Twitxy Saved

Xihong Lin

@XihongLin

Professor of @HarvardBiostats & @HarvardStats, (Bio)statistics,Data Science,Genetics Genomics,Epidemiology,Health, Education,COVID19 analysis,Views are all mine
Boston, MA hsph.harvard.edu/xihong-lin/ Joined August 2017
425 Following 7.8K Followers 2.1K Posts

Latest posts 5

Xihong Lin @XihongLin
The Harvard Statistics dept & Applied Math area seek applicants for a joint tenured professor position in AI for Science. We seek outstanding candidates to advance principled AI grounded in math and stat theory to accelerate scientific discovery.
academicpositions.harvard.edu/postings/16848
academicpositions.harvard.edu Tenured Professor in Applied Mathematics and Statistics The Applied Mathematics area in the John A. Paulson School of Engineering and Applied Sciences (SEAS) and the Department of Statistics in the Faculty of Arts and Sciences (FAS) at Harvard University...
4 replies 11 reposts 44 likes 6.6K views View on X
Xihong Lin @XihongLin
2026 Harvard PQG Conference: Unleashing the Power of Integrated Multimodal Genomic and Health Data, Oct 29-30, 2026
Session 1: Multi-modal Genomics Data
Session 2: Multi-modal Health Data
Session 3: Multimodal AI Systems: Scalable & Trustworthy Integration
hsph.harvard.edu/research/quant…
hsph.harvard.edu Conference of the Program in Quantitative Genomics | Program in Quantitative Genomics | Harvard... The Program in Quantitative Genomics hosts an annual PQG Conference on cutting edge topics in genomics.
1 replies 8 reposts 25 likes 2.3K views View on X
Xihong Lin @XihongLin
Our review article on harnessing synthetic data from generative AI for statistical inference. We discuss generative models for synthetic data & their principled use for valid downstream statistical inference, esp when generative models are misspecified.
arxiv.org/abs/2603.05396.
arxiv.org Harnessing Synthetic Data from Generative AI for Statistical Inference The emergence of generative AI models has dramatically expanded the availability and use of synthetic data across scientific, industrial, and policy domains. While these developments open new...
1 replies 11 reposts 74 likes 8.5K views View on X
Xihong Lin @XihongLin
GATE-STAAR: Rare variant association tests for survival analysis of time-to-event phenotypes, e.g., age at diagnosis in the UK biobank. It fits frailty models by incorporating variant functional annotations and accouning for relatedness, and ancestry PCs
pnas.org/doi/abs/10.107…
www.pnas.org Scalable and accurate rare-variant association tests for whole genome sequencing time-to-event... Whole genome sequencing (WGS) studies in large biobanks provide an unprecedented opportunity to study the rare-variant (RV) effects on the natural ...