Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association

Maher Maalouf, Dirar Humouz, Andrzej Kudlicki

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

1 Citation (Scopus)

Abstract

Gene-gene associations are usually inferred from correlations between pairs of genes in different types of biological data such as microarray expression measurements. However coexpression (or correlation) based associations are meaningful only when data sets share the same experimental conditions. In addition, correlation does not indicate a regulatory relationship between two genes. In this work, we adopt another approach to identify connected (with regulatory relationship) pairs of genes utilizing genome-wide expression data with a wide range of different experimental conditions. The number of connected pairs of genes is typically a very small number compared to the total number of pairs in the whole genome. Thus, we can assume that gene-gene connection is a rare event that can be predicted using special classification algorithms. The algorithm used here is called Rare Event-Weighted Kernel Logistic Regression (RE-WKLR). The features that define each pair of genes are the moments of joint probability distribution of expression levels of these two genes. This approach is applied to Saccharomyces cerevisiae genome. The accuracy of RE-WKLR in predicting gene-gene connections is compared with that of Support Vector Machines (SVM) and is found to be higher than that of SVM.

Original languageEnglish (US)
Title of host publicationIIE Annual Conference and Expo 2014
PublisherInstitute of Industrial Engineers
Pages1356-1360
Number of pages5
ISBN (Print)9780983762430
StatePublished - 2014
EventIIE Annual Conference and Expo 2014 - Montreal, Canada
Duration: May 31 2014Jun 3 2014

Other

OtherIIE Annual Conference and Expo 2014
CountryCanada
CityMontreal
Period5/31/146/3/14

Fingerprint

Logistics
Genes
Association reactions
Support vector machines
Microarrays
Yeast
Probability distributions

Keywords

  • Gene-gene association
  • Kernel Logistic regression
  • Microarray expression Data
  • Rare events
  • Yeast

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering
  • Control and Systems Engineering

Cite this

Maalouf, M., Humouz, D., & Kudlicki, A. (2014). Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association. In IIE Annual Conference and Expo 2014 (pp. 1356-1360). Institute of Industrial Engineers.

Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association. / Maalouf, Maher; Humouz, Dirar; Kudlicki, Andrzej.

IIE Annual Conference and Expo 2014. Institute of Industrial Engineers, 2014. p. 1356-1360.

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

Maalouf, M, Humouz, D & Kudlicki, A 2014, Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association. in IIE Annual Conference and Expo 2014. Institute of Industrial Engineers, pp. 1356-1360, IIE Annual Conference and Expo 2014, Montreal, Canada, 5/31/14.
Maalouf M, Humouz D, Kudlicki A. Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association. In IIE Annual Conference and Expo 2014. Institute of Industrial Engineers. 2014. p. 1356-1360
Maalouf, Maher ; Humouz, Dirar ; Kudlicki, Andrzej. / Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association. IIE Annual Conference and Expo 2014. Institute of Industrial Engineers, 2014. pp. 1356-1360
@inproceedings{331452d044524e7ca31649da7dc6cd30,
title = "Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association",
abstract = "Gene-gene associations are usually inferred from correlations between pairs of genes in different types of biological data such as microarray expression measurements. However coexpression (or correlation) based associations are meaningful only when data sets share the same experimental conditions. In addition, correlation does not indicate a regulatory relationship between two genes. In this work, we adopt another approach to identify connected (with regulatory relationship) pairs of genes utilizing genome-wide expression data with a wide range of different experimental conditions. The number of connected pairs of genes is typically a very small number compared to the total number of pairs in the whole genome. Thus, we can assume that gene-gene connection is a rare event that can be predicted using special classification algorithms. The algorithm used here is called Rare Event-Weighted Kernel Logistic Regression (RE-WKLR). The features that define each pair of genes are the moments of joint probability distribution of expression levels of these two genes. This approach is applied to Saccharomyces cerevisiae genome. The accuracy of RE-WKLR in predicting gene-gene connections is compared with that of Support Vector Machines (SVM) and is found to be higher than that of SVM.",
keywords = "Gene-gene association, Kernel Logistic regression, Microarray expression Data, Rare events, Yeast",
author = "Maher Maalouf and Dirar Humouz and Andrzej Kudlicki",
year = "2014",
language = "English (US)",
isbn = "9780983762430",
pages = "1356--1360",
booktitle = "IIE Annual Conference and Expo 2014",
publisher = "Institute of Industrial Engineers",

}

TY - GEN

T1 - Robust Weighted Kernel Logistic Regression to predict gene-gene regulatory association

AU - Maalouf, Maher

AU - Humouz, Dirar

AU - Kudlicki, Andrzej

PY - 2014

Y1 - 2014

N2 - Gene-gene associations are usually inferred from correlations between pairs of genes in different types of biological data such as microarray expression measurements. However coexpression (or correlation) based associations are meaningful only when data sets share the same experimental conditions. In addition, correlation does not indicate a regulatory relationship between two genes. In this work, we adopt another approach to identify connected (with regulatory relationship) pairs of genes utilizing genome-wide expression data with a wide range of different experimental conditions. The number of connected pairs of genes is typically a very small number compared to the total number of pairs in the whole genome. Thus, we can assume that gene-gene connection is a rare event that can be predicted using special classification algorithms. The algorithm used here is called Rare Event-Weighted Kernel Logistic Regression (RE-WKLR). The features that define each pair of genes are the moments of joint probability distribution of expression levels of these two genes. This approach is applied to Saccharomyces cerevisiae genome. The accuracy of RE-WKLR in predicting gene-gene connections is compared with that of Support Vector Machines (SVM) and is found to be higher than that of SVM.

AB - Gene-gene associations are usually inferred from correlations between pairs of genes in different types of biological data such as microarray expression measurements. However coexpression (or correlation) based associations are meaningful only when data sets share the same experimental conditions. In addition, correlation does not indicate a regulatory relationship between two genes. In this work, we adopt another approach to identify connected (with regulatory relationship) pairs of genes utilizing genome-wide expression data with a wide range of different experimental conditions. The number of connected pairs of genes is typically a very small number compared to the total number of pairs in the whole genome. Thus, we can assume that gene-gene connection is a rare event that can be predicted using special classification algorithms. The algorithm used here is called Rare Event-Weighted Kernel Logistic Regression (RE-WKLR). The features that define each pair of genes are the moments of joint probability distribution of expression levels of these two genes. This approach is applied to Saccharomyces cerevisiae genome. The accuracy of RE-WKLR in predicting gene-gene connections is compared with that of Support Vector Machines (SVM) and is found to be higher than that of SVM.

KW - Gene-gene association

KW - Kernel Logistic regression

KW - Microarray expression Data

KW - Rare events

KW - Yeast

UR - http://www.scopus.com/inward/record.url?scp=84910025585&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84910025585&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9780983762430

SP - 1356

EP - 1360

BT - IIE Annual Conference and Expo 2014

PB - Institute of Industrial Engineers

ER -