Dr. Xihao Li (Assistant Professor) et al. published a paper in Nature Computational Science.
Dr. Xihao Li (Assistant Professor of Genetics & Biostatistics), together with lead authors Yohhan Kumarasinghe (UNC Biostatistics), Dr. Jacob Williams (NCI Division of Cancer Epidemiology & Genetics) and collaborators, published a paper in Nature Computational Science titled “MetaSTAARlite: an all-in-one tool for biobank-scale whole-genome sequencing meta-analysis.” The paper introduces MetaSTAARlite, an open-source, scalable, and resource-efficient pipeline for functionally informed rare variant meta-analysis across multiple biobank-scale whole-genome and whole-exome sequencing studies. By using summary statistics and sparse matrix representations, MetaSTAARlite enables coding, noncoding, ncRNA, and custom meta-analyses without requiring individual-level data sharing across biobanks. In applications to theUK Biobank and All of Us data, the authors demonstrate that MetaSTAARlite preserves the power of pooled analyses while substantially reducing computation, memory, and storage requirements, making cross-biobank rare variant discovery more accessible and privacy-preserving.