Initialization parameter sweep in ATHENA: Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects

Emily R. Holzinger, Carrie C. Buchanan, Scott M. Dudek, Eric C. Torstenson, Stephen D. Turner, Marylyn D. Ritchie

Research output: Chapter in Book/Report/Conference proceedingConference contribution

14 Citations (Scopus)

Abstract

Recent advances in genotyping technology have led to the generation of an enormous quantity of genetic data. Traditional methods of statistical analysis have proved insufficient in extracting all of the information about the genetic components of common, complex human diseases. A contributing factor to the problem of analysis is that amongst the small main effects of each single gene on disease susceptibility, there are non-linear, genegene interactions that can be difficult for traditional, parametric analyses to detect. In addition, exhaustively searching all multilocus combinations has proved computationally impractical. Novel strategies for analysis have been developed to address these issues. The Analysis Tool for Heritable and Environmental Network Associations (ATHENA) is an analytical tool that incorporates grammatical evolution neural networks (GENN) to detect interactions among genetic factors. Initial parameters define how the evolutionary process will be implemented. This research addresses how different parameter settings affect detection of disease models involving interactions. In the current study, we iterate over multiple parameter values to determine which combinations appear optimal for detecting interactions in simulated data for multiple genetic models. Our results indicate that the factors that have the greatest influence on detection are: input variable encoding, population size, and parallel computation.

Original languageEnglish (US)
Title of host publicationProceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10
Pages203-210
Number of pages8
DOIs
StatePublished - Aug 27 2010
Event12th Annual Genetic and Evolutionary Computation Conference, GECCO-2010 - Portland, OR, United States
Duration: Jul 7 2010Jul 11 2010

Publication series

NameProceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10

Other

Other12th Annual Genetic and Evolutionary Computation Conference, GECCO-2010
CountryUnited States
CityPortland, OR
Period7/7/107/11/10

Fingerprint

Main Effect
Sweep
Initialization
Genes
Neural Networks
Gene
Neural networks
Interaction
Grammatical Evolution
Nonlinear Interaction
Statistical methods
Parallel Computation
Population Size
Iterate
Susceptibility
Statistical Analysis
Encoding
Model

All Science Journal Classification (ASJC) codes

  • Computational Theory and Mathematics
  • Theoretical Computer Science

Cite this

Holzinger, E. R., Buchanan, C. C., Dudek, S. M., Torstenson, E. C., Turner, S. D., & Ritchie, M. D. (2010). Initialization parameter sweep in ATHENA: Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects. In Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10 (pp. 203-210). (Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10). https://doi.org/10.1145/1830483.1830519
Holzinger, Emily R. ; Buchanan, Carrie C. ; Dudek, Scott M. ; Torstenson, Eric C. ; Turner, Stephen D. ; Ritchie, Marylyn D. / Initialization parameter sweep in ATHENA : Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects. Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. 2010. pp. 203-210 (Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10).
@inproceedings{0160109342574a8c9cabd8329be4bca8,
title = "Initialization parameter sweep in ATHENA: Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects",
abstract = "Recent advances in genotyping technology have led to the generation of an enormous quantity of genetic data. Traditional methods of statistical analysis have proved insufficient in extracting all of the information about the genetic components of common, complex human diseases. A contributing factor to the problem of analysis is that amongst the small main effects of each single gene on disease susceptibility, there are non-linear, genegene interactions that can be difficult for traditional, parametric analyses to detect. In addition, exhaustively searching all multilocus combinations has proved computationally impractical. Novel strategies for analysis have been developed to address these issues. The Analysis Tool for Heritable and Environmental Network Associations (ATHENA) is an analytical tool that incorporates grammatical evolution neural networks (GENN) to detect interactions among genetic factors. Initial parameters define how the evolutionary process will be implemented. This research addresses how different parameter settings affect detection of disease models involving interactions. In the current study, we iterate over multiple parameter values to determine which combinations appear optimal for detecting interactions in simulated data for multiple genetic models. Our results indicate that the factors that have the greatest influence on detection are: input variable encoding, population size, and parallel computation.",
author = "Holzinger, {Emily R.} and Buchanan, {Carrie C.} and Dudek, {Scott M.} and Torstenson, {Eric C.} and Turner, {Stephen D.} and Ritchie, {Marylyn D.}",
year = "2010",
month = "8",
day = "27",
doi = "10.1145/1830483.1830519",
language = "English (US)",
isbn = "9781450300728",
series = "Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10",
pages = "203--210",
booktitle = "Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10",

