# Drivers of genetic heterogeneity and their impact on polygenic risk prediction

> **NIH NIH F31** · UNIVERSITY OF PENNSYLVANIA · 2021 · $46,036

## Abstract

Abstract
Genome-wide association studies (GWAS) have informed us about the genetic architecture of complex traits
and diseases by identifying common genetic risk variants. Typically many genetic variants conferring small
increases in risk underlie disease variation in a population. Summarizing the effects of these genetic risk variants
can measure an individual’s genetic predisposition, which is done by calculating a polygenic risk score (PRS).
PRS show clinical promise in identifying individuals at the extremes of the genetic risk distribution. However, the
potential benefits of using PRS in risk stratification may be curtailed due to loss of predictive power of PRS in
groups of non-European ancestry. PRS are commonly calculated from GWAS conducted in European
populations because of large sample sizes. PRS derived from European GWAS suffer the greatest loss in
predictive power in African populations. Currently we do not know the reasons for this loss of predictive power.
Investigating the factors limiting the transferability of PRS will inform future avenues of research into developing
more generalizable PRS. Since few GWAS have been conducted in African ancestry individuals, the proposal
will perform a GWAS in Sub-Saharan African populations for height in Aim 1 to contribute to our knowledge of
the genetic architecture of a complex trait in diverse African populations. Through the GWAS in Aim 1, genetic
variation not present in non-African populations that affects height can be uncovered. In Aim 2, the proposal will
investigate the role of differences in linkage disequilibrium (LD) structure across populations in limiting the
transferability of PRS. GWAS rely on LD between a genotyped marker and the causal variant, so the variants
included in a PRS are not necessarily causal. To test whether differences in the tagging of shared causal variants
between populations limit the transferability of PRS, this proposal will identify instances of differential tagging of
causal variants between Europeans and Africans and cases where multiple associations at a locus are obscured
by high LD in Europeans, which can be interrogated with low LD in Africans. In Aim 3, this proposal will study
the mechanisms driving differences in effect size of variants shared across ancestries, specifically the role of
gene by gene (GxG) and gene by environment (GxE) interactions. Variants with the most disparate effect sizes
between Europeans and Africans will be examined for evidence of interaction effects. GxG interactions for a
candidate variant will be tested based on its local genomic ancestry. GxE interactions will be tested using two
groups of similar ancestry but differing by status of living in a rural or urban environment. In all, these analyses
will inform our understanding of the roles of LD structure and GxG and GxE interactions in limiting the
transferability of PRS across populations.

## Key facts

- **NIH application ID:** 10314395
- **Project number:** 1F31HG011813-01A1
- **Recipient organization:** UNIVERSITY OF PENNSYLVANIA
- **Principal Investigator:** Dan Ju
- **Activity code:** F31 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2021
- **Award amount:** $46,036
- **Award type:** 1
- **Project period:** 2021-08-01 → 2022-07-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10314395

## Citation

> US National Institutes of Health, RePORTER application 10314395, Drivers of genetic heterogeneity and their impact on polygenic risk prediction (1F31HG011813-01A1). Retrieved via AI Analytics 2026-05-25 from https://api.ai-analytics.org/grant/nih/10314395. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*