Lin Xu, Ph.D.

With a Ph.D. in Biostatistics and Bioinformatics from Cornell University, I have developed expertise in the integrated analysis of high-dimensional datasets, deep learning, Bayesian modeling, and bioinformatics algorithms, as well as designing clinical studies for both the analytic and clinical validation of biomarkers. During my postdoctoral training, I expanded my expertise by bridging biostatistical methodologies with clinical applications, particularly in translating biomarker discoveries into validated clinical tools. Currently, as a tenure-track Assistant Professor at UT Southwestern, my research centers on developing and applying deep learning, Bayesian statistical models, and bioinformatics tools to identify and validate robust disease-specific biomarkers and disease genes, uncover new therapeutic targets and treatment strategies, and build software platforms and web portals for accessible data analysis and sharing. 

Since establishing my laboratory at UT Southwestern in 2022, I have published 94 peer-reviewed research articles, with 21 as co- or corresponding author. My lab has published a series of deep learning and statistics algorithms to identify clinical biomarkers and disease genes from bulk, single-cell, and spatially resolved omics data, advancing biological and translational insights across cancer, heart, brain, and muscle biology. Among these publications, I served as co-/corresponding authors to develop iExCN 2.0 algorithm for bulk multiomics data integration (Xu et al. Nature Communications, 2023; impact factor (IF) 14.7), ilMPACT algorithm for spatial transcriptomics analysis (Jiang et al. Genome Biology, 2024; IF: 17.9), EpiTR framework for transcriptional factor identification and analysis (Lu et al. Briefings in Bioinformatics, 2024, IF: 14.0), BayeSMART algorithm for multi-sample spatial omics analysis (Guo et al. Briefings in Bioinformatics, 2024),  COSMOS algorithm for spatially resolved multi-omics data integration (Zhou et al. Nature Communications, 2025), SpaSNE algorithm for spatially guided dimension reduction (Zhou et al. GigaScience, 2025; IF: 11.8), and BIT algorithm for epigenomics-based transcriptional factor analysis (Lu et al. Nature Communications, 2025). These algorithms and computational frameworks have been applied in our recent studies published in Nature (Huang et al. 2020, and Engreitz JM, et al. 2025), Cell (Wei et al. 2023), Science (Lebak et al. 2023), Nature Medicine (Cai et al. 2023), Nature Machine Intelligence (Lu et al. 2022), Nature Immunology (Yang et al. 2025), Circulation (Lebak et al. 2023, Bann et al. 2025, and Ding et al. 2025,  IF: 38.6), Cancer Cell (Shi et al. 2022, and Zhu et al. 2023; IF: 44.5), and many others.

More importantly, my laboratory pioneers the application of artificial intelligence (AI) —particularly foundation models and large language models (LLMs)—to drive biological and medical discovery. Our work focuses on developing interpretable and generalizable AI frameworks that extend the capabilities of foundation models beyond text and vision into genomics and precision medicine. As a major milestone, one of our recently developed AI algorithms for clinical diagnosis prediction was invited for an oral presentation at the Association for the Advancement of Artificial Intelligence (AAAI) Conference —one of the oldest and most impactful international conferences in the artificial intelligence field. Founded in 1980, the AAAI Conference is widely recognized as a flagship venue for foundational and cutting-edge advances across the entire spectrum of artificial intelligence—spanning machine learning, reasoning, robotics, and AI for science—and is ranked among the top three AI conferences globally by Google Scholar Metrics. Among fewer than 40 biology-oriented AI studies featured annually, this recognition underscores the novelty and cross-disciplinary impact of my research and reflects my research team’s commitment to bridging AI and biomedical science.

Since I joined UT Southwestern as a tenure-track Assistant Professor, my lab has developed multiple deep learning models and statistics algorithms to identify new disease genes and therapeutic targets [e.g., Huang et al. Nature (2020), Lu et al. Nature Machine Intelligence (2022), Chai et al. Nature Medicine (2022), Takahiko et al. Science Translational Medicine (2022) (Cover Story), Shi et al. Cancer Cell (2022), Lebek et al. Science (2023), Xu et al. Nature Communications (2023), Zhu et al. Cancer Cell (2023), Zhang et al. Journal of Clinical Investigation (2023), He et al. Nature Communications (2023), Yu et al. Nature Communications (2023), Santos et al. Nature Communications (2023), Li et al. Nature Communications (2023), Wei et al. Cell (2023), Hernandez et al. PNAS (2024), Hu L et al. Mol Cell (2024), Eglenen-Polat et al. Nature Communications (2024), Lin CC et al. Nature Communications (2024), Xu L. et al. Nature Communications (2024), Venkateswaran N. et al. Nature Communications (2024), Chen S. et al. Nature Communications (2025), Liao C, Nature Communications (2025), Yang X. et al. Nature Immunology (2025), Lu ZY. et al. Nature Communications (2025), Zhou Y. et al. Nature Communications (2025), Adachi Y et al. J Clin Invest. (2025), Napolitano F, et al. J Clin Invest. (2025), Bann et al. Circulation (2025), Ding et al. Circulation (2025), Caravia XM, et al. Proc. Natl. Acad. Sci. (2025), Guan P, et al. Proc. Natl. Acad. Sci. (2025), Engreitz JM, et al. Nature (2025)]. Since 2021, my lab’s research has been published in a series of high-impact journals, including Nature, Science, Cell, Nature Medicine, Cancer Cell,  Elife, PNAS, Nature Machine Intelligence, Nature Immunology, Nature Metabolism, Nature Communications, Science Translational Medicine, Science Advance, Molecular Cell, Genes & Development, Cell Reports, Journal of Clinical Investigation, Circulation, Circulation Research, and Developmental Cell.