Exome Sequencing of Phenotypic Extremes Identifies CAV2 and TMC6 as Interacting Modifiers of Chronic Pseudomonas aeruginosa Infection in Cystic Fibrosis

Mary J. Emond, Tin Louie, Julia Emerson, Jessica X. Chong, Rasika Mathias, Michael R. Knowles, Mark J. Rieder, Holly K. Tabor, Debbie A. Nickerson, Kathleen C. Barnes, Ronald L. Gibson, Michael J. Bamshad

Research output: Contribution to journalArticle

Abstract

Discovery of rare or low frequency variants in exome or genome data that are associated with complex traits often will require use of very large sample sizes to achieve adequate statistical power. For a fixed sample size, sequencing of individuals sampled from the tails of a phenotype distribution (i.e., extreme phenotypes design) maximizes power and this approach was recently validated empirically with the discovery of variants in DCTN4 that influence the natural history of P. aeruginosa airway infection in persons with cystic fibrosis (CF; MIM219700). The increasing availability of large exome/genome sequence datasets that serve as proxies for population-based controls affords the opportunity to test an alternative, potentially more powerful and generalizable strategy, in which the frequency of rare variants in a single extreme phenotypic group is compared to a control group (i.e., extreme phenotype vs. control population design). As proof-of-principle, we applied this approach to search for variants associated with risk for age-of-onset of chronic P. aeruginosa airway infection among individuals with CF and identified variants in CAV2 and TMC6 that were significantly associated with group status. These results were validated using a large, prospective, longitudinal CF cohort and confirmed a significant association of a variant in CAV2 with increased age-of-onset of P. aeruginosa airway infection (hazard ratio = 0.48, 95% CI=[0.32, 0.88]) and variants in TMC6 with diminished age-of-onset of P. aeruginosa airway infection (HR = 5.4, 95% CI=[2.2, 13.5]) A strong interaction between CAV2 and TMC6 variants was observed (HR=12.1, 95% CI=[3.8, 39]) for children with the deleterious TMC6 variant and without the CAV2 protective variant. Neither gene showed a significant association using an extreme phenotypes design, and conditions for which the power of an extreme phenotype vs. control population design was greater than that for the extreme phenotypes design were explored.

Original languageEnglish (US)
Article numbere1005273
JournalPLoS Genetics
Volume11
Issue number6
DOIs
StatePublished - Jul 1 2015

Fingerprint

Exome
Pseudomonas Infections
modifiers (genes)
cystic fibrosis
Pseudomonas aeruginosa
Cystic Fibrosis
phenotype
Phenotype
Age of Onset
infection
Infection
Sample Size
genome
Genome
Population Control
Proxy
Natural History
natural history
Population
tail

ASJC Scopus subject areas

  • Genetics
  • Molecular Biology
  • Ecology, Evolution, Behavior and Systematics
  • Cancer Research
  • Genetics(clinical)

Cite this

Exome Sequencing of Phenotypic Extremes Identifies CAV2 and TMC6 as Interacting Modifiers of Chronic Pseudomonas aeruginosa Infection in Cystic Fibrosis. / Emond, Mary J.; Louie, Tin; Emerson, Julia; Chong, Jessica X.; Mathias, Rasika; Knowles, Michael R.; Rieder, Mark J.; Tabor, Holly K.; Nickerson, Debbie A.; Barnes, Kathleen C.; Gibson, Ronald L.; Bamshad, Michael J.

In: PLoS Genetics, Vol. 11, No. 6, e1005273, 01.07.2015.

