Maximizing association statistics over genetic models

Juan R. González, Josep L. Carrasco, Frank Dudbridge, Lluís Armengol, Xavier P. Estivill, Victor Moreno

Research output: Contribution to journalArticle

80 Citations (Scopus)

Abstract

The assessment of the association between a candidate locus and a disease may require the assumption of an inheritance model. Most researchers select the additive model and test the association with the Cochran-Armitage trend test. This test assumes a dose-response effect with regard to the number of copies of the variant allele. However, if there is reason to expect dominance or recessiveness in the effect of the variant allele, the heterozygous genotype may be grouped with one of the two homozygous, depending on the inheritance model, and a simple test on the 2 x 2 table can be used to assess independence. When the underlying genetic model is unknown, association may be assessed using the max-statistic, which selects the largest test statistic from the dominant, recessive and additive models. The statistical significance of the maxstatistic has been previously addressed using permutation or Monte Carlo simulation approaches. We aimed to provide simpler alternatives to the max-test to make it feasible in large-scale association studies. Our simulations show that this procedure has an effective number of tests of 2.2, which can be used to correct the significance level or P-values. We also derive the asymptotic distribution of max-statistic, which leads to a simple way to calculate the significance level and allows the derivation of a formula for power calculations in the design of studies that plan to use the max-statistic. A simulation study shows that the use of the max-statistic is a powerful approach that provides safeguard against model uncertainty.

Original languageEnglish
Pages (from-to)246-254
Number of pages9
JournalGenetic Epidemiology
Volume32
Issue number3
DOIs
Publication statusPublished - Apr 2008
Externally publishedYes

Fingerprint

Genetic Models
Alleles
Uncertainty
Genotype
Research Personnel

Keywords

  • Genetic models
  • Max-statistic
  • Mode of inheritance
  • Power

ASJC Scopus subject areas

  • Epidemiology
  • Genetics(clinical)

Cite this

González, J. R., Carrasco, J. L., Dudbridge, F., Armengol, L., Estivill, X. P., & Moreno, V. (2008). Maximizing association statistics over genetic models. Genetic Epidemiology, 32(3), 246-254. https://doi.org/10.1002/gepi.20299

Maximizing association statistics over genetic models. / González, Juan R.; Carrasco, Josep L.; Dudbridge, Frank; Armengol, Lluís; Estivill, Xavier P.; Moreno, Victor.

In: Genetic Epidemiology, Vol. 32, No. 3, 04.2008, p. 246-254.

Research output: Contribution to journalArticle

González, JR, Carrasco, JL, Dudbridge, F, Armengol, L, Estivill, XP & Moreno, V 2008, 'Maximizing association statistics over genetic models', Genetic Epidemiology, vol. 32, no. 3, pp. 246-254. https://doi.org/10.1002/gepi.20299
González JR, Carrasco JL, Dudbridge F, Armengol L, Estivill XP, Moreno V. Maximizing association statistics over genetic models. Genetic Epidemiology. 2008 Apr;32(3):246-254. https://doi.org/10.1002/gepi.20299
González, Juan R. ; Carrasco, Josep L. ; Dudbridge, Frank ; Armengol, Lluís ; Estivill, Xavier P. ; Moreno, Victor. / Maximizing association statistics over genetic models. In: Genetic Epidemiology. 2008 ; Vol. 32, No. 3. pp. 246-254.
@article{47edec8354c44d3ebe8560e9315e878a,
title = "Maximizing association statistics over genetic models",
abstract = "The assessment of the association between a candidate locus and a disease may require the assumption of an inheritance model. Most researchers select the additive model and test the association with the Cochran-Armitage trend test. This test assumes a dose-response effect with regard to the number of copies of the variant allele. However, if there is reason to expect dominance or recessiveness in the effect of the variant allele, the heterozygous genotype may be grouped with one of the two homozygous, depending on the inheritance model, and a simple test on the 2 x 2 table can be used to assess independence. When the underlying genetic model is unknown, association may be assessed using the max-statistic, which selects the largest test statistic from the dominant, recessive and additive models. The statistical significance of the maxstatistic has been previously addressed using permutation or Monte Carlo simulation approaches. We aimed to provide simpler alternatives to the max-test to make it feasible in large-scale association studies. Our simulations show that this procedure has an effective number of tests of 2.2, which can be used to correct the significance level or P-values. We also derive the asymptotic distribution of max-statistic, which leads to a simple way to calculate the significance level and allows the derivation of a formula for power calculations in the design of studies that plan to use the max-statistic. A simulation study shows that the use of the max-statistic is a powerful approach that provides safeguard against model uncertainty.",
keywords = "Genetic models, Max-statistic, Mode of inheritance, Power",
author = "Gonz{\'a}lez, {Juan R.} and Carrasco, {Josep L.} and Frank Dudbridge and Llu{\'i}s Armengol and Estivill, {Xavier P.} and Victor Moreno",
year = "2008",
month = "4",
doi = "10.1002/gepi.20299",
language = "English",
volume = "32",
pages = "246--254",
journal = "Genetic Epidemiology",
issn = "0741-0395",
publisher = "Wiley-Liss Inc.",
number = "3",

