Ancestry deconvolution and partial polygenic score can improve susceptibility predictions in recently admixed individuals

Marnetto, Davide; Pärna, Katri; Läll, Kristi; Molinaro, Ludovica; Montinaro, Francesco; Haller, Toomas; Metspalu, Mait; Mägi, Reedik; Fischer, Krista; Pagani, Luca

doi:10.1038/s41467-020-15464-w

Cited by 88 publications

(112 citation statements)

References 36 publications

Supporting

Mentioning

108

Contrasting

Order By: Relevance

“…We also see this linear trend when we restrict the PRS SNPs to those found in regions of the genome inferred to have European ancestry (Figure 1B) . Consistent with previous results 40 , we conclude that the predictive power of the PRS comes largely, but not entirely, from the European ancestry segments of the admixed genomes.…”

Section: Resultssupporting

confidence: 92%

Polygenic scores for height in admixed populations

Bitarello

Mathieson

2020

Preprint

View full text Add to dashboard Cite

11Polygenic risk scores (PRS) use the results of genome-wide association studies (GWAS) to predict quantitative 12 phenotypes or disease risk at an individual level. This provides a potential route to the use of genetic data in 13 personalized medical care. However, a major barrier to the use of PRS is that the majority of GWAS come from 14 cohorts of European ancestry. The predictive power of PRS constructed from these studies is substantially 15 lower in non-European ancestry cohorts, although the reasons for this are unclear. To address this question, 16we investigate the performance of PRS for height in cohorts with admixed African and European ancestry, 17 allowing us to evaluate ancestry-related differences in PRS predictive accuracy while controlling for 18 environment and cohort differences. We first show that that the predictive accuracy of height PRS increases 19 linearly with European ancestry and is largely explained by European ancestry segments of the admixed 20 genomes. We show that differences in allele frequencies, recombination rate, and marginal effect sizes across 21 ancestries all contribute to the decrease in predictive power, but none of these effects explain the decrease on 22 its own. Finally, we demonstrate that prediction for admixed individuals can be improved by using a linear 23 combination of PRS that includes ancestry-specific effect sizes, although this approach is at present limited by 24 the small size of non-European ancestry discovery cohorts. 60we lifted over SNP positions to hg19 using liftOver. For WHI, JHS and HRS, we flipped alleles to the positive 61 strand using the appropriate strand files from https://www.well.ox.ac.uk/~wrayner/strand/. We identified 62 individuals with admixed African ancestry in each cohort using a combination of genetic clustering and self-63 reported ancestry as follows: 64 65 3 UKB: This dataset contains several ancestry groups. We selected 8,813 individuals with African or admixed 66 African and European ancestry based on PCA ( Figure S1) and refer to them as UKB_afr. We further filtered this 67 set to contain individuals with at least 5% of African ancestry, resulting in 8,700 individuals ( Table 1). We 68 randomly selected 9,998 European ancestry individuals from the "White British" subset to use as a comparison 69 sample and refer to them as "UKB_eur". 71WHI: This dataset contains both African American and Hispanic participants. We ran unsupervised 72 ADMIXTURE 27 with k=3 and identified 7,285 individuals with self-reported "African American" ancestry with at 73 most 0.8 of the first ADMIXTURE component (which we interpret as reflecting European ancestry), and at most 74 0.05 of the second (which we interpret as reflecting Native American ancestry; Figure S2). We further filtered 75 this set to contain individuals with at least 5% of African ancestry and height between ±2 standard deviations 76 (sd) from the mean (see Figure S12), resulting in 6,863 individuals ( Table 1). We refer to them as "WHI_afr". 77 78 HRS: This dataset contains multip...

show abstract

Section: Resultssupporting

confidence: 92%

Polygenic scores for height in admixed populations

Bitarello

Mathieson

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…Another method creates ancestry-specific partial PRS for each individual based on the local ancestry of variants selected from a European GWAS 7 . This approach was found to improve trait predictability, compared to a traditional PRS with population specific or European weights, in East Asians for BMI but not height 7 . In contrast, our simulation found that PRS accuracy was higher with African or fixed-effects meta weighting than with local ancestry in African populations.…”

Section: Discussionmentioning

confidence: 96%

“…Potential explanations for the limited portability of European derived PRS across populations includes differences in population allele frequencies and linkage disequilibrium, the presence of population-specific causal variants or effects, or potential differences in gene-gene or geneenvironment interactions 4 . Recent methods developed to improve PRS accuracy in non-Europeans have prioritized the use of European discovered variants and population specific weighting [5][6][7] . However, only small gains in accuracy are possible with limited sample sizes of non-European cohorts 4 .…”

Section: Introductionmentioning

confidence: 99%

Inclusion of Variants Discovered from Diverse Populations Improves Polygenic Risk Score Transferability

