[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86535-en":3,"doc-seo-86535-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86535,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","NVAITC AI Scientist Governed End-to-End Research System Hypertension GWAS Case Study","Agentic research systems are emerging for coordinating scientific workflows, but biomedical deployment in institutional environments demands governed mechanisms for planning, data access control, workflow orchestration, evidence tracking, reproducibility, and human oversight. NVAITC AI Scientist (NAIS) provides a governed end-to-end agentic framework that keeps protected data within privacy boundaries while integrating proposal review, execution planning, reproducible orchestration, evidence generation, and scientist-in-the-loop review. Validation on a real-world hypertension GWAS using hospital-linked data from 286,422 individuals shows expert-comparable reliability and iterative definition refinement. NAIS also supports drug-induced liver injury prediction with AUC 0.842.","arXiv :2607 . 1 1084v 1 [ cs .AI] 13 Jul 2026  \nNVAITC AI Scientist: A Governed End-to-End Research System  \n—A Hypertension GWAS Case Study  \nEddie Huang 1,* , Ken Liao 1 , Iven Fu 1 , Yang-Hsien Lin 1 , Chao-Shun Zhan 1 , Andy Liao 1 , Virginia Chen 1 , Johnson Sun 1 , Pika Wang 1 , Richard Huang 1 , Jiun-Cheng Jiang 1 , Ting-Yuan Liu2,3 , Hsing-Fang Lu2,3,4 , Ray Y. Lee5 , Chi-Chou Liao2 , Simon See 1 , and Fuu-Jen  \nTsai2,6,7,8,**  \n1 NVIDIA AI Technology Center (NVAITC), NVIDIA Corporation  \n2 Department of Medical Research, China Medical University Hospital, Taichung 40402, Taiwan  \n3 Master Program for Digital Health Innovation, China Medical University, Taichung 406040, Taiwan  \n4 Laboratory for Statistical and Translational Genetics, RIKEN Center for Integrative Medical Sciences, Yokohama, Japan  \n5 AI-Driven Genomic Medicine and Drug Discovery Lab, China Medical University Hospital, Taichung 40402, Taiwan  \n6 School of Chinese Medicine, China Medical University, Taichung 40402, Taiwan  \n7 Division of Pediatric Genetics, Children’s Hospital of China Medical University, Taichung 40447, Taiwan  \n8 Department of Medical Laboratory Science and Biotechnology, Asia University, Taichung 41354, Taiwan  \n* Corresponding author: [tzungchih@nvidia.com](tzungchih@nvidia.com)  \n** Corresponding author: [000704@tool.caaumed.org.tw](000704@tool.caaumed.org.tw)  \nAbstract  \nAgentic research systems are emerging as a new paradigm for coordinating scientific workflows beyond isolated model inference, code generation, or statistical analysis. However, deployment in institutional biomedical environments requires governed mechanisms for research planning, data access, workflow orchestration, evidence tracking, reproducibility, and human oversight. We present NVAITC AI Scientist (NAIS), a governed end-to-end agentic research system designed to support domain-general scientific workflows while keeping protected data within institutional privacy boundaries. NAIS integrates proposal review, execution planning, governed computational routing, reproducible workflow orchestration, evidence generation, and scientist-in-the-loop oversight. We validate NAIS in a real-world hypertension genome-wide association study (GWAS) using hospital-linked genotype and electronic health record (EHR) data from 286,422 individuals under an aggregate-only data policy. The agent planned cohort extraction, orchestrated GWAS execution, generated quality-control summaries and visualizations, and drafted publicationoriented outputs. Systematic comparison with independently curated expert analyses showed that human-AI review identified phenotype discrepancies and enabled iterative refinement of the hypertension definition. After team-directed reconciliation, the agent-orchestrated GWAS reproduced established hypertension-associated loci, including FGF5, ATP2B1, CNNM2, FTO, and GRB14, with the strongest signal at FGF5 reaching − log10 p ≈ 70. As a secondary  \ndemonstration beyond GWAS, NAIS also supported a drug-induced liver injury prediction workflow, achieving a multimodal graph neural network area under the curve (AUC) of 0.842 .  \nThese results demonstrate that governed agentic research systems can support scalable AI-assisted biomedical discovery while producing scientifically reliable outputs comparable to expert-led workflows.  \nKeywords: Agentic AI, AI Scientist, AI governance, Biomedical research, Genome-wide association study (GWAS)  \n1 Introduction  \nAgentic research systems are beginning to extend artificial intelligence from isolated tasks, such as model inference, code generation, literature summarization, or statistical analysis, toward the coordination of complete scientific workflows [1–9] . These systems promise to assist researchers across planning, execution, interpretation, and manuscript preparation. However, most demonstrations of autonomous or semi-autonomous research agents have focused on open data, simulated environments, local software execut","cbCaicwzhviG6GS3","https://ap.wps.com/l/cbCaicwzhviG6GS3","pdf",3390898,4,1,22,"English","en",105,"# Introduction\n## Key challenges in institutional biomedical deployment\n## NAIS overview and governed end-to-end design","[{\"question\":\"What problem does NVAITC AI Scientist (NAIS) address in biomedical agentic research?\",\"answer\":\"NAIS addresses the need for governed mechanisms—covering research planning, protected data access, workflow orchestration, evidence tracking, reproducibility, and human oversight—when deploying agentic systems in institutional biomedical settings.\"},{\"question\":\"How does NAIS handle protected data during end-to-end research?\",\"answer\":\"NAIS separates reasoning and orchestration from data governance: agents submit approved actions through controlled interfaces, protected data stay within institutional infrastructure, and only governed artifacts like aggregate summaries, metrics, plots, logs, and manuscript-oriented evidence are released.\"},{\"question\":\"How was NAIS validated and what were the main outcomes in the hypertension GWAS case study?\",\"answer\":\"NAIS was validated on a real-world hypertension GWAS using hospital-linked genotype and EHR data from 286,422 individuals under an aggregate-only data policy. The system reproduced known hypertension-associated loci, with the strongest signal at FGF5 reaching −log10 p ≈ 70, and human-AI review helped identify phenotype discrepancies for iterative refinement.\"}]",1784212462,55,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"nvaitc-ai-scientist-governed-end-to-end-research-system-hypertension-gwas-case-study","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/nvaitc-ai-scientist-governed-end-to-end-research-system-hypertension-gwas-case-study/86535/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does NVAITC AI Scientist (NAIS) address in biomedical agentic research?","Question",{"text":75,"@type":76},"NAIS addresses the need for governed mechanisms—covering research planning, protected data access, workflow orchestration, evidence tracking, reproducibility, and human oversight—when deploying agentic systems in institutional biomedical settings.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does NAIS handle protected data during end-to-end research?",{"text":80,"@type":76},"NAIS separates reasoning and orchestration from data governance: agents submit approved actions through controlled interfaces, protected data stay within institutional infrastructure, and only governed artifacts like aggregate summaries, metrics, plots, logs, and manuscript-oriented evidence are released.",{"name":82,"@type":73,"acceptedAnswer":83},"How was NAIS validated and what were the main outcomes in the hypertension GWAS case study?",{"text":84,"@type":76},"NAIS was validated on a real-world hypertension GWAS using hospital-linked genotype and EHR data from 286,422 individuals under an aggregate-only data policy. The system reproduced known hypertension-associated loci, with the strongest signal at FGF5 reaching −log10 p ≈ 70, and human-AI review helped identify phenotype discrepancies for iterative refinement.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]