In:
PLOS Genetics, Public Library of Science (PLoS), Vol. 17, No. 5 ( 2021-5-4), p. e1009021-
Abstract:
The predictive utility of polygenic scores is increasing, and many polygenic scoring methods are available, but it is unclear which method performs best. This study evaluates the predictive utility of polygenic scoring methods within a reference-standardized framework, which uses a common set of variants and reference-based estimates of linkage disequilibrium and allele frequencies to construct scores. Eight polygenic score methods were tested: p-value thresholding and clumping (pT+clump), SBLUP, lassosum, LDpred1, LDpred2, PRScs, DBSLMM and SBayesR, evaluating their performance to predict outcomes in UK Biobank and the Twins Early Development Study (TEDS). Strategies to identify optimal p-value thresholds and shrinkage parameters were compared, including 10-fold cross validation, pseudovalidation and infinitesimal models (with no validation sample), and multi-polygenic score elastic net models. LDpred2, lassosum and PRScs performed strongly using 10-fold cross-validation to identify the most predictive p-value threshold or shrinkage parameter, giving a relative improvement of 16–18% over pT+clump in the correlation between observed and predicted outcome values. Using pseudovalidation, the best methods were PRScs, DBSLMM and SBayesR. PRScs pseudovalidation was only 3% worse than the best polygenic score identified by 10-fold cross validation. Elastic net models containing polygenic scores based on a range of parameters consistently improved prediction over any single polygenic score. Within a reference-standardized framework, the best polygenic prediction was achieved using LDpred2, lassosum and PRScs, modeling multiple polygenic scores derived using multiple parameters. This study will help researchers performing polygenic score studies to select the most powerful and predictive analysis methods.
Type of Medium:
Online Resource
ISSN:
1553-7404
DOI:
10.1371/journal.pgen.1009021
DOI:
10.1371/journal.pgen.1009021.g001
DOI:
10.1371/journal.pgen.1009021.g002
DOI:
10.1371/journal.pgen.1009021.g003
DOI:
10.1371/journal.pgen.1009021.t001
DOI:
10.1371/journal.pgen.1009021.t002
DOI:
10.1371/journal.pgen.1009021.t003
DOI:
10.1371/journal.pgen.1009021.s001
DOI:
10.1371/journal.pgen.1009021.s002
DOI:
10.1371/journal.pgen.1009021.s003
DOI:
10.1371/journal.pgen.1009021.s004
DOI:
10.1371/journal.pgen.1009021.s005
DOI:
10.1371/journal.pgen.1009021.s006
DOI:
10.1371/journal.pgen.1009021.s007
DOI:
10.1371/journal.pgen.1009021.s008
DOI:
10.1371/journal.pgen.1009021.s009
DOI:
10.1371/journal.pgen.1009021.s010
DOI:
10.1371/journal.pgen.1009021.s011
DOI:
10.1371/journal.pgen.1009021.s012
DOI:
10.1371/journal.pgen.1009021.s013
DOI:
10.1371/journal.pgen.1009021.s014
DOI:
10.1371/journal.pgen.1009021.s015
DOI:
10.1371/journal.pgen.1009021.s016
DOI:
10.1371/journal.pgen.1009021.s017
DOI:
10.1371/journal.pgen.1009021.s018
DOI:
10.1371/journal.pgen.1009021.s019
DOI:
10.1371/journal.pgen.1009021.s020
DOI:
10.1371/journal.pgen.1009021.s021
DOI:
10.1371/journal.pgen.1009021.s022
DOI:
10.1371/journal.pgen.1009021.s023
DOI:
10.1371/journal.pgen.1009021.s024
DOI:
10.1371/journal.pgen.1009021.s025
DOI:
10.1371/journal.pgen.1009021.s026
DOI:
10.1371/journal.pgen.1009021.s027
DOI:
10.1371/journal.pgen.1009021.s028
DOI:
10.1371/journal.pgen.1009021.s029
DOI:
10.1371/journal.pgen.1009021.s030
DOI:
10.1371/journal.pgen.1009021.s031
DOI:
10.1371/journal.pgen.1009021.s032
DOI:
10.1371/journal.pgen.1009021.s033
DOI:
10.1371/journal.pgen.1009021.s034
DOI:
10.1371/journal.pgen.1009021.s035
DOI:
10.1371/journal.pgen.1009021.r001
DOI:
10.1371/journal.pgen.1009021.r002
DOI:
10.1371/journal.pgen.1009021.r003
DOI:
10.1371/journal.pgen.1009021.r004
Language:
English
Publisher:
Public Library of Science (PLoS)
Publication Date:
2021
detail.hit.zdb_id:
2186725-2
Bookmarklink