Taiwan builds one of the world’s largest precision medicine cohorts

Taipei, Taiwan — [October 22, 2025] — Researchers at Academia Sinica and partner hospitals have established the Taiwan Precision Medicine Initiative (TPMI), a landmark national project that has recruited more than 560,000 participants to create one of the world’s largest and most comprehensive biobank resources for precision medicine. The study, published in Nature, details how this initiative integrates genetic, electronic medical record (EMR), and lifestyle data to improve disease prediction, prevention, and individualized treatment for the Han Chinese population, which represents nearly 20% of the global population.

A national collaboration for precision health

Launched by Academia Sinica in collaboration with 16 major medical centers across Taiwan, the TPMI has built a unified platform combining genetic profiles with long-term EMR data. Participants consented to provide DNA samples, share retrospective and prospective medical records, and receive personalized feedback on genetic risk and health management. “The TPMI represents a major step toward population-based precision health in Asia,” said Dr. Jer-Yuarn Wu, senior corresponding author of the study at the Institute of Medical Sciences, Academia Sinica. “By linking genomic data with clinical outcomes across hospitals, we can transform how diseases are detected and managed.”

Unprecedented scale and data depth

The TPMI cohort includes 565,000 participants with both genotype and EMR data, representing a diverse cross-section of Taiwan’s population. The initiative uses two custom genotyping arrays (TPMv1 and TPMv2) optimized for Han Chinese genetic variation, achieving high accuracy in identifying over 388,000 novel variants not found in other Asian biobanks. EMR data — standardized from 16 hospitals — include laboratory results, diagnoses, imaging, pathology, and clinical notes, enabling longitudinal analyses spanning over a decade for many participants.

Key discoveries and population insights

Genetic analyses revealed fine-scale population structure within Taiwan, including admixture between Han Chinese, Hakka, Minnan, and Indigenous groups, as well as the identification of novel disease-associated variants unique to Taiwanese subpopulations. Large-scale genome-wide association studies (GWAS) using TPMI data replicated known findings for common conditions such as type 2 diabetes, hypertension, and lipid disorders, while uncovering dozens of novel risk loci relevant to Asian populations. Furthermore, TPMI-developed polygenic risk score (PRS) models for 265 diseases and 24 quantitative traits demonstrated strong predictive performance, achieving AUC values up to 0.86 when combined with demographic factors.

Empowering future health innovation

The TPMI data platform, known as the TPMI Data Access Platform (TDAP), securely integrates genetic and clinical data while maintaining participant privacy through advanced de-identification pipelines. Researchers worldwide can apply for access to the data for approved studies, creating new opportunities for collaborative discovery. “The TPMI’s scale and data quality make it a global resource,” said Dr. Hsin-Chou Yang, lead author and Director of the Institute of Statistical Science at Academia Sinica. “It will not only accelerate disease research but also support real-world trials of genetic risk–guided healthcare.”

“The Taiwan Precision Medicine Initiative demonstrates how large-scale, well-coordinated efforts can transform both science and healthcare,” said Dr. Pui-Yan Kwok, co-corresponding author, Academician of Academia Sinica and Professor of Dermatology at the University of California, San Francisco. “By integrating diverse genomic and clinical data, Taiwan is setting a global benchmark for population-specific resources that accelerate discovery and improve human health.”