MSGID: 1:317/3 6274a4d0   
   PID: hpt/lnx 1.9.0-cur 2019-01-08   
   TID: hpt/lnx 1.9.0-cur 2019-01-08   
    New tool more accurately uses genomic data to predict disease risk   
   across diverse populations    
    Integrating data from different ancestries reduces bias in predicting   
   disease risk    
      
     Date:   
         May 5, 2022   
     Source:   
         Massachusetts General Hospital   
     Summary:   
         A modified PRS increases predictive accuracy by integrating data   
         from diverse populations.   
      
      
      
   FULL STORY   
   ==========================================================================   
   Polygenic risk scores (PRS) are promising tools for predicting disease   
   risk, but current versions have built-in bias that can affect their   
   accuracy in some populations and result in health disparities. However,   
   a team of researchers from Massachusetts General Hospital (MGH), the   
   Broad Institute of MIT and Harvard, and Shanghai Jiao Tong University in   
   Shanghai, China, have designed a new method for generating PRS that more   
   accurately predict disease risk across populations, which they report   
   in Nature Genetics.   
      
      
   ==========================================================================   
   Alterations in a gene's DNA sequence can produce a genetic variant   
   that increases the risk for disease. Some genetic variants are closely   
   linked to certain diseases, such as the BRCA1 mutation and breast   
   cancer. "However, most common human diseases -- such as type 2 diabetes,   
   high blood pressure, and depression, for example -- are influenced not   
   by single genes, but by hundreds or thousands of genetic variants across   
   the genome. Each variant contributes a small effect." says Tian Ge,   
   Ph.D., an applied mathematician and biostatistician in the Psychiatric   
   and Neurodevelopmental Genetics Unit, Center for Genomic Medicine at MGH,   
   and co-senior author of the paper. PRS aggregate the effects of genetic   
   variants across the genome and have shown promise for one day being used   
   to predict individual patients' chances of developing diseases. That   
   would allow clinicians to recommend preventive measures and monitor   
   patients closely for early diagnosis and intervention.   
      
   However, a PRS must be "trained" to predict disease risk using data   
   from studies in which genomic information is collected from large   
   groups of individuals. While many disease-causing variants are shared,   
   explains Ge, there are important differences in the genetic basis of   
   a disease between individuals of different ancestries. For example,   
   a common genetic variant that is associated with a specific disease in   
   one population may have a lower frequency or even be missing in other   
   populations. When a genetic variant linked to a disease is shared across   
   different populations, its effect size, or how much it increases risk,   
   may also vary from one ancestral group to another, explains Ge.   
      
   PRS trained using data from one population therefore often have   
   attenuated, or reduced, performance when applied to other populations.   
      
   "A major problem with existing methods for PRS calculation is that, to   
   date, most of the genomic studies used data collected from individuals of   
   European ancestry," says Ge. That creates a Eurocentric bias in existing   
   PRS, he says, producing substantially less-accurate predictions and   
   raising the possibility that they could over- or underestimate disease   
   risk in non-European populations.   
      
   Fortunately, investigators have increased efforts to collect genomic   
   data from underrepresented populations. Leveraging these resources, Ge   
   and his colleagues created a new tool called PRS-CSx that can integrate   
   data from multiple populations and account for genetic similarities and   
   differences between them.   
      
   While there's still significantly more genomic data on individuals of   
   European ancestry, the investigators used computational methods that   
   allowed them to maximize the value of non-European data and improve   
   prediction accuracy in ancestrally diverse individuals.   
      
   In the study, the investigators used genomic data from individuals in   
   several different populations to predict a wide range of physical measures   
   (such as height, body mass index, and blood pressure), blood biomarkers   
   (such as glucose and cholesterol), and the risk for schizophrenia. Then   
   they compared the predicted trait or disease risk with actual measures   
   or reported disease status to measure PRS-CSx's prediction accuracy. The   
   study's results demonstrated that PRS-CSx is significantly more accurate   
   than existing PRS tools in non-European populations.   
      
   "The goal of our work was to narrow the gap between the prediction   
   accuracy in underrepresented populations relative to European individuals,   
   and narrow the gap in health disparities when implementing PRS in   
   clinical settings," says Ge, who notes that the new tool will continue   
   to be refined with the hope that clinicians may one day use it to inform   
   treatment choices and make recommendations about patient care.   
      
