Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression

Biao Zeng, Luke R. Lloyd-Jones, Alexander Holloway, Urko M. Marigorta, Andres Metspalu, Grant W. Montgomery, Tonu Esko, Kenneth L. Brigham, Arshed A. Quyyumi, Youssef Idaghdhour, Jian Yang, Peter M. Visscher, Joseph E. Powell, Greg Gibson

Research output: Contribution to journalArticle

Abstract

Expression quantitative trait locus (eQTL) detection has emerged as an important tool for unraveling of the relationship between genetic risk factors and disease or clinical phenotypes. Most studies use single marker linear regression to discover primary signals, followed by sequential conditional modeling to detect secondary genetic variants affecting gene expression. However, this approach assumes that functional variants are sparsely distributed and that close linkage between them has little impact on estimation of their precise location and the magnitude of effects. We describe a series of simulation studies designed to evaluate the impact of linkage disequilibrium (LD) on the fine mapping of causal variants with typical eQTL effect sizes. In the presence of multisite regulation, even though between 80 and 90% of modeled eSNPs associate with normally distributed traits, up to 10% of all secondary signals could be statistical artifacts, and at least 5% but up to one-quarter of credible intervals of SNPs within r2 > 0.8 of the peak may not even include a causal site. The Bayesian methods eCAVIAR and DAP (Deterministic Approximation of Posteriors) provide only modest improvement in resolution. Given the strong empirical evidence that gene expression is commonly regulated by more than one variant, we conclude that the fine mapping of causal variants needs to be adjusted for multisite influences, as conditional estimates can be highly biased by interference among linked sites, but ultimately experimental verification of individual effects is needed. Presumably similar conclusions apply not just to eQTL mapping, but to multisite influences on fine mapping of most types of quantitative trait.

Original languageEnglish (US)
Pages (from-to)2533-2544
Number of pages12
JournalG3: Genes, Genomes, Genetics
Volume7
Issue number8
DOIs
StatePublished - Jan 1 2017

Fingerprint

Quantitative Trait Loci
Gene Expression Regulation
Gene Expression
Bayes Theorem
Linkage Disequilibrium
Artifacts
Single Nucleotide Polymorphism
Linear Models
Phenotype

Keywords

  • Colocalization
  • Fine mapping
  • Gene regulation
  • Linkage disequilibrium
  • Multivariable regression

ASJC Scopus subject areas

  • Molecular Biology
  • Genetics
  • Genetics(clinical)

Cite this

Zeng, B., Lloyd-Jones, L. R., Holloway, A., Marigorta, U. M., Metspalu, A., Montgomery, G. W., ... Gibson, G. (2017). Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression. G3: Genes, Genomes, Genetics, 7(8), 2533-2544. https://doi.org/10.1534/g3.117.043752

Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression. / Zeng, Biao; Lloyd-Jones, Luke R.; Holloway, Alexander; Marigorta, Urko M.; Metspalu, Andres; Montgomery, Grant W.; Esko, Tonu; Brigham, Kenneth L.; Quyyumi, Arshed A.; Idaghdhour, Youssef; Yang, Jian; Visscher, Peter M.; Powell, Joseph E.; Gibson, Greg.

In: G3: Genes, Genomes, Genetics, Vol. 7, No. 8, 01.01.2017, p. 2533-2544.

Research output: Contribution to journalArticle

Zeng, B, Lloyd-Jones, LR, Holloway, A, Marigorta, UM, Metspalu, A, Montgomery, GW, Esko, T, Brigham, KL, Quyyumi, AA, Idaghdhour, Y, Yang, J, Visscher, PM, Powell, JE & Gibson, G 2017, 'Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression', G3: Genes, Genomes, Genetics, vol. 7, no. 8, pp. 2533-2544. https://doi.org/10.1534/g3.117.043752
Zeng B, Lloyd-Jones LR, Holloway A, Marigorta UM, Metspalu A, Montgomery GW et al. Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression. G3: Genes, Genomes, Genetics. 2017 Jan 1;7(8):2533-2544. https://doi.org/10.1534/g3.117.043752
Zeng, Biao ; Lloyd-Jones, Luke R. ; Holloway, Alexander ; Marigorta, Urko M. ; Metspalu, Andres ; Montgomery, Grant W. ; Esko, Tonu ; Brigham, Kenneth L. ; Quyyumi, Arshed A. ; Idaghdhour, Youssef ; Yang, Jian ; Visscher, Peter M. ; Powell, Joseph E. ; Gibson, Greg. / Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression. In: G3: Genes, Genomes, Genetics. 2017 ; Vol. 7, No. 8. pp. 2533-2544.
@article{a9585764ad8a4b1fa399797516615f08,
title = "Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression",
abstract = "Expression quantitative trait locus (eQTL) detection has emerged as an important tool for unraveling of the relationship between genetic risk factors and disease or clinical phenotypes. Most studies use single marker linear regression to discover primary signals, followed by sequential conditional modeling to detect secondary genetic variants affecting gene expression. However, this approach assumes that functional variants are sparsely distributed and that close linkage between them has little impact on estimation of their precise location and the magnitude of effects. We describe a series of simulation studies designed to evaluate the impact of linkage disequilibrium (LD) on the fine mapping of causal variants with typical eQTL effect sizes. In the presence of multisite regulation, even though between 80 and 90{\%} of modeled eSNPs associate with normally distributed traits, up to 10{\%} of all secondary signals could be statistical artifacts, and at least 5{\%} but up to one-quarter of credible intervals of SNPs within r2 > 0.8 of the peak may not even include a causal site. The Bayesian methods eCAVIAR and DAP (Deterministic Approximation of Posteriors) provide only modest improvement in resolution. Given the strong empirical evidence that gene expression is commonly regulated by more than one variant, we conclude that the fine mapping of causal variants needs to be adjusted for multisite influences, as conditional estimates can be highly biased by interference among linked sites, but ultimately experimental verification of individual effects is needed. Presumably similar conclusions apply not just to eQTL mapping, but to multisite influences on fine mapping of most types of quantitative trait.",
keywords = "Colocalization, Fine mapping, Gene regulation, Linkage disequilibrium, Multivariable regression",
author = "Biao Zeng and Lloyd-Jones, {Luke R.} and Alexander Holloway and Marigorta, {Urko M.} and Andres Metspalu and Montgomery, {Grant W.} and Tonu Esko and Brigham, {Kenneth L.} and Quyyumi, {Arshed A.} and Youssef Idaghdhour and Jian Yang and Visscher, {Peter M.} and Powell, {Joseph E.} and Greg Gibson",
year = "2017",
month = "1",
day = "1",
doi = "10.1534/g3.117.043752",
language = "English (US)",
volume = "7",
pages = "2533--2544",
journal = "G3 (Bethesda, Md.)",
issn = "2160-1836",
publisher = "Genetics Society of America",
number = "8",

