Testing the robustness of the likelihood-ratio test in a variance- component quantitative-trait loci-mapping procedure

David B. Allison, Michael C. Neale, Raffaella Zannolli, Nicholas J. Schork, Christopher I. Amos, John Blangero

Research output: Contribution to journalArticle

255 Citations (Scopus)

Abstract

Detection of linkage to genes for quantitative traits remains a challenging task. Recently, variance components (VC) techniques have emerged as among the more powerful of available methods. As often implemented, such techniques require assumptions about the phenotypic distribution. Usually, multivariate normality is assumed. However, several factors may lead to markedly nonnormal phenotypic data, including (a) the presence of a major gene (not necessarily linked to the markers under study), (b) some types of gene x environment interaction, (c) use of a dichotomous phenotype (i.e., affected vs. unaffected), (d) nonnormality of the population within-genotype (residual) distribution, and (e) selective (extreme) sampling. Using simulation, we have investigated, for sib-pair studies, the robustness of the likelihood-ratio test for a VC quantitative-trait locus-detection procedure to violations of normality that are due to these factors. Results showed (a) that some types of nonnormality, such as leptokurtosis, produced type I error rates in excess of the nominal, or α, levels whereas others did not; and (b) that the degree of type I error-rate inflation appears to be directly related to the residual sibling correlation. Potential solutions to this problem are discussed. Investigators contemplating use of this VC procedure are encouraged to provide evidence that their trait data are normally distributed, to employ a procedure that allows for nonnormal data, or to consider implementation of permutation tests.

Original languageEnglish (US)
Pages (from-to)531-544
Number of pages14
JournalAmerican Journal of Human Genetics
Volume65
Issue number2
DOIs
StatePublished - 1999
Externally publishedYes

Fingerprint

Quantitative Trait Loci
Gene-Environment Interaction
Economic Inflation
Genes
Genotype
Research Personnel
Phenotype
Population

ASJC Scopus subject areas

  • Genetics

Cite this

Testing the robustness of the likelihood-ratio test in a variance- component quantitative-trait loci-mapping procedure. / Allison, David B.; Neale, Michael C.; Zannolli, Raffaella; Schork, Nicholas J.; Amos, Christopher I.; Blangero, John.

In: American Journal of Human Genetics, Vol. 65, No. 2, 1999, p. 531-544.

Research output: Contribution to journalArticle

Allison, David B. ; Neale, Michael C. ; Zannolli, Raffaella ; Schork, Nicholas J. ; Amos, Christopher I. ; Blangero, John. / Testing the robustness of the likelihood-ratio test in a variance- component quantitative-trait loci-mapping procedure. In: American Journal of Human Genetics. 1999 ; Vol. 65, No. 2. pp. 531-544.
@article{2d151a60d48648669dd5b2ebcaca47bd,
title = "Testing the robustness of the likelihood-ratio test in a variance- component quantitative-trait loci-mapping procedure",
abstract = "Detection of linkage to genes for quantitative traits remains a challenging task. Recently, variance components (VC) techniques have emerged as among the more powerful of available methods. As often implemented, such techniques require assumptions about the phenotypic distribution. Usually, multivariate normality is assumed. However, several factors may lead to markedly nonnormal phenotypic data, including (a) the presence of a major gene (not necessarily linked to the markers under study), (b) some types of gene x environment interaction, (c) use of a dichotomous phenotype (i.e., affected vs. unaffected), (d) nonnormality of the population within-genotype (residual) distribution, and (e) selective (extreme) sampling. Using simulation, we have investigated, for sib-pair studies, the robustness of the likelihood-ratio test for a VC quantitative-trait locus-detection procedure to violations of normality that are due to these factors. Results showed (a) that some types of nonnormality, such as leptokurtosis, produced type I error rates in excess of the nominal, or α, levels whereas others did not; and (b) that the degree of type I error-rate inflation appears to be directly related to the residual sibling correlation. Potential solutions to this problem are discussed. Investigators contemplating use of this VC procedure are encouraged to provide evidence that their trait data are normally distributed, to employ a procedure that allows for nonnormal data, or to consider implementation of permutation tests.",
author = "Allison, {David B.} and Neale, {Michael C.} and Raffaella Zannolli and Schork, {Nicholas J.} and Amos, {Christopher I.} and John Blangero",
year = "1999",
doi = "10.1086/302487",
language = "English (US)",
volume = "65",
pages = "531--544",
journal = "American Journal of Human Genetics",
issn = "0002-9297",
publisher = "Cell Press",
number = "2",

}

TY - JOUR

T1 - Testing the robustness of the likelihood-ratio test in a variance- component quantitative-trait loci-mapping procedure

AU - Allison, David B.

AU - Neale, Michael C.

AU - Zannolli, Raffaella

AU - Schork, Nicholas J.

AU - Amos, Christopher I.

AU - Blangero, John

PY - 1999

Y1 - 1999

N2 - Detection of linkage to genes for quantitative traits remains a challenging task. Recently, variance components (VC) techniques have emerged as among the more powerful of available methods. As often implemented, such techniques require assumptions about the phenotypic distribution. Usually, multivariate normality is assumed. However, several factors may lead to markedly nonnormal phenotypic data, including (a) the presence of a major gene (not necessarily linked to the markers under study), (b) some types of gene x environment interaction, (c) use of a dichotomous phenotype (i.e., affected vs. unaffected), (d) nonnormality of the population within-genotype (residual) distribution, and (e) selective (extreme) sampling. Using simulation, we have investigated, for sib-pair studies, the robustness of the likelihood-ratio test for a VC quantitative-trait locus-detection procedure to violations of normality that are due to these factors. Results showed (a) that some types of nonnormality, such as leptokurtosis, produced type I error rates in excess of the nominal, or α, levels whereas others did not; and (b) that the degree of type I error-rate inflation appears to be directly related to the residual sibling correlation. Potential solutions to this problem are discussed. Investigators contemplating use of this VC procedure are encouraged to provide evidence that their trait data are normally distributed, to employ a procedure that allows for nonnormal data, or to consider implementation of permutation tests.

AB - Detection of linkage to genes for quantitative traits remains a challenging task. Recently, variance components (VC) techniques have emerged as among the more powerful of available methods. As often implemented, such techniques require assumptions about the phenotypic distribution. Usually, multivariate normality is assumed. However, several factors may lead to markedly nonnormal phenotypic data, including (a) the presence of a major gene (not necessarily linked to the markers under study), (b) some types of gene x environment interaction, (c) use of a dichotomous phenotype (i.e., affected vs. unaffected), (d) nonnormality of the population within-genotype (residual) distribution, and (e) selective (extreme) sampling. Using simulation, we have investigated, for sib-pair studies, the robustness of the likelihood-ratio test for a VC quantitative-trait locus-detection procedure to violations of normality that are due to these factors. Results showed (a) that some types of nonnormality, such as leptokurtosis, produced type I error rates in excess of the nominal, or α, levels whereas others did not; and (b) that the degree of type I error-rate inflation appears to be directly related to the residual sibling correlation. Potential solutions to this problem are discussed. Investigators contemplating use of this VC procedure are encouraged to provide evidence that their trait data are normally distributed, to employ a procedure that allows for nonnormal data, or to consider implementation of permutation tests.

UR - http://www.scopus.com/inward/record.url?scp=0033358545&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033358545&partnerID=8YFLogxK

U2 - 10.1086/302487

DO - 10.1086/302487

M3 - Article

VL - 65

SP - 531

EP - 544

JO - American Journal of Human Genetics

JF - American Journal of Human Genetics

SN - 0002-9297

IS - 2

ER -