}

TY - JOUR

T1 - Maximizing association statistics over genetic models

AU - González, Juan R.

AU - Carrasco, Josep L.

AU - Dudbridge, Frank

AU - Armengol, Lluís

AU - Estivill, Xavier P.

AU - Moreno, Victor

PY - 2008/4

Y1 - 2008/4

N2 - The assessment of the association between a candidate locus and a disease may require the assumption of an inheritance model. Most researchers select the additive model and test the association with the Cochran-Armitage trend test. This test assumes a dose-response effect with regard to the number of copies of the variant allele. However, if there is reason to expect dominance or recessiveness in the effect of the variant allele, the heterozygous genotype may be grouped with one of the two homozygous, depending on the inheritance model, and a simple test on the 2 x 2 table can be used to assess independence. When the underlying genetic model is unknown, association may be assessed using the max-statistic, which selects the largest test statistic from the dominant, recessive and additive models. The statistical significance of the maxstatistic has been previously addressed using permutation or Monte Carlo simulation approaches. We aimed to provide simpler alternatives to the max-test to make it feasible in large-scale association studies. Our simulations show that this procedure has an effective number of tests of 2.2, which can be used to correct the significance level or P-values. We also derive the asymptotic distribution of max-statistic, which leads to a simple way to calculate the significance level and allows the derivation of a formula for power calculations in the design of studies that plan to use the max-statistic. A simulation study shows that the use of the max-statistic is a powerful approach that provides safeguard against model uncertainty.

AB - The assessment of the association between a candidate locus and a disease may require the assumption of an inheritance model. Most researchers select the additive model and test the association with the Cochran-Armitage trend test. This test assumes a dose-response effect with regard to the number of copies of the variant allele. However, if there is reason to expect dominance or recessiveness in the effect of the variant allele, the heterozygous genotype may be grouped with one of the two homozygous, depending on the inheritance model, and a simple test on the 2 x 2 table can be used to assess independence. When the underlying genetic model is unknown, association may be assessed using the max-statistic, which selects the largest test statistic from the dominant, recessive and additive models. The statistical significance of the maxstatistic has been previously addressed using permutation or Monte Carlo simulation approaches. We aimed to provide simpler alternatives to the max-test to make it feasible in large-scale association studies. Our simulations show that this procedure has an effective number of tests of 2.2, which can be used to correct the significance level or P-values. We also derive the asymptotic distribution of max-statistic, which leads to a simple way to calculate the significance level and allows the derivation of a formula for power calculations in the design of studies that plan to use the max-statistic. A simulation study shows that the use of the max-statistic is a powerful approach that provides safeguard against model uncertainty.

KW - Genetic models

KW - Max-statistic

KW - Mode of inheritance

KW - Power

UR - http://www.scopus.com/inward/record.url?scp=42249096145&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=42249096145&partnerID=8YFLogxK

U2 - 10.1002/gepi.20299

DO - 10.1002/gepi.20299

M3 - Article

C2 - 18228557

AN - SCOPUS:42249096145

VL - 32

SP - 246

EP - 254

JO - Genetic Epidemiology

JF - Genetic Epidemiology

SN - 0741-0395

IS - 3

ER -