Tumor classification by combining PNN classifier ensemble with neighborhood rough set based gene reduction

Shu Lin Wang, Xueling Li, Shanwen Zhang, Jie Gui, De Shuang Huang

Research output: Contribution to journalArticle

73 Citations (Scopus)

Abstract

Since Golub applied gene expression profiles (GEP) to the molecular classification of tumor subtypes for more accurately and reliably clinical diagnosis, a number of studies on GEP-based tumor classification have been done. However, the challenges from high dimension and small sample size of tumor dataset still exist. This paper presents a new tumor classification approach based on an ensemble of probabilistic neural network (PNN) and neighborhood rough set model based gene reduction. Informative genes were initially selected by gene ranking based on an iterative search margin algorithm and then were further refined by gene reduction to select many minimum gene subsets. Finally, the candidate base PNN classifiers trained by each of the selected gene subsets were integrated by majority voting strategy to construct an ensemble classifier. Experiments on tumor datasets showed that this approach can obtain both high and stable classification performance, which is not too sensitive to the number of initially selected genes and competitive to most existing methods. Additionally, the classification results can be cross-verified in a single biomedical experiment by the selected gene subsets, and biologically experimental results also proved that the genes included in the selected gene subsets are functionally related to carcinogenesis, indicating that the performance obtained by the proposed method is convincing.

Original languageEnglish (US)
Pages (from-to)179-189
Number of pages11
JournalComputers in Biology and Medicine
Volume40
Issue number2
DOIs
StatePublished - Feb 2010
Externally publishedYes

Fingerprint

Tumors
Classifiers
Genes
Neural networks
Neoplasms
Transcriptome
Gene expression
Politics
Sample Size
Carcinogenesis
Experiments

Keywords

  • Biological data mining
  • Gene expression profiles
  • Gene selection
  • Neighborhood rough set model
  • Probabilistic neural network ensemble
  • Tumor classification

ASJC Scopus subject areas

  • Computer Science Applications
  • Health Informatics

Cite this

Tumor classification by combining PNN classifier ensemble with neighborhood rough set based gene reduction. / Wang, Shu Lin; Li, Xueling; Zhang, Shanwen; Gui, Jie; Huang, De Shuang.

In: Computers in Biology and Medicine, Vol. 40, No. 2, 02.2010, p. 179-189.

Research output: Contribution to journalArticle

Wang, Shu Lin ; Li, Xueling ; Zhang, Shanwen ; Gui, Jie ; Huang, De Shuang. / Tumor classification by combining PNN classifier ensemble with neighborhood rough set based gene reduction. In: Computers in Biology and Medicine. 2010 ; Vol. 40, No. 2. pp. 179-189.
@article{44b18605ba1c46cf932923bc713f7249,
title = "Tumor classification by combining PNN classifier ensemble with neighborhood rough set based gene reduction",
abstract = "Since Golub applied gene expression profiles (GEP) to the molecular classification of tumor subtypes for more accurately and reliably clinical diagnosis, a number of studies on GEP-based tumor classification have been done. However, the challenges from high dimension and small sample size of tumor dataset still exist. This paper presents a new tumor classification approach based on an ensemble of probabilistic neural network (PNN) and neighborhood rough set model based gene reduction. Informative genes were initially selected by gene ranking based on an iterative search margin algorithm and then were further refined by gene reduction to select many minimum gene subsets. Finally, the candidate base PNN classifiers trained by each of the selected gene subsets were integrated by majority voting strategy to construct an ensemble classifier. Experiments on tumor datasets showed that this approach can obtain both high and stable classification performance, which is not too sensitive to the number of initially selected genes and competitive to most existing methods. Additionally, the classification results can be cross-verified in a single biomedical experiment by the selected gene subsets, and biologically experimental results also proved that the genes included in the selected gene subsets are functionally related to carcinogenesis, indicating that the performance obtained by the proposed method is convincing.",
keywords = "Biological data mining, Gene expression profiles, Gene selection, Neighborhood rough set model, Probabilistic neural network ensemble, Tumor classification",
author = "Wang, {Shu Lin} and Xueling Li and Shanwen Zhang and Jie Gui and Huang, {De Shuang}",
year = "2010",
month = "2",
doi = "10.1016/j.compbiomed.2009.11.014",
language = "English (US)",
volume = "40",
pages = "179--189",
journal = "Computers in Biology and Medicine",
issn = "0010-4825",
publisher = "Elsevier Limited",
number = "2",

}

TY - JOUR

T1 - Tumor classification by combining PNN classifier ensemble with neighborhood rough set based gene reduction

AU - Wang, Shu Lin

AU - Li, Xueling

AU - Zhang, Shanwen

AU - Gui, Jie

AU - Huang, De Shuang

PY - 2010/2

Y1 - 2010/2

N2 - Since Golub applied gene expression profiles (GEP) to the molecular classification of tumor subtypes for more accurately and reliably clinical diagnosis, a number of studies on GEP-based tumor classification have been done. However, the challenges from high dimension and small sample size of tumor dataset still exist. This paper presents a new tumor classification approach based on an ensemble of probabilistic neural network (PNN) and neighborhood rough set model based gene reduction. Informative genes were initially selected by gene ranking based on an iterative search margin algorithm and then were further refined by gene reduction to select many minimum gene subsets. Finally, the candidate base PNN classifiers trained by each of the selected gene subsets were integrated by majority voting strategy to construct an ensemble classifier. Experiments on tumor datasets showed that this approach can obtain both high and stable classification performance, which is not too sensitive to the number of initially selected genes and competitive to most existing methods. Additionally, the classification results can be cross-verified in a single biomedical experiment by the selected gene subsets, and biologically experimental results also proved that the genes included in the selected gene subsets are functionally related to carcinogenesis, indicating that the performance obtained by the proposed method is convincing.

AB - Since Golub applied gene expression profiles (GEP) to the molecular classification of tumor subtypes for more accurately and reliably clinical diagnosis, a number of studies on GEP-based tumor classification have been done. However, the challenges from high dimension and small sample size of tumor dataset still exist. This paper presents a new tumor classification approach based on an ensemble of probabilistic neural network (PNN) and neighborhood rough set model based gene reduction. Informative genes were initially selected by gene ranking based on an iterative search margin algorithm and then were further refined by gene reduction to select many minimum gene subsets. Finally, the candidate base PNN classifiers trained by each of the selected gene subsets were integrated by majority voting strategy to construct an ensemble classifier. Experiments on tumor datasets showed that this approach can obtain both high and stable classification performance, which is not too sensitive to the number of initially selected genes and competitive to most existing methods. Additionally, the classification results can be cross-verified in a single biomedical experiment by the selected gene subsets, and biologically experimental results also proved that the genes included in the selected gene subsets are functionally related to carcinogenesis, indicating that the performance obtained by the proposed method is convincing.

KW - Biological data mining

KW - Gene expression profiles

KW - Gene selection

KW - Neighborhood rough set model

KW - Probabilistic neural network ensemble

KW - Tumor classification

UR - http://www.scopus.com/inward/record.url?scp=77649237177&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77649237177&partnerID=8YFLogxK

U2 - 10.1016/j.compbiomed.2009.11.014

DO - 10.1016/j.compbiomed.2009.11.014

M3 - Article

VL - 40

SP - 179

EP - 189

JO - Computers in Biology and Medicine

JF - Computers in Biology and Medicine

SN - 0010-4825

IS - 2

ER -