GPNN: Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease

Alison A. Motsinger, Stephen L. Lee, George Mellick, Marylyn Deriggi Ritchie

Research output: Contribution to journalArticle

60 Citations (Scopus)

Abstract

Background: The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results: We show that GPNN has high power to detect even relatively small genetic effects (2-3% heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (<1%) or when interactions involved more than three loci. We tested GPNN on a real dataset comprised of Parkinson's disease cases and controls and found a two locus interaction between the DLST gene and sex. Conclusion: These results indicate that GPNN may be a useful pattern recognition approach for detecting gene-gene and gene-environment interactions.

Original languageEnglish (US)
Article number39
JournalBMC Bioinformatics
Volume7
DOIs
StatePublished - Jan 25 2006

Fingerprint

Genetic programming
Genetic Programming
Genes
Neural Networks
Gene
Neural networks
Interaction
Heritability
Locus
Parkinson's Disease
Parkinson Disease
Genetic Epidemiology
Gene-environment Interaction
Gene-Environment Interaction
Gene Order
Molecular Epidemiology
Human
Environmental Factors
Limit of Detection
High Power

All Science Journal Classification (ASJC) codes

  • Structural Biology
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Applied Mathematics

Cite this

Motsinger, Alison A. ; Lee, Stephen L. ; Mellick, George ; Ritchie, Marylyn Deriggi. / GPNN : Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease. In: BMC Bioinformatics. 2006 ; Vol. 7.
@article{cbc919ff2535437c95a075c016efd989,
title = "GPNN: Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease",
abstract = "Background: The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results: We show that GPNN has high power to detect even relatively small genetic effects (2-3{\%} heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (<1{\%}) or when interactions involved more than three loci. We tested GPNN on a real dataset comprised of Parkinson's disease cases and controls and found a two locus interaction between the DLST gene and sex. Conclusion: These results indicate that GPNN may be a useful pattern recognition approach for detecting gene-gene and gene-environment interactions.",
author = "Motsinger, {Alison A.} and Lee, {Stephen L.} and George Mellick and Ritchie, {Marylyn Deriggi}",
year = "2006",
month = "1",
day = "25",
doi = "10.1186/1471-2105-7-39",
language = "English (US)",
volume = "7",
journal = "BMC Bioinformatics",
issn = "1471-2105",
publisher = "BioMed Central",

}

GPNN : Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease. / Motsinger, Alison A.; Lee, Stephen L.; Mellick, George; Ritchie, Marylyn Deriggi.

In: BMC Bioinformatics, Vol. 7, 39, 25.01.2006.

Research output: Contribution to journalArticle

TY - JOUR

T1 - GPNN

T2 - Power studies and applications of a neural network method for detecting gene-gene interactions in studies of human disease

AU - Motsinger, Alison A.

AU - Lee, Stephen L.

AU - Mellick, George

AU - Ritchie, Marylyn Deriggi

PY - 2006/1/25

Y1 - 2006/1/25

N2 - Background: The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results: We show that GPNN has high power to detect even relatively small genetic effects (2-3% heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (<1%) or when interactions involved more than three loci. We tested GPNN on a real dataset comprised of Parkinson's disease cases and controls and found a two locus interaction between the DLST gene and sex. Conclusion: These results indicate that GPNN may be a useful pattern recognition approach for detecting gene-gene and gene-environment interactions.

AB - Background: The identification and characterization of genes that influence the risk of common, complex multifactorial disease primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. We have previously introduced a genetic programming optimized neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. The goal of this study was to evaluate the power of GPNN for identifying high-order gene-gene interactions. We were also interested in applying GPNN to a real data analysis in Parkinson's disease. Results: We show that GPNN has high power to detect even relatively small genetic effects (2-3% heritability) in simulated data models involving two and three locus interactions. The limits of detection were reached under conditions with very small heritability (<1%) or when interactions involved more than three loci. We tested GPNN on a real dataset comprised of Parkinson's disease cases and controls and found a two locus interaction between the DLST gene and sex. Conclusion: These results indicate that GPNN may be a useful pattern recognition approach for detecting gene-gene and gene-environment interactions.

UR - http://www.scopus.com/inward/record.url?scp=33644747451&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33644747451&partnerID=8YFLogxK

U2 - 10.1186/1471-2105-7-39

DO - 10.1186/1471-2105-7-39

M3 - Article

C2 - 16436204

AN - SCOPUS:33644747451

VL - 7

JO - BMC Bioinformatics

JF - BMC Bioinformatics

SN - 1471-2105

M1 - 39

ER -