Identifying differentially expressed genes in meta-analysis via Bayesian model-based clustering

Yoon Young Jung, Man Suk Oh, Dong Wan Shin, Seung Ho Kang, Hyun Sook Oh

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

A Bayesian model-based clustering approach is proposed for identifying differentially expressed genes in meta-analysis. A Bayesian hierarchical model is used as a scientific tool for combining information from different studies, and a mixture prior is used to separate differentially expressed genes from non-differentially expressed genes. Posterior estimation of the parameters and missing observations are done by using a simple Markov chain Monte Carlo method. From the estimated mixture model, useful measure of significance of a test such as the Bayesian false discovery rate (FDR), the local FDR (Efron et al., 2001), and the integration-driven discovery rate (IDR; Choi et al., 2003) can be easily computed. The model-based approach is also compared with commonly used permutation methods, and it is shown that the model-based approach is superior to the permutation methods when there are excessive under-expressed genes compared to over-expressed genes or vice versa. The proposed method is applied to four publicly available prostate cancer gene expression data sets and simulated data sets.

Original languageEnglish
Pages (from-to)435-450
Number of pages16
JournalBiometrical Journal
Volume48
Issue number3
DOIs
Publication statusPublished - 2006 Jun 1

Fingerprint

Model-based Clustering
Bayesian Model
Gene
Permutation
Model-based
Bayesian Hierarchical Model
Missing Observations
Prostate Cancer
Markov Chain Monte Carlo Methods
Gene Expression Data
Mixture Model
Bayesian model
Meta-analysis
Clustering

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Jung, Yoon Young ; Oh, Man Suk ; Shin, Dong Wan ; Kang, Seung Ho ; Oh, Hyun Sook. / Identifying differentially expressed genes in meta-analysis via Bayesian model-based clustering. In: Biometrical Journal. 2006 ; Vol. 48, No. 3. pp. 435-450.
@article{788ee1f1aa2c4ce3ab6f04351ecc84d9,
title = "Identifying differentially expressed genes in meta-analysis via Bayesian model-based clustering",
abstract = "A Bayesian model-based clustering approach is proposed for identifying differentially expressed genes in meta-analysis. A Bayesian hierarchical model is used as a scientific tool for combining information from different studies, and a mixture prior is used to separate differentially expressed genes from non-differentially expressed genes. Posterior estimation of the parameters and missing observations are done by using a simple Markov chain Monte Carlo method. From the estimated mixture model, useful measure of significance of a test such as the Bayesian false discovery rate (FDR), the local FDR (Efron et al., 2001), and the integration-driven discovery rate (IDR; Choi et al., 2003) can be easily computed. The model-based approach is also compared with commonly used permutation methods, and it is shown that the model-based approach is superior to the permutation methods when there are excessive under-expressed genes compared to over-expressed genes or vice versa. The proposed method is applied to four publicly available prostate cancer gene expression data sets and simulated data sets.",
author = "Jung, {Yoon Young} and Oh, {Man Suk} and Shin, {Dong Wan} and Kang, {Seung Ho} and Oh, {Hyun Sook}",
year = "2006",
month = "6",
day = "1",
doi = "10.1002/bimj.200410230",
language = "English",
volume = "48",
pages = "435--450",
journal = "Biometrical Journal",
issn = "0323-3847",
publisher = "Wiley-VCH Verlag",
number = "3",

}

Identifying differentially expressed genes in meta-analysis via Bayesian model-based clustering. / Jung, Yoon Young; Oh, Man Suk; Shin, Dong Wan; Kang, Seung Ho; Oh, Hyun Sook.

In: Biometrical Journal, Vol. 48, No. 3, 01.06.2006, p. 435-450.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Identifying differentially expressed genes in meta-analysis via Bayesian model-based clustering

AU - Jung, Yoon Young

AU - Oh, Man Suk

AU - Shin, Dong Wan

AU - Kang, Seung Ho

AU - Oh, Hyun Sook

PY - 2006/6/1

Y1 - 2006/6/1

N2 - A Bayesian model-based clustering approach is proposed for identifying differentially expressed genes in meta-analysis. A Bayesian hierarchical model is used as a scientific tool for combining information from different studies, and a mixture prior is used to separate differentially expressed genes from non-differentially expressed genes. Posterior estimation of the parameters and missing observations are done by using a simple Markov chain Monte Carlo method. From the estimated mixture model, useful measure of significance of a test such as the Bayesian false discovery rate (FDR), the local FDR (Efron et al., 2001), and the integration-driven discovery rate (IDR; Choi et al., 2003) can be easily computed. The model-based approach is also compared with commonly used permutation methods, and it is shown that the model-based approach is superior to the permutation methods when there are excessive under-expressed genes compared to over-expressed genes or vice versa. The proposed method is applied to four publicly available prostate cancer gene expression data sets and simulated data sets.

AB - A Bayesian model-based clustering approach is proposed for identifying differentially expressed genes in meta-analysis. A Bayesian hierarchical model is used as a scientific tool for combining information from different studies, and a mixture prior is used to separate differentially expressed genes from non-differentially expressed genes. Posterior estimation of the parameters and missing observations are done by using a simple Markov chain Monte Carlo method. From the estimated mixture model, useful measure of significance of a test such as the Bayesian false discovery rate (FDR), the local FDR (Efron et al., 2001), and the integration-driven discovery rate (IDR; Choi et al., 2003) can be easily computed. The model-based approach is also compared with commonly used permutation methods, and it is shown that the model-based approach is superior to the permutation methods when there are excessive under-expressed genes compared to over-expressed genes or vice versa. The proposed method is applied to four publicly available prostate cancer gene expression data sets and simulated data sets.

UR - http://www.scopus.com/inward/record.url?scp=33745654577&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33745654577&partnerID=8YFLogxK

U2 - 10.1002/bimj.200410230

DO - 10.1002/bimj.200410230

M3 - Article

C2 - 16845907

AN - SCOPUS:33745654577

VL - 48

SP - 435

EP - 450

JO - Biometrical Journal

JF - Biometrical Journal

SN - 0323-3847

IS - 3

ER -