In the rapidly evolving landscape of genetic research, the quest to unravel the nuanced interplay between heredity and disease has taken a pivotal turn with a groundbreaking investigation into population-specific polygenic risk scores (PRS) focused on Han Chinese ancestry. This latest study, published in Nature, probes deep into the genetic underpinnings that differ across populations and sheds light on the critical limitations of applying broadly generalized genetic models on diverse ethnic groups. The findings hold profound implications for personalized medicine and the global applicability of genetic risk prediction.
Geneticists have long been aware that the architecture of common complex diseases varies among ethnicities, but quantifying these differences and their impact on disease prediction accuracy has remained a significant challenge. The current research relies on an extensive genome-wide association study (GWAS) conducted on a Han Chinese cohort derived from the Taiwan Precision Medicine Initiative (TPMI). By comparing these results with those from a European population-based GWAS from the UK Biobank (UKB), the study rigorously evaluates the transethnic genetic-effect correlations that govern polygenic traits and diseases.
One of the central breakthroughs revealed in the paper is the heterogeneous nature of genetic correlation across populations for different traits. For complex diseases such as cholelithiasis, an extraordinarily high transethnic genetic-effect correlation (>0.999) was observed, suggesting almost identical genetic determinants between the Han Chinese and European groups. This finding underscores that certain genetically mediated conditions may possess highly conserved causal variants across human populations, offering opportunities for universal predictive genetic markers.
However, the study also exposes contrasting scenarios. For pervasive metabolic diseases like type 2 diabetes and ischaemic heart disease, while still significantly correlated across populations, the genetic-effect correlations were more moderate—0.829 and 0.756 respectively—suggesting substantial but not complete overlap in genetic architecture. These intermediate correlations imply that while some loci contribute similarly to disease risk among different ancestries, others may be population-specific or exert varying effect sizes.
More strikingly, the genetic correlations drop markedly for diseases such as gout and psoriasis. Gout showed a moderate correlation of 0.616, while psoriasis exhibited only a weak correlation of 0.418, pointing toward distinctly differentiated genetic mechanisms. This sharp decline hints not only at divergent allele frequencies and variant effects but also implicates complex gene-environment interactions and evolutionary histories that uniquely shape disease prevalence and manifestation in different ethnic backgrounds.
These findings have practical consequences for the design and utility of polygenic risk scores. PRS models developed predominantly with European-ancestry datasets often underperform or produce biased risk estimates when applied to non-European populations. The demonstrated variability in cross-population genetic effect sizes renders a “one-size-fits-all” approach ineffective, emphasizing the critical need for ancestry-specific genetic data to refine risk prediction algorithms.
Crucially, the study highlights the disease case numbers within each cohort, underscoring how sample size disparities might influence correlation estimates. For example, the gout case count in TPMI was 24,411, considerably larger than the 3,179 cases in UKB, reflecting differential disease burdens and data availability. Psoriasis cases were 4,166 in TPMI and 2,197 in UKB. Such discrepancies further advocate for tailored cohort construction to yield robust and representative genetic insights.
Technologically, the researchers employed advanced statistical methodologies for cross-population genetic-effect correlation assessment, building upon previous frameworks but extending them to capture subtle allele frequency and linkage disequilibrium differences inherent to the distinct biogeographical groups. This rigorous analytical approach ensures the identification of both shared and unique genetic variants implicated in complex diseases across ancestries.
From an evolutionary biology perspective, understanding these transethnic correlations offers glimpses into historic population divergence, selective pressures, and migration patterns that have sculpted the genetic landscape of chronic diseases. It reveals how natural selection and genetic drift could differentially influence variant distributions, modifying disease susceptibilities in various human populations.
Implications for genetic counseling and public health are profound. Incorporating population-specific PRS can lead to more equitable healthcare by providing precise risk stratification for individuals of Han Chinese descent and potentially other underrepresented groups. This can enhance early disease detection, inform preventive strategies, and optimize personalized treatment plans, thereby narrowing health disparities amplified by Eurocentric genomic research biases.
Moreover, the paper champions the systematic expansion of large-scale genomic databases encompassing diverse ancestries, thus urging the scientific community and funding bodies to invest in global collaborations and inclusive recruitment paradigms. Only through such broadened data representation can polygenic risk prediction achieve accuracy and fairness across the world’s heterogeneous populations.
Looking forward, the study paves the way for integrating multi-omic and environmental data layers with population-specific genetic scores. This multi-dimensional approach promises to unravel even more refined predictors of disease risk and progression, pushing the frontier of precision medicine beyond genetic variants alone.
In conclusion, this seminal work by Chen and colleagues powerfully demonstrates that genetic efficacy in disease prediction necessitates acknowledging and incorporating population-specific genetic architectures. Their comprehensive analysis reinforces the scientific mandate to design polygenic risk scoring frameworks that are culturally and genetically inclusive, revolutionizing genomic medicine by transcending ancestral boundaries.
Subject of Research: Population-specific polygenic risk scores and transethnic genetic-effect correlations in Han Chinese versus European ancestries.
Article Title: Population-specific polygenic risk scores for people of Han Chinese ancestry.
Article References:
Chen, HH., Chen, CH., Hou, MC. et al. Population-specific polygenic risk scores for people of Han Chinese ancestry. Nature (2025). https://doi.org/10.1038/s41586-025-09350-y
Image Credits: AI Generated
