Extension of multifactor dimensionality reduction for identifying multilocus effects in the GAW14 simulated data

Hao Mei, Deqiong Ma, Allison Ashley-Koch, Eden R Martin

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

The multifactor dimensionality reduction (MDR) is a model-free approach that can identify gene x gene or gene x environment effects in a case-control study. Here we explore several modifications of the MDR method. We extended MDR to provide model selection without crossvalidation, and use a chi-square statistic as an alternative to prediction error (PE). We also modified the permutation test to provide different levels of stringency. The extended MDR (EMDR) includes three permutation tests (fixed, non-fixed, and omnibus) to obtain p-values of multilocus models. The goal of this study was to compare the different approaches implemented in the EMDR method and evaluate the ability to identify genetic effects in the Genetic Analysis Workshop 14 simulated data. We used three replicates from the simulated family data, generating matched pairs from family triads. The results showed: 1) chi-square and PE statistics give nearly consistent results; 2) results of EMDR without cross-validation matched that of EMDR with 10-fold cross-validation; 3) the fixed permutation test reports false-positive results in data from loci unrelated to the disease, but the non-fixed and omnibus permutation tests perform well in preventing false positives, with the omnibus test being the most conservative. We conclude that the non-cross-validation test can provide accurate results with the advantage of high efficiency compared to 10-cross-validation, and the non-fixed permutation test provides a good compromise between power and false-positive rate.

Original languageEnglish
Article numberS145
JournalBMC Genetics
Volume6
Issue numberSUPPL.1
DOIs
StatePublished - Dec 30 2005
Externally publishedYes

Fingerprint

Multifactor Dimensionality Reduction
Genes
Case-Control Studies
Education

ASJC Scopus subject areas

  • Genetics
  • Genetics(clinical)

Cite this

Extension of multifactor dimensionality reduction for identifying multilocus effects in the GAW14 simulated data. / Mei, Hao; Ma, Deqiong; Ashley-Koch, Allison; Martin, Eden R.

In: BMC Genetics, Vol. 6, No. SUPPL.1, S145, 30.12.2005.

Research output: Contribution to journalArticle

@article{65d3d834db2a43abb354475f85773eaf,
title = "Extension of multifactor dimensionality reduction for identifying multilocus effects in the GAW14 simulated data",
abstract = "The multifactor dimensionality reduction (MDR) is a model-free approach that can identify gene x gene or gene x environment effects in a case-control study. Here we explore several modifications of the MDR method. We extended MDR to provide model selection without crossvalidation, and use a chi-square statistic as an alternative to prediction error (PE). We also modified the permutation test to provide different levels of stringency. The extended MDR (EMDR) includes three permutation tests (fixed, non-fixed, and omnibus) to obtain p-values of multilocus models. The goal of this study was to compare the different approaches implemented in the EMDR method and evaluate the ability to identify genetic effects in the Genetic Analysis Workshop 14 simulated data. We used three replicates from the simulated family data, generating matched pairs from family triads. The results showed: 1) chi-square and PE statistics give nearly consistent results; 2) results of EMDR without cross-validation matched that of EMDR with 10-fold cross-validation; 3) the fixed permutation test reports false-positive results in data from loci unrelated to the disease, but the non-fixed and omnibus permutation tests perform well in preventing false positives, with the omnibus test being the most conservative. We conclude that the non-cross-validation test can provide accurate results with the advantage of high efficiency compared to 10-cross-validation, and the non-fixed permutation test provides a good compromise between power and false-positive rate.",
author = "Hao Mei and Deqiong Ma and Allison Ashley-Koch and Martin, {Eden R}",
year = "2005",
month = "12",
day = "30",
doi = "10.1186/1471-2156-6-S1-S145",
language = "English",
volume = "6",
journal = "BMC Genetics",
issn = "1471-2156",
publisher = "BioMed Central",
number = "SUPPL.1",

}

TY - JOUR

T1 - Extension of multifactor dimensionality reduction for identifying multilocus effects in the GAW14 simulated data

AU - Mei, Hao

AU - Ma, Deqiong

AU - Ashley-Koch, Allison

AU - Martin, Eden R

PY - 2005/12/30

Y1 - 2005/12/30

N2 - The multifactor dimensionality reduction (MDR) is a model-free approach that can identify gene x gene or gene x environment effects in a case-control study. Here we explore several modifications of the MDR method. We extended MDR to provide model selection without crossvalidation, and use a chi-square statistic as an alternative to prediction error (PE). We also modified the permutation test to provide different levels of stringency. The extended MDR (EMDR) includes three permutation tests (fixed, non-fixed, and omnibus) to obtain p-values of multilocus models. The goal of this study was to compare the different approaches implemented in the EMDR method and evaluate the ability to identify genetic effects in the Genetic Analysis Workshop 14 simulated data. We used three replicates from the simulated family data, generating matched pairs from family triads. The results showed: 1) chi-square and PE statistics give nearly consistent results; 2) results of EMDR without cross-validation matched that of EMDR with 10-fold cross-validation; 3) the fixed permutation test reports false-positive results in data from loci unrelated to the disease, but the non-fixed and omnibus permutation tests perform well in preventing false positives, with the omnibus test being the most conservative. We conclude that the non-cross-validation test can provide accurate results with the advantage of high efficiency compared to 10-cross-validation, and the non-fixed permutation test provides a good compromise between power and false-positive rate.

AB - The multifactor dimensionality reduction (MDR) is a model-free approach that can identify gene x gene or gene x environment effects in a case-control study. Here we explore several modifications of the MDR method. We extended MDR to provide model selection without crossvalidation, and use a chi-square statistic as an alternative to prediction error (PE). We also modified the permutation test to provide different levels of stringency. The extended MDR (EMDR) includes three permutation tests (fixed, non-fixed, and omnibus) to obtain p-values of multilocus models. The goal of this study was to compare the different approaches implemented in the EMDR method and evaluate the ability to identify genetic effects in the Genetic Analysis Workshop 14 simulated data. We used three replicates from the simulated family data, generating matched pairs from family triads. The results showed: 1) chi-square and PE statistics give nearly consistent results; 2) results of EMDR without cross-validation matched that of EMDR with 10-fold cross-validation; 3) the fixed permutation test reports false-positive results in data from loci unrelated to the disease, but the non-fixed and omnibus permutation tests perform well in preventing false positives, with the omnibus test being the most conservative. We conclude that the non-cross-validation test can provide accurate results with the advantage of high efficiency compared to 10-cross-validation, and the non-fixed permutation test provides a good compromise between power and false-positive rate.

UR - http://www.scopus.com/inward/record.url?scp=28444496622&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=28444496622&partnerID=8YFLogxK

U2 - 10.1186/1471-2156-6-S1-S145

DO - 10.1186/1471-2156-6-S1-S145

M3 - Article

VL - 6

JO - BMC Genetics

JF - BMC Genetics

SN - 1471-2156

IS - SUPPL.1

M1 - S145

ER -