Fast and powerful genome wide association of dense genetic data with high dimensional imaging phenotypes

Habib Ganjgahi, Anderson M. Winkler, David C. Glahn, John Blangero, Brian Donohue, Peter Kochunov, Thomas E. Nichols

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Genome wide association (GWA) analysis of brain imaging phenotypes can advance our understanding of the genetic basis of normal and disorder-related variation in the brain. GWA approaches typically use linear mixed effect models to account for non-independence amongst subjects due to factors, such as family relatedness and population structure. The use of these models with high-dimensional imaging phenotypes presents enormous challenges in terms of computational intensity and the need to account multiple testing in both the imaging and genetic domain. Here we present a method that makes mixed models practical with high-dimensional traits by a combination of a transformation applied to the data and model, and the use of a non-iterative variance component estimator. With such speed enhancements permutation tests are feasible, which allows inference on powerful spatial tests like the cluster size statistic.

Original languageEnglish (US)
Article number3254
JournalNature Communications
Volume9
Issue number1
DOIs
StatePublished - Dec 1 2018
Externally publishedYes

Fingerprint

phenotype
genome
Genes
Genome
Phenotype
Imaging techniques
Genome-Wide Association Study
Neuroimaging
brain
Brain
permutations
inference
estimators
Population
Statistics
statistics
disorders
augmentation
Testing

ASJC Scopus subject areas

  • Chemistry(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Physics and Astronomy(all)

Cite this

Ganjgahi, H., Winkler, A. M., Glahn, D. C., Blangero, J., Donohue, B., Kochunov, P., & Nichols, T. E. (2018). Fast and powerful genome wide association of dense genetic data with high dimensional imaging phenotypes. Nature Communications, 9(1), [3254]. https://doi.org/10.1038/s41467-018-05444-6

Fast and powerful genome wide association of dense genetic data with high dimensional imaging phenotypes. / Ganjgahi, Habib; Winkler, Anderson M.; Glahn, David C.; Blangero, John; Donohue, Brian; Kochunov, Peter; Nichols, Thomas E.

In: Nature Communications, Vol. 9, No. 1, 3254, 01.12.2018.

Research output: Contribution to journalArticle

Ganjgahi, H, Winkler, AM, Glahn, DC, Blangero, J, Donohue, B, Kochunov, P & Nichols, TE 2018, 'Fast and powerful genome wide association of dense genetic data with high dimensional imaging phenotypes', Nature Communications, vol. 9, no. 1, 3254. https://doi.org/10.1038/s41467-018-05444-6
Ganjgahi, Habib ; Winkler, Anderson M. ; Glahn, David C. ; Blangero, John ; Donohue, Brian ; Kochunov, Peter ; Nichols, Thomas E. / Fast and powerful genome wide association of dense genetic data with high dimensional imaging phenotypes. In: Nature Communications. 2018 ; Vol. 9, No. 1.
@article{48acf5837d3a4ac8a3fe674204056d5e,
title = "Fast and powerful genome wide association of dense genetic data with high dimensional imaging phenotypes",
abstract = "Genome wide association (GWA) analysis of brain imaging phenotypes can advance our understanding of the genetic basis of normal and disorder-related variation in the brain. GWA approaches typically use linear mixed effect models to account for non-independence amongst subjects due to factors, such as family relatedness and population structure. The use of these models with high-dimensional imaging phenotypes presents enormous challenges in terms of computational intensity and the need to account multiple testing in both the imaging and genetic domain. Here we present a method that makes mixed models practical with high-dimensional traits by a combination of a transformation applied to the data and model, and the use of a non-iterative variance component estimator. With such speed enhancements permutation tests are feasible, which allows inference on powerful spatial tests like the cluster size statistic.",
author = "Habib Ganjgahi and Winkler, {Anderson M.} and Glahn, {David C.} and John Blangero and Brian Donohue and Peter Kochunov and Nichols, {Thomas E.}",
year = "2018",
month = "12",
day = "1",
doi = "10.1038/s41467-018-05444-6",
language = "English (US)",
volume = "9",
journal = "Nature Communications",
issn = "2041-1723",
publisher = "Nature Publishing Group",
number = "1",

}

TY - JOUR

T1 - Fast and powerful genome wide association of dense genetic data with high dimensional imaging phenotypes

AU - Ganjgahi, Habib

AU - Winkler, Anderson M.

AU - Glahn, David C.

AU - Blangero, John

AU - Donohue, Brian

AU - Kochunov, Peter

AU - Nichols, Thomas E.

PY - 2018/12/1

Y1 - 2018/12/1

N2 - Genome wide association (GWA) analysis of brain imaging phenotypes can advance our understanding of the genetic basis of normal and disorder-related variation in the brain. GWA approaches typically use linear mixed effect models to account for non-independence amongst subjects due to factors, such as family relatedness and population structure. The use of these models with high-dimensional imaging phenotypes presents enormous challenges in terms of computational intensity and the need to account multiple testing in both the imaging and genetic domain. Here we present a method that makes mixed models practical with high-dimensional traits by a combination of a transformation applied to the data and model, and the use of a non-iterative variance component estimator. With such speed enhancements permutation tests are feasible, which allows inference on powerful spatial tests like the cluster size statistic.

AB - Genome wide association (GWA) analysis of brain imaging phenotypes can advance our understanding of the genetic basis of normal and disorder-related variation in the brain. GWA approaches typically use linear mixed effect models to account for non-independence amongst subjects due to factors, such as family relatedness and population structure. The use of these models with high-dimensional imaging phenotypes presents enormous challenges in terms of computational intensity and the need to account multiple testing in both the imaging and genetic domain. Here we present a method that makes mixed models practical with high-dimensional traits by a combination of a transformation applied to the data and model, and the use of a non-iterative variance component estimator. With such speed enhancements permutation tests are feasible, which allows inference on powerful spatial tests like the cluster size statistic.

UR - http://www.scopus.com/inward/record.url?scp=85051673084&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85051673084&partnerID=8YFLogxK

U2 - 10.1038/s41467-018-05444-6

DO - 10.1038/s41467-018-05444-6

M3 - Article

VL - 9

JO - Nature Communications

JF - Nature Communications

SN - 2041-1723

IS - 1

M1 - 3254

ER -