Research output: Contribution to journalArticle

Emond, MJ, Louie, T, Emerson, J, Chong, JX, Mathias, R, Knowles, MR, Rieder, MJ, Tabor, HK, Nickerson, DA, Barnes, KC, Gibson, RL & Bamshad, MJ 2015, 'Exome Sequencing of Phenotypic Extremes Identifies CAV2 and TMC6 as Interacting Modifiers of Chronic Pseudomonas aeruginosa Infection in Cystic Fibrosis', PLoS Genetics, vol. 11, no. 6, e1005273. https://doi.org/10.1371/journal.pgen.1005273
Emond, Mary J. ; Louie, Tin ; Emerson, Julia ; Chong, Jessica X. ; Mathias, Rasika ; Knowles, Michael R. ; Rieder, Mark J. ; Tabor, Holly K. ; Nickerson, Debbie A. ; Barnes, Kathleen C. ; Gibson, Ronald L. ; Bamshad, Michael J. / Exome Sequencing of Phenotypic Extremes Identifies CAV2 and TMC6 as Interacting Modifiers of Chronic Pseudomonas aeruginosa Infection in Cystic Fibrosis. In: PLoS Genetics. 2015 ; Vol. 11, No. 6.
@article{c390dca7391d428b879619beb255a43a,
title = "Exome Sequencing of Phenotypic Extremes Identifies CAV2 and TMC6 as Interacting Modifiers of Chronic Pseudomonas aeruginosa Infection in Cystic Fibrosis",
abstract = "Discovery of rare or low frequency variants in exome or genome data that are associated with complex traits often will require use of very large sample sizes to achieve adequate statistical power. For a fixed sample size, sequencing of individuals sampled from the tails of a phenotype distribution (i.e., extreme phenotypes design) maximizes power and this approach was recently validated empirically with the discovery of variants in DCTN4 that influence the natural history of P. aeruginosa airway infection in persons with cystic fibrosis (CF; MIM219700). The increasing availability of large exome/genome sequence datasets that serve as proxies for population-based controls affords the opportunity to test an alternative, potentially more powerful and generalizable strategy, in which the frequency of rare variants in a single extreme phenotypic group is compared to a control group (i.e., extreme phenotype vs. control population design). As proof-of-principle, we applied this approach to search for variants associated with risk for age-of-onset of chronic P. aeruginosa airway infection among individuals with CF and identified variants in CAV2 and TMC6 that were significantly associated with group status. These results were validated using a large, prospective, longitudinal CF cohort and confirmed a significant association of a variant in CAV2 with increased age-of-onset of P. aeruginosa airway infection (hazard ratio = 0.48, 95{\%} CI=[0.32, 0.88]) and variants in TMC6 with diminished age-of-onset of P. aeruginosa airway infection (HR = 5.4, 95{\%} CI=[2.2, 13.5]) A strong interaction between CAV2 and TMC6 variants was observed (HR=12.1, 95{\%} CI=[3.8, 39]) for children with the deleterious TMC6 variant and without the CAV2 protective variant. Neither gene showed a significant association using an extreme phenotypes design, and conditions for which the power of an extreme phenotype vs. control population design was greater than that for the extreme phenotypes design were explored.",
author = "Emond, {Mary J.} and Tin Louie and Julia Emerson and Chong, {Jessica X.} and Rasika Mathias and Knowles, {Michael R.} and Rieder, {Mark J.} and Tabor, {Holly K.} and Nickerson, {Debbie A.} and Barnes, {Kathleen C.} and Gibson, {Ronald L.} and Bamshad, {Michael J.}",
year = "2015",
month = "7",
day = "1",
doi = "10.1371/journal.pgen.1005273",
language = "English (US)",
volume = "11",
journal = "PLoS Genetics",
issn = "1553-7390",
publisher = "Public Library of Science",
number = "6",

}

TY - JOUR

T1 - Exome Sequencing of Phenotypic Extremes Identifies CAV2 and TMC6 as Interacting Modifiers of Chronic Pseudomonas aeruginosa Infection in Cystic Fibrosis

AU - Emond, Mary J.

AU - Louie, Tin

AU - Emerson, Julia

AU - Chong, Jessica X.

AU - Mathias, Rasika

AU - Knowles, Michael R.

AU - Rieder, Mark J.

AU - Tabor, Holly K.

AU - Nickerson, Debbie A.

