Factor analysis for cross-platform tumor classification based on gene expression profiles

Shu Lin Wang, Jie Gui, Xueling Li

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

Previous studies on tumor classification based on feature extraction from gene expression profiles (GEP) were proven to be effective, but some of such methods lack biomedical meaning to some extent. To deal with this problem, we proposed a novel feature extraction method whose experimental results are of biomedical interpretability and helpful for gaining insight into the structure analysis of gene expression dataset. This method first applied rank sum test to roughly select a set of informative genes and then adopted factor analysis to extract latent factors for tumor classification. Experiments on three pairs of cross-platform tumor datasets indicated that the proposed method can obviously improve the performance of cross-platform classification and only several latent factors, which can represent a large number of informative genes, would obtain very high predictive accuracy on test set. The results also suggested that the classification model trained on one dataset can successfully predict another tumor dataset with the same tumor subtype obtained on different experimental platforms.

Original languageEnglish (US)
Pages (from-to)243-258
Number of pages16
JournalJournal of Circuits, Systems and Computers
Volume19
Issue number1
DOIs
StatePublished - Feb 2010
Externally publishedYes

Fingerprint

Factor analysis
Gene expression
Tumors
Feature extraction
Genes
Experiments

Keywords

  • Cross-platform analysis
  • Factor analysis
  • Feature extraction
  • Gene expression profiles
  • Tumor classification

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Hardware and Architecture

Cite this

Factor analysis for cross-platform tumor classification based on gene expression profiles. / Wang, Shu Lin; Gui, Jie; Li, Xueling.

In: Journal of Circuits, Systems and Computers, Vol. 19, No. 1, 02.2010, p. 243-258.

Research output: Contribution to journalArticle

@article{fc69fc4ac5ed45ebaab8e319509eb1af,
title = "Factor analysis for cross-platform tumor classification based on gene expression profiles",
abstract = "Previous studies on tumor classification based on feature extraction from gene expression profiles (GEP) were proven to be effective, but some of such methods lack biomedical meaning to some extent. To deal with this problem, we proposed a novel feature extraction method whose experimental results are of biomedical interpretability and helpful for gaining insight into the structure analysis of gene expression dataset. This method first applied rank sum test to roughly select a set of informative genes and then adopted factor analysis to extract latent factors for tumor classification. Experiments on three pairs of cross-platform tumor datasets indicated that the proposed method can obviously improve the performance of cross-platform classification and only several latent factors, which can represent a large number of informative genes, would obtain very high predictive accuracy on test set. The results also suggested that the classification model trained on one dataset can successfully predict another tumor dataset with the same tumor subtype obtained on different experimental platforms.",
keywords = "Cross-platform analysis, Factor analysis, Feature extraction, Gene expression profiles, Tumor classification",
author = "Wang, {Shu Lin} and Jie Gui and Xueling Li",
year = "2010",
month = "2",
doi = "10.1142/S0218126610006074",
language = "English (US)",
volume = "19",
pages = "243--258",
journal = "Journal of Circuits, Systems and Computers",
issn = "0218-1266",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "1",

}

TY - JOUR

T1 - Factor analysis for cross-platform tumor classification based on gene expression profiles

AU - Wang, Shu Lin

AU - Gui, Jie

AU - Li, Xueling

PY - 2010/2

Y1 - 2010/2

N2 - Previous studies on tumor classification based on feature extraction from gene expression profiles (GEP) were proven to be effective, but some of such methods lack biomedical meaning to some extent. To deal with this problem, we proposed a novel feature extraction method whose experimental results are of biomedical interpretability and helpful for gaining insight into the structure analysis of gene expression dataset. This method first applied rank sum test to roughly select a set of informative genes and then adopted factor analysis to extract latent factors for tumor classification. Experiments on three pairs of cross-platform tumor datasets indicated that the proposed method can obviously improve the performance of cross-platform classification and only several latent factors, which can represent a large number of informative genes, would obtain very high predictive accuracy on test set. The results also suggested that the classification model trained on one dataset can successfully predict another tumor dataset with the same tumor subtype obtained on different experimental platforms.

AB - Previous studies on tumor classification based on feature extraction from gene expression profiles (GEP) were proven to be effective, but some of such methods lack biomedical meaning to some extent. To deal with this problem, we proposed a novel feature extraction method whose experimental results are of biomedical interpretability and helpful for gaining insight into the structure analysis of gene expression dataset. This method first applied rank sum test to roughly select a set of informative genes and then adopted factor analysis to extract latent factors for tumor classification. Experiments on three pairs of cross-platform tumor datasets indicated that the proposed method can obviously improve the performance of cross-platform classification and only several latent factors, which can represent a large number of informative genes, would obtain very high predictive accuracy on test set. The results also suggested that the classification model trained on one dataset can successfully predict another tumor dataset with the same tumor subtype obtained on different experimental platforms.

KW - Cross-platform analysis

KW - Factor analysis

KW - Feature extraction

KW - Gene expression profiles

KW - Tumor classification

UR - http://www.scopus.com/inward/record.url?scp=77951618458&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951618458&partnerID=8YFLogxK

U2 - 10.1142/S0218126610006074

DO - 10.1142/S0218126610006074

M3 - Article

AN - SCOPUS:77951618458

VL - 19

SP - 243

EP - 258

JO - Journal of Circuits, Systems and Computers

JF - Journal of Circuits, Systems and Computers

SN - 0218-1266

IS - 1

ER -