Cavazos

Witte

2020

Preprint

View full text Add to dashboard Cite

The majority of polygenic risk scores (PRS) have been developed and optimized in individuals of European ancestry, and may have limited generalizability across other ancestral populations.Understanding aspects of PRS that contribute to this issue and determining solutions is complicated by disease-specific genetic architecture and limited knowledge of trans-ethnic sharing of causal variants and effect sizes. Motivated by these challenges, we undertook a simulation study to assess the relationship between ancestry and the potential bias in PRS developed in European populations. Our simulations show that the magnitude of this bias increases with increasing divergence from European ancestry, and this is attributed to population differences in linkage disequilibrium and allele frequencies of European discovered variants, likely as a result of genetic drift. Importantly, we find that including into the PRS variants discovered in African ancestry individuals has the potential to achieve unbiased estimates of genetic risk across global populations and admixed individuals. We confirm our simulation findings in an analysis of HbA1c in the UK Biobank. Given the demonstrated improvement in PRS prediction accuracy, recruiting larger diverse cohorts will be crucial-and potentially even necessary-for enabling accurate and equitable genetic risk prediction across populations.

show abstract

“…The fact that we find such an abundance of epistatic signals in the African subgroup underscores that African populations, and non-European ancestries in general, are particularly useful for complex trait genetics [66-68, 72, 147-152]. Past research has shown that African ancestry genomes offer a more complete characterization of the the genetic architecture of skin pigmentation [63,64], reveal the evolutionary histories of FOXP2 and other loci [153,154], and are needed for more transferable polygenic risk scores [65,70]. While many studies have generated a call for more GWA studies to be conducted in individuals of non-European ancestry [71,[73][74][75]155], we believe this study reveals that our understanding of the role of epistasis in human complex trait architecture and broad-sense heritability will also expand with multiethnic analyses.…”

Section: Discussionmentioning

confidence: 72%

“…In Nakka et al [62] we showed that enrichment analyses applied to multiple ancestries can identify genes and gene networks contributing to disease risk that ancestry-specific enrichment analyses fail to find. Recent multiethnic GWA studies have also found that using non-European populations offer new insights into additive genetic architecture [63][64][65][66][67][68][69][70]. However, despite this growing body of work and increasing efforts to promote conducting GWA studies in diverse ancestries [68,[71][72][73][74][75], few studies have investigated the role of epistasis in shaping multiethnic human genetic variation (but see [76][77][78][79]).…”

Section: Introductionmentioning

confidence: 99%

Pathway Analysis within Multiple Human Ancestries Reveals Novel Signals for Epistasis in Complex Traits

Turchin

Darnell

Crawford

et al. 2020

Preprint

View full text Add to dashboard Cite

Genome-wide association (GWA) studies have identified thousands of significant genetic associations in humans across a number of complex traits. However, the majority of these studies focus on linear additive relationships between genotypic and phenotypic variation. Epistasis, or non-additive genetic interactions, has been identified as a major driver of both complex trait architecture and evolution in multiple model organisms; yet, this same phenomenon is not considered to be a significant factor underlying human complex traits. There are two possible reasons for this assumption. First, most large GWA studies are conducted solely with European cohorts; therefore, our understanding of broad-sense heritability for many complex traits is limited to just one ancestry group. Second, current epistasis mapping methods commonly identify significant genetic interactions by exhaustively searching across all possible pairs of SNPs. In these frameworks, estimated epistatic effects size are often small and power can be low due to the multiple testing burden. Here, we present a case study that uses a novel region-based mapping approach to analyze sets of variants for the presence of epistatic effects across six diverse subgroups within the UK Biobank. We refer to this method as the “MArginal ePIstasis Test for Regions” or MAPIT-R. Even with limited sample sizes, we find a total of 245 pathways within the KEGG and REACTOME databases that are significantly enriched for epistatic effects in height and body mass index (BMI), with 67% of these pathways being detected within individuals of African ancestry. As a secondary analysis, we introduce a novel region-based “leave-one-out” approach to localize pathway-level epistatic signals to specific interacting genes in BMI. Overall, our results indicate that non-European ancestry populations may be better suited for the discovery of non-additive genetic variation in human complex traits — further underscoring the need for publicly available, biobank-sized datasets of diverse groups of individuals.

show abstract

Ancestry deconvolution and partial polygenic score can improve susceptibility predictions in recently admixed individuals

Cited by 88 publications

References 36 publications

Polygenic scores for height in admixed populations

Polygenic scores for height in admixed populations

Inclusion of Variants Discovered from Diverse Populations Improves Polygenic Risk Score Transferability

Pathway Analysis within Multiple Human Ancestries Reveals Novel Signals for Epistasis in Complex Traits

Contact Info

Product

Resources

About