Bayesian models based on test statistics for multiple hypothesis testing problems

Yuan Ji, Yiling Lu, Gordon Mills

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Motivation: We propose a Bayesian method for the problem of multiple hypothesis testing that is routinely encountered in bioinformatics research, such as the differential gene expression analysis. Our algorithm is based on modeling the distributions of test statistics under both null and alternative hypotheses. We substantially reduce the complexity of the process of defining posterior model probabilities by modeling the test statistics directly instead of modeling the full data. Computationally, we apply a Bayesian FDR approach to control the number of rejections of null hypotheses. To check if our model assumptions for the test statistics are valid for various bioinformatics experiments, we also propose a simple graphical model-assessment tool. Results: Using extensive simulations, we demonstrate the performance of our models and the utility of the model-assessment tool. In the end, we apply the proposed methodology to an siRNA screening and a gene expression experiment.

Original languageEnglish (US)
Pages (from-to)943-949
Number of pages7
JournalBioinformatics
Volume24
Issue number7
DOIs
StatePublished - Apr 1 2008
Externally publishedYes

Fingerprint

Multiple Hypothesis Testing
Bayesian Model
Test Statistic
Bayes Theorem
Statistics
Model-based
Computational Biology
Bioinformatics
Testing
Modeling
Gene Expression Analysis
Gene Expression
Gene expression
Differential Expression
Probability Model
Graphical Models
Bayesian Methods
Rejection
Bayesian Approach
Null hypothesis

ASJC Scopus subject areas

  • Statistics and Probability
  • Medicine(all)
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Computational Mathematics

Cite this

Bayesian models based on test statistics for multiple hypothesis testing problems. / Ji, Yuan; Lu, Yiling; Mills, Gordon.

In: Bioinformatics, Vol. 24, No. 7, 01.04.2008, p. 943-949.

Research output: Contribution to journalArticle

@article{b9ab8da5b5e04254989182af016688c5,
title = "Bayesian models based on test statistics for multiple hypothesis testing problems",
abstract = "Motivation: We propose a Bayesian method for the problem of multiple hypothesis testing that is routinely encountered in bioinformatics research, such as the differential gene expression analysis. Our algorithm is based on modeling the distributions of test statistics under both null and alternative hypotheses. We substantially reduce the complexity of the process of defining posterior model probabilities by modeling the test statistics directly instead of modeling the full data. Computationally, we apply a Bayesian FDR approach to control the number of rejections of null hypotheses. To check if our model assumptions for the test statistics are valid for various bioinformatics experiments, we also propose a simple graphical model-assessment tool. Results: Using extensive simulations, we demonstrate the performance of our models and the utility of the model-assessment tool. In the end, we apply the proposed methodology to an siRNA screening and a gene expression experiment.",
author = "Yuan Ji and Yiling Lu and Gordon Mills",
year = "2008",
month = "4",
day = "1",
doi = "10.1093/bioinformatics/btn049",
language = "English (US)",
volume = "24",
pages = "943--949",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "7",

}

TY - JOUR

T1 - Bayesian models based on test statistics for multiple hypothesis testing problems

AU - Ji, Yuan

AU - Lu, Yiling

AU - Mills, Gordon

PY - 2008/4/1

Y1 - 2008/4/1

N2 - Motivation: We propose a Bayesian method for the problem of multiple hypothesis testing that is routinely encountered in bioinformatics research, such as the differential gene expression analysis. Our algorithm is based on modeling the distributions of test statistics under both null and alternative hypotheses. We substantially reduce the complexity of the process of defining posterior model probabilities by modeling the test statistics directly instead of modeling the full data. Computationally, we apply a Bayesian FDR approach to control the number of rejections of null hypotheses. To check if our model assumptions for the test statistics are valid for various bioinformatics experiments, we also propose a simple graphical model-assessment tool. Results: Using extensive simulations, we demonstrate the performance of our models and the utility of the model-assessment tool. In the end, we apply the proposed methodology to an siRNA screening and a gene expression experiment.

AB - Motivation: We propose a Bayesian method for the problem of multiple hypothesis testing that is routinely encountered in bioinformatics research, such as the differential gene expression analysis. Our algorithm is based on modeling the distributions of test statistics under both null and alternative hypotheses. We substantially reduce the complexity of the process of defining posterior model probabilities by modeling the test statistics directly instead of modeling the full data. Computationally, we apply a Bayesian FDR approach to control the number of rejections of null hypotheses. To check if our model assumptions for the test statistics are valid for various bioinformatics experiments, we also propose a simple graphical model-assessment tool. Results: Using extensive simulations, we demonstrate the performance of our models and the utility of the model-assessment tool. In the end, we apply the proposed methodology to an siRNA screening and a gene expression experiment.

UR - http://www.scopus.com/inward/record.url?scp=41349103439&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=41349103439&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btn049

DO - 10.1093/bioinformatics/btn049

M3 - Article

C2 - 18245123

AN - SCOPUS:41349103439

VL - 24

SP - 943

EP - 949

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 7

ER -