Feature selection for high-dimensional integrated data

Charles Zheng; Scott Schwartz; Robert S. Chapkin; Raymond J. Carroll; Ivan Ivanov

doi:10.1137/1.9781611972825.98

Feature selection for high-dimensional integrated data

Charles Zheng, Scott Schwartz, Robert S. Chapkin, Raymond J. Carroll, Ivan Ivanov

Research output: Chapter in Book/Report/Conference proceeding › Chapter

1 Scopus citations

Abstract

Motivated by the problem of identifying correlations between genes or features of two related biological systems, we propose a model of feature selection in which only a subset of the predictors Xt are dependent on the multidimensional variate Y, and the remainder of the predictors constitute a “noise set” Xu independent of Y. Using Monte Carlo simulations, we investigated the relative performance of two methods: thresholding and singular-value decomposition, in combination with stochastic optimization to determine “empirical bounds” on the small-sample accuracy of an asymptotic approximation. We demonstrate utility of the thresholding and SVD feature selection methods to with respect to a recent infant intestinal gene expression and metagenomics dataset.

Original language	English (US)
Title of host publication	Proceedings of the 2012 SIAM International Conference on Data Mining
Publisher	Society for Industrial & Applied Mathematics (SIAM)
ISBN (Print)	9781611972320
DOIs	https://doi.org/10.1137/1.9781611972825.98
State	Published - Dec 18 2013
Externally published	Yes

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledged KAUST grant number(s): KUS-C1-016-04
Acknowledgements: We are indebted to the Texas A& M Brazos Computing Cluster and Institute of Developmentaland Molecular Biology for access to computingresources, and to professors David B. Dahl,Mohsen Pourahmadi, and Joel Zinn for helpful discussions.The infant microarray-metagenomics data wasprovided courtesy of Sharon M. Donovan, of the Divisionof Nutritional Sciences, U. of Illinois, Urbana, IL.This publication is based in part on work supported byAward No. KUS-C1-016-04, made by King AbdullahUniversity of Science and Technology (KAUST).
This publication acknowledges KAUST support, but has no KAUST affiliated authors.

Access to Document

10.1137/1.9781611972825.98

Cite this

@inbook{68871148e8994391ab3a3fbc951f2983,

title = "Feature selection for high-dimensional integrated data",

abstract = "Motivated by the problem of identifying correlations between genes or features of two related biological systems, we propose a model of feature selection in which only a subset of the predictors Xt are dependent on the multidimensional variate Y, and the remainder of the predictors constitute a “noise set” Xu independent of Y. Using Monte Carlo simulations, we investigated the relative performance of two methods: thresholding and singular-value decomposition, in combination with stochastic optimization to determine “empirical bounds” on the small-sample accuracy of an asymptotic approximation. We demonstrate utility of the thresholding and SVD feature selection methods to with respect to a recent infant intestinal gene expression and metagenomics dataset.",

author = "Charles Zheng and Scott Schwartz and Chapkin, {Robert S.} and Carroll, {Raymond J.} and Ivan Ivanov",

note = "KAUST Repository Item: Exported on 2020-10-01 Acknowledged KAUST grant number(s): KUS-C1-016-04 Acknowledgements: We are indebted to the Texas A& M Brazos Computing Cluster and Institute of Developmentaland Molecular Biology for access to computingresources, and to professors David B. Dahl,Mohsen Pourahmadi, and Joel Zinn for helpful discussions.The infant microarray-metagenomics data wasprovided courtesy of Sharon M. Donovan, of the Divisionof Nutritional Sciences, U. of Illinois, Urbana, IL.This publication is based in part on work supported byAward No. KUS-C1-016-04, made by King AbdullahUniversity of Science and Technology (KAUST). This publication acknowledges KAUST support, but has no KAUST affiliated authors.",

year = "2013",

month = dec,

day = "18",

doi = "10.1137/1.9781611972825.98",

language = "English (US)",

isbn = "9781611972320",

booktitle = "Proceedings of the 2012 SIAM International Conference on Data Mining",

publisher = "Society for Industrial & Applied Mathematics (SIAM)",

}

TY - CHAP

T1 - Feature selection for high-dimensional integrated data

AU - Zheng, Charles

AU - Schwartz, Scott

AU - Chapkin, Robert S.

AU - Carroll, Raymond J.

AU - Ivanov, Ivan

N1 - KAUST Repository Item: Exported on 2020-10-01 Acknowledged KAUST grant number(s): KUS-C1-016-04 Acknowledgements: We are indebted to the Texas A& M Brazos Computing Cluster and Institute of Developmentaland Molecular Biology for access to computingresources, and to professors David B. Dahl,Mohsen Pourahmadi, and Joel Zinn for helpful discussions.The infant microarray-metagenomics data wasprovided courtesy of Sharon M. Donovan, of the Divisionof Nutritional Sciences, U. of Illinois, Urbana, IL.This publication is based in part on work supported byAward No. KUS-C1-016-04, made by King AbdullahUniversity of Science and Technology (KAUST). This publication acknowledges KAUST support, but has no KAUST affiliated authors.

PY - 2013/12/18

Y1 - 2013/12/18

N2 - Motivated by the problem of identifying correlations between genes or features of two related biological systems, we propose a model of feature selection in which only a subset of the predictors Xt are dependent on the multidimensional variate Y, and the remainder of the predictors constitute a “noise set” Xu independent of Y. Using Monte Carlo simulations, we investigated the relative performance of two methods: thresholding and singular-value decomposition, in combination with stochastic optimization to determine “empirical bounds” on the small-sample accuracy of an asymptotic approximation. We demonstrate utility of the thresholding and SVD feature selection methods to with respect to a recent infant intestinal gene expression and metagenomics dataset.

AB - Motivated by the problem of identifying correlations between genes or features of two related biological systems, we propose a model of feature selection in which only a subset of the predictors Xt are dependent on the multidimensional variate Y, and the remainder of the predictors constitute a “noise set” Xu independent of Y. Using Monte Carlo simulations, we investigated the relative performance of two methods: thresholding and singular-value decomposition, in combination with stochastic optimization to determine “empirical bounds” on the small-sample accuracy of an asymptotic approximation. We demonstrate utility of the thresholding and SVD feature selection methods to with respect to a recent infant intestinal gene expression and metagenomics dataset.

UR - http://hdl.handle.net/10754/598323

UR - https://epubs.siam.org/doi/10.1137/1.9781611972825.98

U2 - 10.1137/1.9781611972825.98

DO - 10.1137/1.9781611972825.98

M3 - Chapter

SN - 9781611972320

BT - Proceedings of the 2012 SIAM International Conference on Data Mining

PB - Society for Industrial & Applied Mathematics (SIAM)

ER -

Feature selection for high-dimensional integrated data

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this