Comparative Data Mining for Microarrays: A Case Study Based on Multiple Myeloma

David Page; Fenghuang Zhan; James Cussens; Michael Waddell; Johanna Hardin; Bart Barlogie; John Shaughnessy, Jr.

Comparative Data Mining for Microarrays: A Case Study Based on Multiple Myeloma

David Page, Fenghuang Zhan, James Cussens, Michael Waddell, Johanna Hardin, Bart Barlogie, John Shaughnessy, Jr.

Computer Science

Research output: Working paper

Abstract

Supervised machine learning and data mining tools have become popular for the analysis of gene expression microarray data. They have the potential to uncover new therapeutic targets for diseases, to predict how patients will respond to specific treatments, and to uncover regulatory relationships among genes in normal and disease situations. Comparative experiments are needed to identify the advantages of the leading supervised learning algorithms for microarray data, as well as to give direction in methodological decisions. This paper compares support vector machines, Bayesian networks, decision trees, boosted decision trees, and voting (ensembles of decision stumps) on a new microarray data set for cancer with over 100 samples. The paper provides evidence for several important lessons for mining microarray data, including: (1) Bayes nets and ensembles perform at least as well as other approaches but arguably provide more direct insight; (2) the common practice of throwing out low or negative average differences, or those accompanied by an absent call, is a mistake; (3) looking for consistent differences in expression may be more important than large differences.

Original language	English
Publication status	Published - 1 Nov 2002

Access to Document

ftp://ftp.cs.wisc.edu/pub/tech-reports/reports/2002/tr1453.ps.Z

Cite this

@techreport{e28264f4b21b4dc7a603dc9ec4cc080f,

title = "Comparative Data Mining for Microarrays: A Case Study Based on Multiple Myeloma",

abstract = "Supervised machine learning and data mining tools have become popular for the analysis of gene expression microarray data. They have the potential to uncover new therapeutic targets for diseases, to predict how patients will respond to specific treatments, and to uncover regulatory relationships among genes in normal and disease situations. Comparative experiments are needed to identify the advantages of the leading supervised learning algorithms for microarray data, as well as to give direction in methodological decisions. This paper compares support vector machines, Bayesian networks, decision trees, boosted decision trees, and voting (ensembles of decision stumps) on a new microarray data set for cancer with over 100 samples. The paper provides evidence for several important lessons for mining microarray data, including: (1) Bayes nets and ensembles perform at least as well as other approaches but arguably provide more direct insight; (2) the common practice of throwing out low or negative average differences, or those accompanied by an absent call, is a mistake; (3) looking for consistent differences in expression may be more important than large differences.",

author = "David Page and Fenghuang Zhan and James Cussens and Michael Waddell and Johanna Hardin and Bart Barlogie and {Shaughnessy, Jr.}, John",

year = "2002",

month = nov,

day = "1",

language = "English",

type = "WorkingPaper",

}

TY - UNPB

T1 - Comparative Data Mining for Microarrays: A Case Study Based on Multiple Myeloma

AU - Page, David

AU - Zhan, Fenghuang

AU - Cussens, James

AU - Waddell, Michael

AU - Hardin, Johanna

AU - Barlogie, Bart

AU - Shaughnessy, Jr., John

PY - 2002/11/1

Y1 - 2002/11/1

N2 - Supervised machine learning and data mining tools have become popular for the analysis of gene expression microarray data. They have the potential to uncover new therapeutic targets for diseases, to predict how patients will respond to specific treatments, and to uncover regulatory relationships among genes in normal and disease situations. Comparative experiments are needed to identify the advantages of the leading supervised learning algorithms for microarray data, as well as to give direction in methodological decisions. This paper compares support vector machines, Bayesian networks, decision trees, boosted decision trees, and voting (ensembles of decision stumps) on a new microarray data set for cancer with over 100 samples. The paper provides evidence for several important lessons for mining microarray data, including: (1) Bayes nets and ensembles perform at least as well as other approaches but arguably provide more direct insight; (2) the common practice of throwing out low or negative average differences, or those accompanied by an absent call, is a mistake; (3) looking for consistent differences in expression may be more important than large differences.

AB - Supervised machine learning and data mining tools have become popular for the analysis of gene expression microarray data. They have the potential to uncover new therapeutic targets for diseases, to predict how patients will respond to specific treatments, and to uncover regulatory relationships among genes in normal and disease situations. Comparative experiments are needed to identify the advantages of the leading supervised learning algorithms for microarray data, as well as to give direction in methodological decisions. This paper compares support vector machines, Bayesian networks, decision trees, boosted decision trees, and voting (ensembles of decision stumps) on a new microarray data set for cancer with over 100 samples. The paper provides evidence for several important lessons for mining microarray data, including: (1) Bayes nets and ensembles perform at least as well as other approaches but arguably provide more direct insight; (2) the common practice of throwing out low or negative average differences, or those accompanied by an absent call, is a mistake; (3) looking for consistent differences in expression may be more important than large differences.

M3 - Working paper

BT - Comparative Data Mining for Microarrays: A Case Study Based on Multiple Myeloma

ER -