026041202193870000_2025.04.01.646731v1.21","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2404.18021"},{"key":"2026041202193870000_2025.04.01.646731v1.22","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06792-0"},{"key":"2026041202193870000_2025.04.01.646731v1.23","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2502.18864"},{"key":"2026041202193870000_2025.04.01.646731v1.24","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2408.06292"},{"key":"2026041202193870000_2025.04.01.646731v1.25","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-019-1168-5"},{"key":"2026041202193870000_2025.04.01.646731v1.26","doi-asserted-by":"publisher","DOI":"10.1016\/j.stem.2019.11.011"},{"key":"2026041202193870000_2025.04.01.646731v1.27","doi-asserted-by":"publisher","DOI":"10.1016\/j.celrep.2021.108850"},{"key":"2026041202193870000_2025.04.01.646731v1.28","doi-asserted-by":"publisher","DOI":"10.1016\/j.xpro.2019.100007"},{"key":"2026041202193870000_2025.04.01.646731v1.29","doi-asserted-by":"publisher","DOI":"10.1126\/science.aat5691"},{"key":"2026041202193870000_2025.04.01.646731v1.30","unstructured":"Anthropic. Claude 3.7 (2025). https:\/\/www.anthropic.com"},{"key":"2026041202193870000_2025.04.01.646731v1.31","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2303.08774"},{"key":"2026041202193870000_2025.04.01.646731v1.32","unstructured":"Lewis, P. et al. Retrieval-augmented generation for knowledge-intensive NLP tasks. Proceedings of the 34th International Conference on Neural Information Processing Systems 9459-9474 (NeurIPS, 2020). https:\/\/dl.acm.org\/doi\/abs\/10.5555\/3495724.3496517"},{"key":"2026041202193870000_2025.04.01.646731v1.33","unstructured":"33. Chroma Github https:\/\/github.com\/chroma-core\/chroma (2022)."},{"key":"2026041202193870000_2025.04.01.646731v1.34","unstructured":"34. Streamlit Inc. Streamlit. GitHub https:\/\/github.com\/streamlit (2024)."},{"key":"2026041202193870000_2025.04.01.646731v1.35","unstructured":"35. Chase, H. LangChain. GitHub https:\/\/github.com\/hwchase17\/langchain (2022)."},{"key":"2026041202193870000_2025.04.01.646731v1.36","unstructured":"36. SerpApi, LLC. Github https:\/\/github.com\/serpapi\/serpapi-python (2023)."},{"key":"2026041202193870000_2025.04.01.646731v1.37","doi-asserted-by":"publisher","DOI":"10.1038\/s43588-023-00528-w"},{"key":"2026041202193870000_2025.04.01.646731v1.38","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-021-01264-7"},{"key":"2026041202193870000_2025.04.01.646731v1.39","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-021-21246-9"},{"key":"2026041202193870000_2025.04.01.646731v1.40","doi-asserted-by":"publisher","DOI":"10.1186\/s13059-017-1382-0"},{"key":"2026041202193870000_2025.04.01.646731v1.41","doi-asserted-by":"publisher","DOI":"10.1038\/s41592-021-01358-2"},{"key":"2026041202193870000_2025.04.01.646731v1.42","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-019-11517-x"},{"key":"2026041202193870000_2025.04.01.646731v1.43","doi-asserted-by":"publisher","DOI":"10.1369\/0022155415583535"},{"key":"2026041202193870000_2025.04.01.646731v1.44","doi-asserted-by":"publisher","DOI":"10.1016\/j.trre.2012.11.003"},{"key":"2026041202193870000_2025.04.01.646731v1.45","doi-asserted-by":"publisher","DOI":"10.3390\/biomedicines11051426"},{"key":"2026041202193870000_2025.04.01.646731v1.46","doi-asserted-by":"publisher","DOI":"10.1900\/RDS.2010.7.132"},{"key":"2026041202193870000_2025.04.01.646731v1.47","doi-asserted-by":"publisher","DOI":"10.1155\/2013\/352315"},{"key":"2026041202193870000_2025.04.01.646731v1.48","doi-asserted-by":"publisher","DOI":"10.1016\/j.xpro.2023.102580"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.04.01.646731","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,12]],"date-time":"2026-04-12T09:19:49Z","timestamp":1775985589000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.1101\/2025.04.01.646731"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4,4]]},"references-count":48,"URL":"https:\/\/doi.org\/10.1101\/2025.04.01.646731","relation":{},"subject":[],"published":{"date-parts":[[2025,4,4]]},"subtype":"preprint"}}