AU - Barnes, Kathleen C.

AU - Gibson, Ronald L.

AU - Bamshad, Michael J.

PY - 2015/7/1

Y1 - 2015/7/1

N2 - Discovery of rare or low frequency variants in exome or genome data that are associated with complex traits often will require use of very large sample sizes to achieve adequate statistical power. For a fixed sample size, sequencing of individuals sampled from the tails of a phenotype distribution (i.e., extreme phenotypes design) maximizes power and this approach was recently validated empirically with the discovery of variants in DCTN4 that influence the natural history of P. aeruginosa airway infection in persons with cystic fibrosis (CF; MIM219700). The increasing availability of large exome/genome sequence datasets that serve as proxies for population-based controls affords the opportunity to test an alternative, potentially more powerful and generalizable strategy, in which the frequency of rare variants in a single extreme phenotypic group is compared to a control group (i.e., extreme phenotype vs. control population design). As proof-of-principle, we applied this approach to search for variants associated with risk for age-of-onset of chronic P. aeruginosa airway infection among individuals with CF and identified variants in CAV2 and TMC6 that were significantly associated with group status. These results were validated using a large, prospective, longitudinal CF cohort and confirmed a significant association of a variant in CAV2 with increased age-of-onset of P. aeruginosa airway infection (hazard ratio = 0.48, 95% CI=[0.32, 0.88]) and variants in TMC6 with diminished age-of-onset of P. aeruginosa airway infection (HR = 5.4, 95% CI=[2.2, 13.5]) A strong interaction between CAV2 and TMC6 variants was observed (HR=12.1, 95% CI=[3.8, 39]) for children with the deleterious TMC6 variant and without the CAV2 protective variant. Neither gene showed a significant association using an extreme phenotypes design, and conditions for which the power of an extreme phenotype vs. control population design was greater than that for the extreme phenotypes design were explored.

AB - Discovery of rare or low frequency variants in exome or genome data that are associated with complex traits often will require use of very large sample sizes to achieve adequate statistical power. For a fixed sample size, sequencing of individuals sampled from the tails of a phenotype distribution (i.e., extreme phenotypes design) maximizes power and this approach was recently validated empirically with the discovery of variants in DCTN4 that influence the natural history of P. aeruginosa airway infection in persons with cystic fibrosis (CF; MIM219700). The increasing availability of large exome/genome sequence datasets that serve as proxies for population-based controls affords the opportunity to test an alternative, potentially more powerful and generalizable strategy, in which the frequency of rare variants in a single extreme phenotypic group is compared to a control group (i.e., extreme phenotype vs. control population design). As proof-of-principle, we applied this approach to search for variants associated with risk for age-of-onset of chronic P. aeruginosa airway infection among individuals with CF and identified variants in CAV2 and TMC6 that were significantly associated with group status. These results were validated using a large, prospective, longitudinal CF cohort and confirmed a significant association of a variant in CAV2 with increased age-of-onset of P. aeruginosa airway infection (hazard ratio = 0.48, 95% CI=[0.32, 0.88]) and variants in TMC6 with diminished age-of-onset of P. aeruginosa airway infection (HR = 5.4, 95% CI=[2.2, 13.5]) A strong interaction between CAV2 and TMC6 variants was observed (HR=12.1, 95% CI=[3.8, 39]) for children with the deleterious TMC6 variant and without the CAV2 protective variant. Neither gene showed a significant association using an extreme phenotypes design, and conditions for which the power of an extreme phenotype vs. control population design was greater than that for the extreme phenotypes design were explored.

UR - http://www.scopus.com/inward/record.url?scp=84937788259&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84937788259&partnerID=8YFLogxK

U2 - 10.1371/journal.pgen.1005273

DO - 10.1371/journal.pgen.1005273

M3 - Article

VL - 11

JO - PLoS Genetics

JF - PLoS Genetics

SN - 1553-7390

IS - 6

M1 - e1005273

ER -