Projecting partial least square and principle component regression across microarray studies

Chi Cheng Hunag, Shin Hsin Tu, Heng Hui Lien, Ching Shui Huang, Eric Y. Chuang, Liang Chuan Lai

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

Abstract

The study was to compare principle component (PC) versus partial least square (PLS) regression, the former unsupervised and the latter supervised gene component analysis, for highly complicated and correlated microarray gene expression profile. Projection of derived classifiers into independent samples for clinical phenotype prediction was evaluated as well. Previous studies had suggested that PLS might be superior to PC regression in the task of tumor classification since the covariance between predictive and respondent variables was maximized for latent factor extraction. We applied both algorithms for classifier construction and validated their prediction performance on independent microarray experiments. The statistical strategy could reduce high-dimensionality of microarray features and avoid the collinearity problem inherited in gene expression profiles. Proposed predictive model could discriminate breast cancers with positive and negative estrogen receptor status successfully and was feasible for both Taiwanese and Chinese females, both with the same Han Chinese ethnic origin.

Original languageEnglish
Title of host publication2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010
Pages506-511
Number of pages6
DOIs
Publication statusPublished - 2010
Event2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010 - HongKong, China
Duration: Dec 18 2010Dec 21 2010

Other

Other2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010
CountryChina
CityHongKong
Period12/18/1012/21/10

Fingerprint

Microarrays
Least-Squares Analysis
Transcriptome
Gene Components
Gene expression
Classifiers
Estrogen Receptors
Breast Neoplasms
Phenotype
Tumors
Genes
Neoplasms
Experiments
Surveys and Questionnaires

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics

Cite this

Hunag, C. C., Tu, S. H., Lien, H. H., Huang, C. S., Chuang, E. Y., & Lai, L. C. (2010). Projecting partial least square and principle component regression across microarray studies. In 2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010 (pp. 506-511). [5703853] https://doi.org/10.1109/BIBMW.2010.5703853

Projecting partial least square and principle component regression across microarray studies. / Hunag, Chi Cheng; Tu, Shin Hsin; Lien, Heng Hui; Huang, Ching Shui; Chuang, Eric Y.; Lai, Liang Chuan.

2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010. 2010. p. 506-511 5703853.

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

Hunag, CC, Tu, SH, Lien, HH, Huang, CS, Chuang, EY & Lai, LC 2010, Projecting partial least square and principle component regression across microarray studies. in 2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010., 5703853, pp. 506-511, 2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010, HongKong, China, 12/18/10. https://doi.org/10.1109/BIBMW.2010.5703853
Hunag CC, Tu SH, Lien HH, Huang CS, Chuang EY, Lai LC. Projecting partial least square and principle component regression across microarray studies. In 2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010. 2010. p. 506-511. 5703853 https://doi.org/10.1109/BIBMW.2010.5703853
Hunag, Chi Cheng ; Tu, Shin Hsin ; Lien, Heng Hui ; Huang, Ching Shui ; Chuang, Eric Y. ; Lai, Liang Chuan. / Projecting partial least square and principle component regression across microarray studies. 2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010. 2010. pp. 506-511
@inproceedings{0899997050ae4e778c6f9da57fe46ae5,
title = "Projecting partial least square and principle component regression across microarray studies",
abstract = "The study was to compare principle component (PC) versus partial least square (PLS) regression, the former unsupervised and the latter supervised gene component analysis, for highly complicated and correlated microarray gene expression profile. Projection of derived classifiers into independent samples for clinical phenotype prediction was evaluated as well. Previous studies had suggested that PLS might be superior to PC regression in the task of tumor classification since the covariance between predictive and respondent variables was maximized for latent factor extraction. We applied both algorithms for classifier construction and validated their prediction performance on independent microarray experiments. The statistical strategy could reduce high-dimensionality of microarray features and avoid the collinearity problem inherited in gene expression profiles. Proposed predictive model could discriminate breast cancers with positive and negative estrogen receptor status successfully and was feasible for both Taiwanese and Chinese females, both with the same Han Chinese ethnic origin.",
author = "Hunag, {Chi Cheng} and Tu, {Shin Hsin} and Lien, {Heng Hui} and Huang, {Ching Shui} and Chuang, {Eric Y.} and Lai, {Liang Chuan}",
year = "2010",
doi = "10.1109/BIBMW.2010.5703853",
language = "English",
isbn = "9781424483044",
pages = "506--511",
booktitle = "2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010",

}

TY - GEN

T1 - Projecting partial least square and principle component regression across microarray studies

AU - Hunag, Chi Cheng

AU - Tu, Shin Hsin

AU - Lien, Heng Hui

AU - Huang, Ching Shui

AU - Chuang, Eric Y.

AU - Lai, Liang Chuan

PY - 2010

Y1 - 2010

N2 - The study was to compare principle component (PC) versus partial least square (PLS) regression, the former unsupervised and the latter supervised gene component analysis, for highly complicated and correlated microarray gene expression profile. Projection of derived classifiers into independent samples for clinical phenotype prediction was evaluated as well. Previous studies had suggested that PLS might be superior to PC regression in the task of tumor classification since the covariance between predictive and respondent variables was maximized for latent factor extraction. We applied both algorithms for classifier construction and validated their prediction performance on independent microarray experiments. The statistical strategy could reduce high-dimensionality of microarray features and avoid the collinearity problem inherited in gene expression profiles. Proposed predictive model could discriminate breast cancers with positive and negative estrogen receptor status successfully and was feasible for both Taiwanese and Chinese females, both with the same Han Chinese ethnic origin.

AB - The study was to compare principle component (PC) versus partial least square (PLS) regression, the former unsupervised and the latter supervised gene component analysis, for highly complicated and correlated microarray gene expression profile. Projection of derived classifiers into independent samples for clinical phenotype prediction was evaluated as well. Previous studies had suggested that PLS might be superior to PC regression in the task of tumor classification since the covariance between predictive and respondent variables was maximized for latent factor extraction. We applied both algorithms for classifier construction and validated their prediction performance on independent microarray experiments. The statistical strategy could reduce high-dimensionality of microarray features and avoid the collinearity problem inherited in gene expression profiles. Proposed predictive model could discriminate breast cancers with positive and negative estrogen receptor status successfully and was feasible for both Taiwanese and Chinese females, both with the same Han Chinese ethnic origin.

UR - http://www.scopus.com/inward/record.url?scp=79952028829&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952028829&partnerID=8YFLogxK

U2 - 10.1109/BIBMW.2010.5703853

DO - 10.1109/BIBMW.2010.5703853

M3 - Conference contribution

AN - SCOPUS:79952028829

SN - 9781424483044

SP - 506

EP - 511

BT - 2010 IEEE International Conference on Bioinformatics and Biomedicine Workshops, BIBMW 2010

ER -