Combining results of microarray experiments: A rank aggregation approach

Robert P. DeConde, Sarah Hawley, Seth Falcon, Nigel Clegg, Beatrice Knudsen, Ruth Etzioni

Research output: Contribution to journalArticle

99 Citations (Scopus)

Abstract

As technology for microarray analysis becomes widespread, it is becoming increasingly important to be able to compare and combine the results of experiments that explore the same scientific question. In this article, we present a rank-aggregation approach for combining results from several microarray studies. The motivation for this approach is twofold; first, the final results of microarray studies are typically expressed as lists of genes, rank-ordered by a measure of the strength of evidence that they are functionally involved in the disease process, and second, using the information on this rank-ordered metric means that we do not have to concern ourselves with data on the actual expression levels, which may not be comparable across experiments. Our approach draws on methods for combining top-k lists from the computer science literature on meta-search. The meta-search problem shares several important features with that of combining microarray experiments, including the fact that there are typically few lists with many elements and the elements may not be common to all lists. We implement two meta-search algorithms, which use a Markov chain framework to convert pairwise preferences between list elements into a stationary distribution that represents an aggregate ranking (Dwork et al, 2001). We explore the behavior of the algorithms in hypothetical examples and a simulated dataset and compare their performance with that of an algorithm based on the order-statistics model of Thurstone (Thurstone, 1927). We apply all three algorithms to aggregate the results of five microarray studies of prostate cancer.

Original languageEnglish (US)
Article number15
JournalStatistical Applications in Genetics and Molecular Biology
Volume5
Issue number1
DOIs
StatePublished - Jan 1 2006
Externally publishedYes

Fingerprint

Rank Aggregation
Microarrays
Microarray
Agglomeration
Experiment
Experiments
Microarray Analysis
Markov Chains
Prostate Cancer
Search Problems
Stationary Distribution
Order Statistics
Search Algorithm
Convert
Pairwise
Markov chain
Prostatic Neoplasms
Ranking
Computer Science
Markov processes

Keywords

  • Markov chains
  • Meta-analysis
  • Microarrays
  • Order-statistic models
  • Rank aggregation

ASJC Scopus subject areas

  • Statistics and Probability
  • Molecular Biology
  • Genetics
  • Computational Mathematics

Cite this

Combining results of microarray experiments : A rank aggregation approach. / DeConde, Robert P.; Hawley, Sarah; Falcon, Seth; Clegg, Nigel; Knudsen, Beatrice; Etzioni, Ruth.

In: Statistical Applications in Genetics and Molecular Biology, Vol. 5, No. 1, 15, 01.01.2006.

Research output: Contribution to journalArticle

DeConde, Robert P. ; Hawley, Sarah ; Falcon, Seth ; Clegg, Nigel ; Knudsen, Beatrice ; Etzioni, Ruth. / Combining results of microarray experiments : A rank aggregation approach. In: Statistical Applications in Genetics and Molecular Biology. 2006 ; Vol. 5, No. 1.
@article{bbaa9544c1bf433f8fee1a13a81f62a0,
title = "Combining results of microarray experiments: A rank aggregation approach",
abstract = "As technology for microarray analysis becomes widespread, it is becoming increasingly important to be able to compare and combine the results of experiments that explore the same scientific question. In this article, we present a rank-aggregation approach for combining results from several microarray studies. The motivation for this approach is twofold; first, the final results of microarray studies are typically expressed as lists of genes, rank-ordered by a measure of the strength of evidence that they are functionally involved in the disease process, and second, using the information on this rank-ordered metric means that we do not have to concern ourselves with data on the actual expression levels, which may not be comparable across experiments. Our approach draws on methods for combining top-k lists from the computer science literature on meta-search. The meta-search problem shares several important features with that of combining microarray experiments, including the fact that there are typically few lists with many elements and the elements may not be common to all lists. We implement two meta-search algorithms, which use a Markov chain framework to convert pairwise preferences between list elements into a stationary distribution that represents an aggregate ranking (Dwork et al, 2001). We explore the behavior of the algorithms in hypothetical examples and a simulated dataset and compare their performance with that of an algorithm based on the order-statistics model of Thurstone (Thurstone, 1927). We apply all three algorithms to aggregate the results of five microarray studies of prostate cancer.",
keywords = "Markov chains, Meta-analysis, Microarrays, Order-statistic models, Rank aggregation",
author = "DeConde, {Robert P.} and Sarah Hawley and Seth Falcon and Nigel Clegg and Beatrice Knudsen and Ruth Etzioni",
year = "2006",
month = "1",
day = "1",
doi = "10.2202/1544-6115.1204",
language = "English (US)",
volume = "5",
journal = "Statistical Applications in Genetics and Molecular Biology",
issn = "1544-6115",
publisher = "Berkeley Electronic Press",
number = "1",

