Leveraging fine-mapping and multipopulation training data to improve cross-population polygenic risk scores

Omer Weissbrod; Masahiro Kanai; Huwenbo Shi; Steven Gazal; Wouter J Peyrot; Amit V Khera; Yukinori Okada; Biobank Japan Project; Alicia R Martin; Hilary K Finucane; Alkes L Price

doi:10.1038/s41588-022-01036-9

Leveraging fine-mapping and multipopulation training data to improve cross-population polygenic risk scores

Nat Genet. 2022 Apr;54(4):450-458. doi: 10.1038/s41588-022-01036-9. Epub 2022 Apr 7.

Authors

Omer Weissbrod^#¹, Masahiro Kanai^#^{2

3}, Huwenbo Shi^#^{4

5}, Steven Gazal^{4

6

7}, Wouter J Peyrot^{4

8}, Amit V Khera^{2

9}, Yukinori Okada^{3

10}; Biobank Japan Project; Alicia R Martin², Hilary K Finucane^{2

11}, Alkes L Price^{12

13}

Collaborators

Affiliations

¹ Epidemiology Department, Harvard School of Public Health, Boston, MA, USA. oweissbrod@hsph.harvard.edu.
² Broad Institute of MIT and Harvard, Cambridge, MA, USA.
³ Department of Statistical Genetics, Osaka University Graduate School of Medicine, Suita, Japan.
⁴ Epidemiology Department, Harvard School of Public Health, Boston, MA, USA.
⁵ OMNI Bioinformatics, San Francisco, CA, USA.
⁶ Department of Population and Public Health Sciences, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA.
⁷ Center for Genetic Epidemiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA.
⁸ Department of Psychiatry, Amsterdam UMC, Vrije Universiteit, Amsterdam, the Netherlands.
⁹ Verve Therapeutics, Cambridge, MA, USA.
¹⁰ Laboratory for Systems Genetics, RIKEN Center for Integrative Medical Sciences, Yokohama, Japan.
¹¹ Department of Medicine, Massachusetts General Hospital, Boston, MA, USA.
¹² Epidemiology Department, Harvard School of Public Health, Boston, MA, USA. aprice@hsph.harvard.edu.
¹³ Broad Institute of MIT and Harvard, Cambridge, MA, USA. aprice@hsph.harvard.edu.

^# Contributed equally.

Abstract

Polygenic risk scores suffer reduced accuracy in non-European populations, exacerbating health disparities. We propose PolyPred, a method that improves cross-population polygenic risk scores by combining two predictors: a new predictor that leverages functionally informed fine-mapping to estimate causal effects (instead of tagging effects), addressing linkage disequilibrium differences, and BOLT-LMM, a published predictor. When a large training sample is available in the non-European target population, we propose PolyPred⁺, which further incorporates the non-European training data. We applied PolyPred to 49 diseases/traits in four UK Biobank populations using UK Biobank British training data, and observed relative improvements versus BOLT-LMM ranging from +7% in south Asians to +32% in Africans, consistent with simulations. We applied PolyPred⁺ to 23 diseases/traits in UK Biobank east Asians using both UK Biobank British and Biobank Japan training data, and observed improvements of +24% versus BOLT-LMM and +12% versus PolyPred. Summary statistics-based analogs of PolyPred and PolyPred⁺ attained similar improvements.

Publication types

Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't

MeSH terms

Genome-Wide Association Study*
Humans
Linkage Disequilibrium
Multifactorial Inheritance* / genetics
Polymorphism, Single Nucleotide / genetics
Risk Factors

Abstract

Publication types

MeSH terms

Grants and funding