Microarray classification from several twogene expression comparisons

Donald Geman, Bahman Afsari, Aik Choon Tan, Daniel Q. Naiman

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

Abstract

We describe our contribution to the ICMLA2008 "Automated Micro-Array Classification Challenge". The design of our classifier is motivated by the special scenario encountered in molecular cancer classification based on the mRNA concentrations provided by gene microarray data. Our classifier is rank-based; it only depends on expression comparisons among selected pairs of genes. Such comparisons are invariant to most of the transformations involved in preprocessing and normalization. Every pair of genes determines a binary classifier - choose the class for which the observed ordering is most likely. Pairs are scored by maximizing accuracy. In our k-TSP (k-disjoint Top Scoring Pairs) classifier, k disjoint pairs of genes are learned from training data; the discriminant function is simply the difference in the number of votes for the two classes. This rule involves exactly 2k genes, is readily interpretable, and provides some state-of-the-art results in cancer diagnosis and prognosis for small values of k, even k=1.

Original languageEnglish (US)
Title of host publicationProceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008
Pages583-585
Number of pages3
DOIs
StatePublished - Dec 1 2008
Externally publishedYes
Event7th International Conference on Machine Learning and Applications, ICMLA 2008 - San Diego, CA, United States
Duration: Dec 11 2008Dec 13 2008

Other

Other7th International Conference on Machine Learning and Applications, ICMLA 2008
CountryUnited States
CitySan Diego, CA
Period12/11/0812/13/08

Fingerprint

Microarrays
Genes
Classifiers

Keywords

  • Cancer diagnosis
  • Gene expression
  • Maximum likelihood
  • Molecular classification
  • Rank-based

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Software

Cite this

Geman, D., Afsari, B., Tan, A. C., & Naiman, D. Q. (2008). Microarray classification from several twogene expression comparisons. In Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008 (pp. 583-585). [4725033] https://doi.org/10.1109/ICMLA.2008.152

Microarray classification from several twogene expression comparisons. / Geman, Donald; Afsari, Bahman; Tan, Aik Choon; Naiman, Daniel Q.

Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008. 2008. p. 583-585 4725033.

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

Geman, D, Afsari, B, Tan, AC & Naiman, DQ 2008, Microarray classification from several twogene expression comparisons. in Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008., 4725033, pp. 583-585, 7th International Conference on Machine Learning and Applications, ICMLA 2008, San Diego, CA, United States, 12/11/08. https://doi.org/10.1109/ICMLA.2008.152
Geman D, Afsari B, Tan AC, Naiman DQ. Microarray classification from several twogene expression comparisons. In Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008. 2008. p. 583-585. 4725033 https://doi.org/10.1109/ICMLA.2008.152
Geman, Donald ; Afsari, Bahman ; Tan, Aik Choon ; Naiman, Daniel Q. / Microarray classification from several twogene expression comparisons. Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008. 2008. pp. 583-585
@inproceedings{63d56ed3910c4fb7be5e433345907ac7,
title = "Microarray classification from several twogene expression comparisons",
abstract = "We describe our contribution to the ICMLA2008 {"}Automated Micro-Array Classification Challenge{"}. The design of our classifier is motivated by the special scenario encountered in molecular cancer classification based on the mRNA concentrations provided by gene microarray data. Our classifier is rank-based; it only depends on expression comparisons among selected pairs of genes. Such comparisons are invariant to most of the transformations involved in preprocessing and normalization. Every pair of genes determines a binary classifier - choose the class for which the observed ordering is most likely. Pairs are scored by maximizing accuracy. In our k-TSP (k-disjoint Top Scoring Pairs) classifier, k disjoint pairs of genes are learned from training data; the discriminant function is simply the difference in the number of votes for the two classes. This rule involves exactly 2k genes, is readily interpretable, and provides some state-of-the-art results in cancer diagnosis and prognosis for small values of k, even k=1.",
keywords = "Cancer diagnosis, Gene expression, Maximum likelihood, Molecular classification, Rank-based",
author = "Donald Geman and Bahman Afsari and Tan, {Aik Choon} and Naiman, {Daniel Q.}",
year = "2008",
month = "12",
day = "1",
doi = "10.1109/ICMLA.2008.152",
language = "English (US)",
isbn = "9780769534954",
pages = "583--585",
booktitle = "Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008",

}

TY - GEN

T1 - Microarray classification from several twogene expression comparisons

AU - Geman, Donald

AU - Afsari, Bahman

AU - Tan, Aik Choon

AU - Naiman, Daniel Q.

PY - 2008/12/1

Y1 - 2008/12/1

N2 - We describe our contribution to the ICMLA2008 "Automated Micro-Array Classification Challenge". The design of our classifier is motivated by the special scenario encountered in molecular cancer classification based on the mRNA concentrations provided by gene microarray data. Our classifier is rank-based; it only depends on expression comparisons among selected pairs of genes. Such comparisons are invariant to most of the transformations involved in preprocessing and normalization. Every pair of genes determines a binary classifier - choose the class for which the observed ordering is most likely. Pairs are scored by maximizing accuracy. In our k-TSP (k-disjoint Top Scoring Pairs) classifier, k disjoint pairs of genes are learned from training data; the discriminant function is simply the difference in the number of votes for the two classes. This rule involves exactly 2k genes, is readily interpretable, and provides some state-of-the-art results in cancer diagnosis and prognosis for small values of k, even k=1.

AB - We describe our contribution to the ICMLA2008 "Automated Micro-Array Classification Challenge". The design of our classifier is motivated by the special scenario encountered in molecular cancer classification based on the mRNA concentrations provided by gene microarray data. Our classifier is rank-based; it only depends on expression comparisons among selected pairs of genes. Such comparisons are invariant to most of the transformations involved in preprocessing and normalization. Every pair of genes determines a binary classifier - choose the class for which the observed ordering is most likely. Pairs are scored by maximizing accuracy. In our k-TSP (k-disjoint Top Scoring Pairs) classifier, k disjoint pairs of genes are learned from training data; the discriminant function is simply the difference in the number of votes for the two classes. This rule involves exactly 2k genes, is readily interpretable, and provides some state-of-the-art results in cancer diagnosis and prognosis for small values of k, even k=1.

KW - Cancer diagnosis

KW - Gene expression

KW - Maximum likelihood

KW - Molecular classification

KW - Rank-based

UR - http://www.scopus.com/inward/record.url?scp=60649120943&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=60649120943&partnerID=8YFLogxK

U2 - 10.1109/ICMLA.2008.152

DO - 10.1109/ICMLA.2008.152

M3 - Conference contribution

AN - SCOPUS:60649120943

SN - 9780769534954

SP - 583

EP - 585

BT - Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008

ER -