}

TY - JOUR

T1 - Constraints on eQTL fine mapping in the presence of multisite local regulation of gene expression

AU - Zeng, Biao

AU - Lloyd-Jones, Luke R.

AU - Holloway, Alexander

AU - Marigorta, Urko M.

AU - Metspalu, Andres

AU - Montgomery, Grant W.

AU - Esko, Tonu

AU - Brigham, Kenneth L.

AU - Quyyumi, Arshed A.

AU - Idaghdhour, Youssef

AU - Yang, Jian

AU - Visscher, Peter M.

AU - Powell, Joseph E.

AU - Gibson, Greg

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Expression quantitative trait locus (eQTL) detection has emerged as an important tool for unraveling of the relationship between genetic risk factors and disease or clinical phenotypes. Most studies use single marker linear regression to discover primary signals, followed by sequential conditional modeling to detect secondary genetic variants affecting gene expression. However, this approach assumes that functional variants are sparsely distributed and that close linkage between them has little impact on estimation of their precise location and the magnitude of effects. We describe a series of simulation studies designed to evaluate the impact of linkage disequilibrium (LD) on the fine mapping of causal variants with typical eQTL effect sizes. In the presence of multisite regulation, even though between 80 and 90% of modeled eSNPs associate with normally distributed traits, up to 10% of all secondary signals could be statistical artifacts, and at least 5% but up to one-quarter of credible intervals of SNPs within r2 > 0.8 of the peak may not even include a causal site. The Bayesian methods eCAVIAR and DAP (Deterministic Approximation of Posteriors) provide only modest improvement in resolution. Given the strong empirical evidence that gene expression is commonly regulated by more than one variant, we conclude that the fine mapping of causal variants needs to be adjusted for multisite influences, as conditional estimates can be highly biased by interference among linked sites, but ultimately experimental verification of individual effects is needed. Presumably similar conclusions apply not just to eQTL mapping, but to multisite influences on fine mapping of most types of quantitative trait.

AB - Expression quantitative trait locus (eQTL) detection has emerged as an important tool for unraveling of the relationship between genetic risk factors and disease or clinical phenotypes. Most studies use single marker linear regression to discover primary signals, followed by sequential conditional modeling to detect secondary genetic variants affecting gene expression. However, this approach assumes that functional variants are sparsely distributed and that close linkage between them has little impact on estimation of their precise location and the magnitude of effects. We describe a series of simulation studies designed to evaluate the impact of linkage disequilibrium (LD) on the fine mapping of causal variants with typical eQTL effect sizes. In the presence of multisite regulation, even though between 80 and 90% of modeled eSNPs associate with normally distributed traits, up to 10% of all secondary signals could be statistical artifacts, and at least 5% but up to one-quarter of credible intervals of SNPs within r2 > 0.8 of the peak may not even include a causal site. The Bayesian methods eCAVIAR and DAP (Deterministic Approximation of Posteriors) provide only modest improvement in resolution. Given the strong empirical evidence that gene expression is commonly regulated by more than one variant, we conclude that the fine mapping of causal variants needs to be adjusted for multisite influences, as conditional estimates can be highly biased by interference among linked sites, but ultimately experimental verification of individual effects is needed. Presumably similar conclusions apply not just to eQTL mapping, but to multisite influences on fine mapping of most types of quantitative trait.

KW - Colocalization

KW - Fine mapping

KW - Gene regulation

KW - Linkage disequilibrium

KW - Multivariable regression

UR - http://www.scopus.com/inward/record.url?scp=85021624557&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85021624557&partnerID=8YFLogxK

U2 - 10.1534/g3.117.043752

DO - 10.1534/g3.117.043752

M3 - Article

VL - 7

SP - 2533

EP - 2544

JO - G3 (Bethesda, Md.)

JF - G3 (Bethesda, Md.)

SN - 2160-1836

IS - 8

ER -