Bingxin Zhao
I am an associate professor in Statistics and Data Science at the Wharton School of the University of Pennsylvania, with a secondary appointment in the Department of Medicine at the Perelman School of Medicine.
My research develops statistical and AI methods for science and medicine, with a focus on integrating multimodal data, such as imaging, genomics, and electronic health records, to enable discovery and translation. My work has advanced understanding of the brain and inter-organ connections (e.g., heart–brain links) in relation to brain disorders and healthy aging. I have also contributed to unifying statistical rigor, software engineering, and agentic AI to build practical systems for scientific and medical research.
Email / Google Scholar / Wharton Faculty Page / LinkedIn / CV
News
- Oct 2026Our review, Multi-organ imaging genetics: towards an integrated view of brain, body and brain diseases, with Zirui Fan, is published in Nature Reviews Neurology. We introduce brain-focused multi-organ imaging genetics and outline a roadmap for agentic AI co-scientists toward an integrated view of brain health and disease.
- Aug 2026Our paper, Self-rankings as a predictor of scientific impact beyond peer review, with Buxin Su, Weijie Su, and colleagues, is published in Nature Computational Science, featured in Nature Computational Science News & Views and a Nature Index research highlight, and covered by Penn Today, The Daily Pennsylvanian, and Wharton AI & Analytics Insight. In a large-scale experiment at a leading AI conference, papers ranked highest by their own authors received twice as many citations, and self-rankings predicted citations better than peer-review scores.
- Mar 2026Our agentic AI platform was selected as a finalist for the Alzheimer’s Insights AI Prize, a $1M award supporting the use of agentic AI in Alzheimer’s disease research.
- Feb 2026My article, Engineering AI co-scientists for statistical genetics applications, is published in Nature Genetics. It argues that AI co-scientists can accelerate genetic discovery and translation when built on domain-specific data and infrastructure, with community standards for rigor and responsible use.
- Dec 2025Our work on the brain and inter-organ connections, such as heart–brain links, in relation to brain disorders and healthy aging, was recognized by the inaugural UK Biobank Scientific Impact Awards.
- Mar 2025Received the IMS (Institute of Mathematical Statistics) Tweedie New Researcher Award for my statistical research.
Research
My work mainly spans across four directions:
- Inter-organ connections: Heart-brain connections (Science 2023), Multi-organ imaging genetics (Nature Reviews Neurology 2026), Multi-organ Mendelian randomization (Nature Biomedical Engineering 2026), Rare coding variants across organs (Nature Communications 2026), Plasma proteomics and organ imaging (Nature Communications 2026), Eye-brain connections (Nature Communications 2024)
- Genetics of the human brain: White matter microstructure (Science 2021), Functional networks (Nature Genetics 2022), Regional brain volumes (Nature Genetics 2019), White matter and mental health traits (Molecular Psychiatry 2021), Transcriptome-wide associations of brain structures (Nature Communications 2021)
- Statistical methods and theory: LD score regression (Annals of Statistics 2025), Block-wise and reference panel-based estimators (Annals of Statistics 2024), Trans-ancestry genetic correlation (Journal of the American Statistical Association 2024), Genetic correlation estimation (Journal of the American Statistical Association 2022), Cross-trait prediction accuracy (Biometrics 2023), Polygenic risk scores (Biometrics 2022)
- AI for science and research platforms: AI co-scientists (Nature Genetics 2026), Self-rankings (Nature Computational Science 2026), BIGA (Journal of the American Statistical Association 2026), PennPRS (Nature Methods, accepted)
Platforms for science and medicine
We build publicly accessible platforms that bring advanced statistics, genetics, and AI to researchers worldwide, from cloud computing for genetic analyses to AI systems that generate genetic evidence for drug development and evaluate scientific manuscripts.
PennPRS
Train polygenic risk score models online from GWAS summary statistics, with no individual-level data or local computing needed. Supports single- and multi-ancestry training with multiple methods, direct access to GWAS Catalog data, and free pretrained weights for thousands of phenotypes.
BIGA
Analyze the genetics shared across traits and diseases from GWAS summary statistics in the cloud. Offers unified pipelines such as genetic correlation, curated GWAS data for thousands of phenotypes, and tools to upload data, submit jobs, and share results.
TigerAI
Query AI-generated genetic evidence on drug targets to judge which target–indication pairs are likely to succeed in clinical development. Its evidence is as informative as expert curation with far broader coverage, and the platform also benchmarks the latest large language models.
PASS
An agentic AI system that evaluates a manuscript and positions it within its field: it finds the closest related work, traces how the topic has developed, and predicts which journals it fits best. It outperforms general-purpose LLMs and established journal-finder tools.
Group
Students and researchers interested in working with our group are welcome to email me with a CV and a short note on your interests.
Postdoctoral researchers
Zihang Wang (Penn Statistics and Data Science)
Jiawen Chen (Penn Statistics and Data Science)
Zhirui Li (Penn Statistics and Data Science)
Zichen Zhang (Penn Statistics and Data Science)
Ph.D. thesis students
Zirui Fan (Penn Statistics and Data Science)
Yujue Li (Purdue Statistics; co-advised with Fei Xue)
Buxin Su (Penn AMCS)
Eddie Yang (Penn Statistics and Data Science; co-advised with Nancy Zhang)
Yuxin Li (Penn AMCS)
Graduate students
Yuxin Guo (Purdue PULSe Program, the Paschou Lab)
Bingxuan Li (UIUC Computer Science)
Cameron Beeche (Penn Bioengineering, the CorLab)
Jiyao Zhang (Penn AMCS, the Jin Lab)
Zhijie Duan (Weill Cornell Population Health Sciences)
Xiaojing Sun (Purdue Statistics, the Xue Group)
Undergraduate researchers
Shenmin Zhang (Penn Computer Science)
Lucy Pu (Penn Bioengineering)
Alumni
Ph.D. thesis
Xiaochen Yang (Purdue Statistics; current position: Software Engineer, Google)
Kai Xia (UNC Biostatistics; current position: Associate Professor, Department of Psychiatry, UNC Chapel Hill)
Master’s thesis
Jingxuan Bao (Penn GCB Program)
Qipeng Zhan (Penn AMCS Program)
Max Dudek (Penn GCB Program)
Graduate and undergraduate
Juan Shu (Purdue Statistics)
Yijun Fan (Penn AMCS)
Xiyao Wang (Columbia Computer Science)
Tianqi Shang (Penn GCB Program, the Shen Lab)
Fred Zhang (Penn Statistics and Data Science)
Zhaoqian Xue (Penn Biostatistics, the Jin Lab)
Haoxuan Fu (Penn Data Science)
Zhenyi Wu (Purdue Statistics)
Dezheng Ding (Penn Systems Engineering)
Carlos Copana (Purdue Statistics)
Shan Su (Purdue Statistics)
Danyang Wang (Purdue Statistics)
Xieheng Wang (Tsinghua University)
Chun Wang (Purdue Mathematics)
Ethan Yu (2023 Penn Undergraduate Research Mentoring Program)
Yixiao Wang (University of Science and Technology of China)
Lavanya Neti (Wharton Statistics & Finance)
Eleftherios Lazarinis (2024 Penn Undergraduate Research Mentoring Program)
Aravind Krishnan (Vagelos Program in Life Sciences & Management; 2024 Truman Scholar)
William Huang (Penn Computer Science)
Luka Adzic (2026 Penn Undergraduate Research Mentoring Program)
Kate Li (2026 Penn Undergraduate Research Mentoring Program)
Bole Ying (Penn Mathematics)