}

TY - JOUR

T1 - Combining results of microarray experiments

T2 - A rank aggregation approach

AU - DeConde, Robert P.

AU - Hawley, Sarah

AU - Falcon, Seth

AU - Clegg, Nigel

AU - Knudsen, Beatrice

AU - Etzioni, Ruth

PY - 2006/1/1

Y1 - 2006/1/1

N2 - As technology for microarray analysis becomes widespread, it is becoming increasingly important to be able to compare and combine the results of experiments that explore the same scientific question. In this article, we present a rank-aggregation approach for combining results from several microarray studies. The motivation for this approach is twofold; first, the final results of microarray studies are typically expressed as lists of genes, rank-ordered by a measure of the strength of evidence that they are functionally involved in the disease process, and second, using the information on this rank-ordered metric means that we do not have to concern ourselves with data on the actual expression levels, which may not be comparable across experiments. Our approach draws on methods for combining top-k lists from the computer science literature on meta-search. The meta-search problem shares several important features with that of combining microarray experiments, including the fact that there are typically few lists with many elements and the elements may not be common to all lists. We implement two meta-search algorithms, which use a Markov chain framework to convert pairwise preferences between list elements into a stationary distribution that represents an aggregate ranking (Dwork et al, 2001). We explore the behavior of the algorithms in hypothetical examples and a simulated dataset and compare their performance with that of an algorithm based on the order-statistics model of Thurstone (Thurstone, 1927). We apply all three algorithms to aggregate the results of five microarray studies of prostate cancer.

AB - As technology for microarray analysis becomes widespread, it is becoming increasingly important to be able to compare and combine the results of experiments that explore the same scientific question. In this article, we present a rank-aggregation approach for combining results from several microarray studies. The motivation for this approach is twofold; first, the final results of microarray studies are typically expressed as lists of genes, rank-ordered by a measure of the strength of evidence that they are functionally involved in the disease process, and second, using the information on this rank-ordered metric means that we do not have to concern ourselves with data on the actual expression levels, which may not be comparable across experiments. Our approach draws on methods for combining top-k lists from the computer science literature on meta-search. The meta-search problem shares several important features with that of combining microarray experiments, including the fact that there are typically few lists with many elements and the elements may not be common to all lists. We implement two meta-search algorithms, which use a Markov chain framework to convert pairwise preferences between list elements into a stationary distribution that represents an aggregate ranking (Dwork et al, 2001). We explore the behavior of the algorithms in hypothetical examples and a simulated dataset and compare their performance with that of an algorithm based on the order-statistics model of Thurstone (Thurstone, 1927). We apply all three algorithms to aggregate the results of five microarray studies of prostate cancer.

KW - Markov chains

KW - Meta-analysis

KW - Microarrays

KW - Order-statistic models

KW - Rank aggregation

UR - http://www.scopus.com/inward/record.url?scp=85047234321&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047234321&partnerID=8YFLogxK

U2 - 10.2202/1544-6115.1204

DO - 10.2202/1544-6115.1204

M3 - Article

C2 - 17049026

AN - SCOPUS:85047234321

VL - 5

JO - Statistical Applications in Genetics and Molecular Biology

JF - Statistical Applications in Genetics and Molecular Biology

SN - 1544-6115

IS - 1

M1 - 15

ER -