Participation bias in the UK Biobank distorts genetic associations and downstream analyses

Tabea Schoeler; Doug Speed; Eleonora Porcu; Nicola Pirastu; Jean Baptiste Pingault; Zoltán Kutalik

doi:10.1038/s41562-023-01579-9

Participation bias in the UK Biobank distorts genetic associations and downstream analyses

Tabea Schoeler^*, Doug Speed, Eleonora Porcu, Nicola Pirastu, Jean Baptiste Pingault, Zoltán Kutalik^*

^*Corresponding author for this work

Center for Quantitative Genetics and Genomics

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

132 Citations (Scopus)

6 Downloads (Pure)

Abstract

While volunteer-based studies such as the UK Biobank have become the cornerstone of genetic epidemiology, the participating individuals are rarely representative of their target population. To evaluate the impact of selective participation, here we derived UK Biobank participation probabilities on the basis of 14 variables harmonized across the UK Biobank and a representative sample. We then conducted weighted genome-wide association analyses on 19 traits. Comparing the output from weighted genome-wide association analyses (n _effective = 94,643 to 102,215) with that from standard genome-wide association analyses (n = 263,464 to 283,749), we found that increasing representativeness led to changes in SNP effect sizes and identified novel SNP associations for 12 traits. While heritability estimates were less impacted by weighting (maximum change in h ², 5%), we found substantial discrepancies for genetic correlations (maximum change in r _g, 0.31) and Mendelian randomization estimates (maximum change in β _STD, 0.15) for socio-behavioural traits. We urge the field to increase representativeness in biobank samples, especially when studying genetic correlates of behaviour, lifestyles and social outcomes.

Original language	English
Journal	Nature Human Behaviour
Volume	7
Issue	7
Pages (from-to)	1216-1227
Number of pages	12
ISSN	2397-3374
DOIs	https://doi.org/10.1038/s41562-023-01579-9
Publication status	Published - Jul 2023

Keywords

Biological Specimen Banks
Genome-Wide Association Study
Humans
Phenotype
United Kingdom/epidemiology

Access to Document

10.1038/s41562-023-01579-9

s41562-023-01579-9Final published version, 1.95 MBLicence: CC BY

Cite this

@article{c11bf2763ea24f59ba7cd549fcbd1596,

title = "Participation bias in the UK Biobank distorts genetic associations and downstream analyses",

abstract = "While volunteer-based studies such as the UK Biobank have become the cornerstone of genetic epidemiology, the participating individuals are rarely representative of their target population. To evaluate the impact of selective participation, here we derived UK Biobank participation probabilities on the basis of 14 variables harmonized across the UK Biobank and a representative sample. We then conducted weighted genome-wide association analyses on 19 traits. Comparing the output from weighted genome-wide association analyses (n effective = 94,643 to 102,215) with that from standard genome-wide association analyses (n = 263,464 to 283,749), we found that increasing representativeness led to changes in SNP effect sizes and identified novel SNP associations for 12 traits. While heritability estimates were less impacted by weighting (maximum change in h 2, 5%), we found substantial discrepancies for genetic correlations (maximum change in r g, 0.31) and Mendelian randomization estimates (maximum change in β STD, 0.15) for socio-behavioural traits. We urge the field to increase representativeness in biobank samples, especially when studying genetic correlates of behaviour, lifestyles and social outcomes.",

keywords = "Biological Specimen Banks, Genome-Wide Association Study, Humans, Phenotype, United Kingdom/epidemiology",

author = "Tabea Schoeler and Doug Speed and Eleonora Porcu and Nicola Pirastu and Pingault, {Jean Baptiste} and Zolt{\'a}n Kutalik",

note = "Publisher Copyright: {\textcopyright} 2023, The Author(s).",

year = "2023",

month = jul,

doi = "10.1038/s41562-023-01579-9",

language = "English",

volume = "7",

pages = "1216--1227",

journal = "Nature Human Behaviour",

issn = "2397-3374",

publisher = "Nature Publishing Group",

number = "7",

}

TY - JOUR

T1 - Participation bias in the UK Biobank distorts genetic associations and downstream analyses

AU - Schoeler, Tabea

AU - Speed, Doug

AU - Porcu, Eleonora

AU - Pirastu, Nicola

AU - Pingault, Jean Baptiste

AU - Kutalik, Zoltán

PY - 2023/7

Y1 - 2023/7

N2 - While volunteer-based studies such as the UK Biobank have become the cornerstone of genetic epidemiology, the participating individuals are rarely representative of their target population. To evaluate the impact of selective participation, here we derived UK Biobank participation probabilities on the basis of 14 variables harmonized across the UK Biobank and a representative sample. We then conducted weighted genome-wide association analyses on 19 traits. Comparing the output from weighted genome-wide association analyses (n effective = 94,643 to 102,215) with that from standard genome-wide association analyses (n = 263,464 to 283,749), we found that increasing representativeness led to changes in SNP effect sizes and identified novel SNP associations for 12 traits. While heritability estimates were less impacted by weighting (maximum change in h 2, 5%), we found substantial discrepancies for genetic correlations (maximum change in r g, 0.31) and Mendelian randomization estimates (maximum change in β STD, 0.15) for socio-behavioural traits. We urge the field to increase representativeness in biobank samples, especially when studying genetic correlates of behaviour, lifestyles and social outcomes.

AB - While volunteer-based studies such as the UK Biobank have become the cornerstone of genetic epidemiology, the participating individuals are rarely representative of their target population. To evaluate the impact of selective participation, here we derived UK Biobank participation probabilities on the basis of 14 variables harmonized across the UK Biobank and a representative sample. We then conducted weighted genome-wide association analyses on 19 traits. Comparing the output from weighted genome-wide association analyses (n effective = 94,643 to 102,215) with that from standard genome-wide association analyses (n = 263,464 to 283,749), we found that increasing representativeness led to changes in SNP effect sizes and identified novel SNP associations for 12 traits. While heritability estimates were less impacted by weighting (maximum change in h 2, 5%), we found substantial discrepancies for genetic correlations (maximum change in r g, 0.31) and Mendelian randomization estimates (maximum change in β STD, 0.15) for socio-behavioural traits. We urge the field to increase representativeness in biobank samples, especially when studying genetic correlates of behaviour, lifestyles and social outcomes.

KW - Biological Specimen Banks

KW - Genome-Wide Association Study

KW - Humans

KW - Phenotype

KW - United Kingdom/epidemiology

U2 - 10.1038/s41562-023-01579-9

DO - 10.1038/s41562-023-01579-9

M3 - Journal article

C2 - 37106081

AN - SCOPUS:85153782421

SN - 2397-3374

VL - 7

SP - 1216

EP - 1227

JO - Nature Human Behaviour

JF - Nature Human Behaviour

IS - 7

ER -

Participation bias in the UK Biobank distorts genetic associations and downstream analyses

Abstract

Keywords

Access to Document

Fingerprint

Cite this