   PRS-CSx could also have a role in basic research, says the study's   
   lead author, Yunfeng Ruan, Ph.D., a postdoctoral research fellow at   
   the Broad Institute of MIT and Harvard. It could be used, for example,   
   to explore gene-environment interactions, such as how the effect of   
   genetic risk would depend on the level of environmental risk factors in   
   global populations.   
      
   Even with PRS-CSx, the gap in prediction accuracy between European and   
   non- European populations remains considerable. Broadening the sample   
   diversity across global populations is crucial to further improve the   
   prediction accuracy of PRS in diverse populations. "The expansion of   
   non-European genomic resources, coupled with advanced analytic methods   
   like PRS-CSx, will accelerate the equitable deployment of PRS in clinical   
   settings," says Hailiang Huang, Ph.D., a statistical geneticist in the   
   Analytic and Translational Genetics Unit at MGH and the Stanley Center   
   for Psychiatric Research at the Broad Institute, and co-senior author   
   of the paper.   
      
   Ge is also an assistant professor of Psychiatry at Harvard Medical School   
   (HMS). Huang is an assistant professor of Medicine at HMS.   
      
   This work was supported by the National Institute on Aging, National   
   Human Genome Research Institute, the National Institute of Diabetes   
   and Digestive and Kidney Diseases, the National Institute of Mental   
   Health, the Brain & Behavior Research Foundation, the Zhengxu and Ying   
   He Foundation, and the Stanley Center for Psychiatric Research.   
      
      
   ==========================================================================   
   Story Source: Materials provided by Massachusetts_General_Hospital. Note:   
   Content may be edited for style and length.   
      
      
   ==========================================================================   
   Journal Reference:   
      1. Yunfeng Ruan, Yen-Feng Lin, Yen-Chen Anne Feng, Chia-Yen Chen,   
      Max Lam,   
         Zhenglin Guo, Yong Min Ahn, Kazufumi Akiyama, Makoto Arai, Ji   
         Hyun Baek, Wei J. Chen, Young-Chul Chung, Gang Feng, Kumiko Fujii,   
         Stephen J. Glatt, Kyooseob Ha, Kotaro Hattori, Teruhiko Higuchi,   
         Akitoyo Hishimoto, Kyung Sue Hong, Yasue Horiuchi, Hai-Gwo Hwu,   
         Masashi Ikeda, Sayuri Ishiwata, Masanari Itokawa, Nakao Iwata,   
         Eun-Jeong Joo, Rene S. Kahn, Sung-Wan Kim, Se Joo Kim, Se Hyun Kim,   
         Makoto Kinoshita, Hiroshi Kunugi, Agung Kusumawardhani, Jimmy Lee,   
         Byung Dae Lee, Heon-Jeong Lee, Jianjun Liu, Ruize Liu, Xiancang   
         Ma, Woojae Myung, Shusuke Numata, Tetsuro Ohmori, Ikuo Otsuka,   
         Yuji Ozeki, Sibylle G. Schwab, Wenzhao Shi, Kazutaka Shimoda, Kang   
         Sim, Ichiro Sora, Jinsong Tang, Tomoko Toyota, Ming Tsuang, Dieter   
         B. Wildenauer, Hong-Hee Won, Takeo Yoshikawa, Alice Zheng, Feng Zhu,   
         Lin He, Akira Sawa, Alicia R. Martin, Shengying Qin, Hailiang Huang,   
         Tian Ge. Improving polygenic prediction in ancestrally diverse   
         populations. Nature Genetics, 2022; DOI: 10.1038/s41588-022-01054-7   
   ==========================================================================   
      
   Link to news story:   
   https://www.sciencedaily.com/releases/2022/05/220505143814.htm   
      
   --- up 9 weeks, 3 days, 10 hours, 50 minutes   
    * Origin: -=> Castle Rock BBS <=- Now Husky HPT Powered! (1:317/3)   
   SEEN-BY: 15/0 106/201 114/705 123/120 129/330 331 153/7715 218/700   
   SEEN-BY: 229/110 111 317 400 426 428 470 664 700 292/854 298/25 305/3   
   SEEN-BY: 317/3 320/219 396/45   
   PATH: 317/3 229/426