}

Holzinger, ER, Buchanan, CC, Dudek, SM, Torstenson, EC, Turner, SD & Ritchie, MD 2010, Initialization parameter sweep in ATHENA: Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects. in Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10, pp. 203-210, 12th Annual Genetic and Evolutionary Computation Conference, GECCO-2010, Portland, OR, United States, 7/7/10. https://doi.org/10.1145/1830483.1830519

Initialization parameter sweep in ATHENA : Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects. / Holzinger, Emily R.; Buchanan, Carrie C.; Dudek, Scott M.; Torstenson, Eric C.; Turner, Stephen D.; Ritchie, Marylyn D.

Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. 2010. p. 203-210 (Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Initialization parameter sweep in ATHENA

T2 - Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects

AU - Holzinger, Emily R.

AU - Buchanan, Carrie C.

AU - Dudek, Scott M.

AU - Torstenson, Eric C.

AU - Turner, Stephen D.

AU - Ritchie, Marylyn D.

PY - 2010/8/27

Y1 - 2010/8/27

N2 - Recent advances in genotyping technology have led to the generation of an enormous quantity of genetic data. Traditional methods of statistical analysis have proved insufficient in extracting all of the information about the genetic components of common, complex human diseases. A contributing factor to the problem of analysis is that amongst the small main effects of each single gene on disease susceptibility, there are non-linear, genegene interactions that can be difficult for traditional, parametric analyses to detect. In addition, exhaustively searching all multilocus combinations has proved computationally impractical. Novel strategies for analysis have been developed to address these issues. The Analysis Tool for Heritable and Environmental Network Associations (ATHENA) is an analytical tool that incorporates grammatical evolution neural networks (GENN) to detect interactions among genetic factors. Initial parameters define how the evolutionary process will be implemented. This research addresses how different parameter settings affect detection of disease models involving interactions. In the current study, we iterate over multiple parameter values to determine which combinations appear optimal for detecting interactions in simulated data for multiple genetic models. Our results indicate that the factors that have the greatest influence on detection are: input variable encoding, population size, and parallel computation.

AB - Recent advances in genotyping technology have led to the generation of an enormous quantity of genetic data. Traditional methods of statistical analysis have proved insufficient in extracting all of the information about the genetic components of common, complex human diseases. A contributing factor to the problem of analysis is that amongst the small main effects of each single gene on disease susceptibility, there are non-linear, genegene interactions that can be difficult for traditional, parametric analyses to detect. In addition, exhaustively searching all multilocus combinations has proved computationally impractical. Novel strategies for analysis have been developed to address these issues. The Analysis Tool for Heritable and Environmental Network Associations (ATHENA) is an analytical tool that incorporates grammatical evolution neural networks (GENN) to detect interactions among genetic factors. Initial parameters define how the evolutionary process will be implemented. This research addresses how different parameter settings affect detection of disease models involving interactions. In the current study, we iterate over multiple parameter values to determine which combinations appear optimal for detecting interactions in simulated data for multiple genetic models. Our results indicate that the factors that have the greatest influence on detection are: input variable encoding, population size, and parallel computation.

UR - http://www.scopus.com/inward/record.url?scp=77955883431&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955883431&partnerID=8YFLogxK

U2 - 10.1145/1830483.1830519

DO - 10.1145/1830483.1830519

M3 - Conference contribution

AN - SCOPUS:77955883431

SN - 9781450300728

T3 - Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10

SP - 203

EP - 210

BT - Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10

ER -

Holzinger ER, Buchanan CC, Dudek SM, Torstenson EC, Turner SD, Ritchie MD. Initialization parameter sweep in ATHENA: Optimizing neural networks for detecting gene-gene interactions in the presence of small main effects. In Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10. 2010. p. 203-210. (Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, GECCO '10). https://doi.org/10.1145/1830483.1830519