An integrated system for class prediction using gene expression profiling

Dechang Chen, Donald D. Hua, Zhenqiu Liu, Zhi Fu Cheng

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

1 Citation (Scopus)

Abstract

Motivation: Gene expression profiles have been successfully applied to class prediction. Due to a large number of genes (features) and a small number of samples in gene expression data, feature selection is essential when performing the prediction task. Many methods have been proposed to select features in microarray data analysis, but there is no unique method which performs uniformly well for all the learning algorithms. It is then practical to find a feature selction method and a learning algorithm that give superior performance. Results: In this paper, we present an integrated scheme to perform the task of class prediction based on gene expression profiles. The scheme incorporates a simple novel feature selection procedure into naive Bayes models. Each selected gene has a high score of discriminatory power determined by the Brown-Forsythe test statistic. Any pair of selected genes have a low correlation. This facilitates the use of the conditional independence among genes assumed by the naive Bayes models. To demonstrate the effectiveness, the proposed scheme was applied to three commonly used expression data sets COLON, OVARIAN, and LEUKEMIA. The results show that the numbers of misclassified samples are 0, 0, and 4, respectively.

Original languageEnglish (US)
Title of host publication2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)
Pages1023-1028
Number of pages6
StatePublished - Dec 1 2004
Event8th International Conference on Control, Automation, Robotics and Vision (ICARCV) - Kunming, China
Duration: Dec 6 2004Dec 9 2004

Publication series

Name2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)
Volume2

Other

Other8th International Conference on Control, Automation, Robotics and Vision (ICARCV)
CountryChina
CityKunming
Period12/6/0412/9/04

Fingerprint

Gene expression
Genes
Learning algorithms
Feature extraction
Microarrays
Statistics

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Chen, D., Hua, D. D., Liu, Z., & Cheng, Z. F. (2004). An integrated system for class prediction using gene expression profiling. In 2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV) (pp. 1023-1028). (2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV); Vol. 2).
Chen, Dechang ; Hua, Donald D. ; Liu, Zhenqiu ; Cheng, Zhi Fu. / An integrated system for class prediction using gene expression profiling. 2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV). 2004. pp. 1023-1028 (2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)).
@inproceedings{c5a6d32644ca48f2a3ef4571de3569f4,
title = "An integrated system for class prediction using gene expression profiling",
abstract = "Motivation: Gene expression profiles have been successfully applied to class prediction. Due to a large number of genes (features) and a small number of samples in gene expression data, feature selection is essential when performing the prediction task. Many methods have been proposed to select features in microarray data analysis, but there is no unique method which performs uniformly well for all the learning algorithms. It is then practical to find a feature selction method and a learning algorithm that give superior performance. Results: In this paper, we present an integrated scheme to perform the task of class prediction based on gene expression profiles. The scheme incorporates a simple novel feature selection procedure into naive Bayes models. Each selected gene has a high score of discriminatory power determined by the Brown-Forsythe test statistic. Any pair of selected genes have a low correlation. This facilitates the use of the conditional independence among genes assumed by the naive Bayes models. To demonstrate the effectiveness, the proposed scheme was applied to three commonly used expression data sets COLON, OVARIAN, and LEUKEMIA. The results show that the numbers of misclassified samples are 0, 0, and 4, respectively.",
author = "Dechang Chen and Hua, {Donald D.} and Zhenqiu Liu and Cheng, {Zhi Fu}",
year = "2004",
month = "12",
day = "1",
language = "English (US)",
isbn = "0780386531",
series = "2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)",
pages = "1023--1028",
booktitle = "2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)",

}

Chen, D, Hua, DD, Liu, Z & Cheng, ZF 2004, An integrated system for class prediction using gene expression profiling. in 2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV). 2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV), vol. 2, pp. 1023-1028, 8th International Conference on Control, Automation, Robotics and Vision (ICARCV), Kunming, China, 12/6/04.

An integrated system for class prediction using gene expression profiling. / Chen, Dechang; Hua, Donald D.; Liu, Zhenqiu; Cheng, Zhi Fu.

2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV). 2004. p. 1023-1028 (2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV); Vol. 2).

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

TY - GEN

T1 - An integrated system for class prediction using gene expression profiling

AU - Chen, Dechang

AU - Hua, Donald D.

AU - Liu, Zhenqiu

AU - Cheng, Zhi Fu

PY - 2004/12/1

Y1 - 2004/12/1

N2 - Motivation: Gene expression profiles have been successfully applied to class prediction. Due to a large number of genes (features) and a small number of samples in gene expression data, feature selection is essential when performing the prediction task. Many methods have been proposed to select features in microarray data analysis, but there is no unique method which performs uniformly well for all the learning algorithms. It is then practical to find a feature selction method and a learning algorithm that give superior performance. Results: In this paper, we present an integrated scheme to perform the task of class prediction based on gene expression profiles. The scheme incorporates a simple novel feature selection procedure into naive Bayes models. Each selected gene has a high score of discriminatory power determined by the Brown-Forsythe test statistic. Any pair of selected genes have a low correlation. This facilitates the use of the conditional independence among genes assumed by the naive Bayes models. To demonstrate the effectiveness, the proposed scheme was applied to three commonly used expression data sets COLON, OVARIAN, and LEUKEMIA. The results show that the numbers of misclassified samples are 0, 0, and 4, respectively.

AB - Motivation: Gene expression profiles have been successfully applied to class prediction. Due to a large number of genes (features) and a small number of samples in gene expression data, feature selection is essential when performing the prediction task. Many methods have been proposed to select features in microarray data analysis, but there is no unique method which performs uniformly well for all the learning algorithms. It is then practical to find a feature selction method and a learning algorithm that give superior performance. Results: In this paper, we present an integrated scheme to perform the task of class prediction based on gene expression profiles. The scheme incorporates a simple novel feature selection procedure into naive Bayes models. Each selected gene has a high score of discriminatory power determined by the Brown-Forsythe test statistic. Any pair of selected genes have a low correlation. This facilitates the use of the conditional independence among genes assumed by the naive Bayes models. To demonstrate the effectiveness, the proposed scheme was applied to three commonly used expression data sets COLON, OVARIAN, and LEUKEMIA. The results show that the numbers of misclassified samples are 0, 0, and 4, respectively.

UR - http://www.scopus.com/inward/record.url?scp=21244441090&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=21244441090&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:21244441090

SN - 0780386531

T3 - 2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)

SP - 1023

EP - 1028

BT - 2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)

ER -

Chen D, Hua DD, Liu Z, Cheng ZF. An integrated system for class prediction using gene expression profiling. In 2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV). 2004. p. 1023-1028. (2004 8th International Conference on Control, Automation, Robotics and Vision (ICARCV)).