Empirical bayes model comparisons for differential methylation analysis

Mingxiang Teng, Yadong Wang, Seongho Kim, Lang Li, Changyu Shen, Guohua Wang, Yunlong Liu, Tim H M Huang, Kenneth Nephew, Curt Balch

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

A number of empirical Bayes models (each with different statistical distribution assumptions) have now been developed to analyze differential DNA methylation using high-density oligonucleotide tiling arrays. However, it remains unclear which model performs best. For example, for analysis of differentially methylated regions for conservative and functional sequence characteristics (e.g., enrichment of transcription factor-binding sites (TFBSs)), the sensitivity of such analyses, using various empirical Bayes models, remains unclear. In this paper, five empirical Bayes models were constructed, based on either a gamma distribution or a log-normal distribution, for the identification of differential methylated loci and their cell division(1, 3, and 5) and drug-treatment-(cisplatin) dependent methylation patterns. While differential methylation patterns generated by log-normal models were enriched with numerous TFBSs, we observed almost no TFBS-enriched sequences using gamma assumption models. Statistical and biological results suggest log-normal, rather than gamma, empirical Bayes model distribution to be a highly accurate and precise method for differential methylation microarray analysis. In addition, we presented one of the log-normal models for differential methylation analysis and tested its reproducibility by simulation study. We believe this research to be the first extensive comparison of statistical modeling for the analysis of differential DNA methylation, an important biological phenomenon that precisely regulates gene transcription.

Original languageEnglish
Article number376706
JournalComparative and Functional Genomics
Volume2012
DOIs
StatePublished - 2012

Fingerprint

Methylation
Transcription Factors
Binding Sites
DNA Methylation
Statistical Distributions
Biological Phenomena
Normal Distribution
Microarray Analysis
Oligonucleotide Array Sequence Analysis
Cell Division
Cisplatin
Research
Pharmaceutical Preparations
Genes

ASJC Scopus subject areas

  • Genetics
  • Molecular Biology
  • Biotechnology

Cite this

Teng, M., Wang, Y., Kim, S., Li, L., Shen, C., Wang, G., ... Balch, C. (2012). Empirical bayes model comparisons for differential methylation analysis. Comparative and Functional Genomics, 2012, [376706]. https://doi.org/10.1155/2012/376706

Empirical bayes model comparisons for differential methylation analysis. / Teng, Mingxiang; Wang, Yadong; Kim, Seongho; Li, Lang; Shen, Changyu; Wang, Guohua; Liu, Yunlong; Huang, Tim H M; Nephew, Kenneth; Balch, Curt.

In: Comparative and Functional Genomics, Vol. 2012, 376706, 2012.

Research output: Contribution to journalArticle

Teng, Mingxiang ; Wang, Yadong ; Kim, Seongho ; Li, Lang ; Shen, Changyu ; Wang, Guohua ; Liu, Yunlong ; Huang, Tim H M ; Nephew, Kenneth ; Balch, Curt. / Empirical bayes model comparisons for differential methylation analysis. In: Comparative and Functional Genomics. 2012 ; Vol. 2012.
@article{d5adb1651ee649709d1d923372cb85cf,
title = "Empirical bayes model comparisons for differential methylation analysis",
abstract = "A number of empirical Bayes models (each with different statistical distribution assumptions) have now been developed to analyze differential DNA methylation using high-density oligonucleotide tiling arrays. However, it remains unclear which model performs best. For example, for analysis of differentially methylated regions for conservative and functional sequence characteristics (e.g., enrichment of transcription factor-binding sites (TFBSs)), the sensitivity of such analyses, using various empirical Bayes models, remains unclear. In this paper, five empirical Bayes models were constructed, based on either a gamma distribution or a log-normal distribution, for the identification of differential methylated loci and their cell division(1, 3, and 5) and drug-treatment-(cisplatin) dependent methylation patterns. While differential methylation patterns generated by log-normal models were enriched with numerous TFBSs, we observed almost no TFBS-enriched sequences using gamma assumption models. Statistical and biological results suggest log-normal, rather than gamma, empirical Bayes model distribution to be a highly accurate and precise method for differential methylation microarray analysis. In addition, we presented one of the log-normal models for differential methylation analysis and tested its reproducibility by simulation study. We believe this research to be the first extensive comparison of statistical modeling for the analysis of differential DNA methylation, an important biological phenomenon that precisely regulates gene transcription.",
author = "Mingxiang Teng and Yadong Wang and Seongho Kim and Lang Li and Changyu Shen and Guohua Wang and Yunlong Liu and Huang, {Tim H M} and Kenneth Nephew and Curt Balch",
year = "2012",
doi = "10.1155/2012/376706",
language = "English",
volume = "2012",
journal = "Comparative and Functional Genomics",
issn = "1531-6912",
publisher = "Hindawi Publishing Corporation",

}

TY - JOUR

T1 - Empirical bayes model comparisons for differential methylation analysis

AU - Teng, Mingxiang

AU - Wang, Yadong

AU - Kim, Seongho

AU - Li, Lang

AU - Shen, Changyu

AU - Wang, Guohua

AU - Liu, Yunlong

AU - Huang, Tim H M

AU - Nephew, Kenneth

AU - Balch, Curt

PY - 2012

Y1 - 2012

N2 - A number of empirical Bayes models (each with different statistical distribution assumptions) have now been developed to analyze differential DNA methylation using high-density oligonucleotide tiling arrays. However, it remains unclear which model performs best. For example, for analysis of differentially methylated regions for conservative and functional sequence characteristics (e.g., enrichment of transcription factor-binding sites (TFBSs)), the sensitivity of such analyses, using various empirical Bayes models, remains unclear. In this paper, five empirical Bayes models were constructed, based on either a gamma distribution or a log-normal distribution, for the identification of differential methylated loci and their cell division(1, 3, and 5) and drug-treatment-(cisplatin) dependent methylation patterns. While differential methylation patterns generated by log-normal models were enriched with numerous TFBSs, we observed almost no TFBS-enriched sequences using gamma assumption models. Statistical and biological results suggest log-normal, rather than gamma, empirical Bayes model distribution to be a highly accurate and precise method for differential methylation microarray analysis. In addition, we presented one of the log-normal models for differential methylation analysis and tested its reproducibility by simulation study. We believe this research to be the first extensive comparison of statistical modeling for the analysis of differential DNA methylation, an important biological phenomenon that precisely regulates gene transcription.

AB - A number of empirical Bayes models (each with different statistical distribution assumptions) have now been developed to analyze differential DNA methylation using high-density oligonucleotide tiling arrays. However, it remains unclear which model performs best. For example, for analysis of differentially methylated regions for conservative and functional sequence characteristics (e.g., enrichment of transcription factor-binding sites (TFBSs)), the sensitivity of such analyses, using various empirical Bayes models, remains unclear. In this paper, five empirical Bayes models were constructed, based on either a gamma distribution or a log-normal distribution, for the identification of differential methylated loci and their cell division(1, 3, and 5) and drug-treatment-(cisplatin) dependent methylation patterns. While differential methylation patterns generated by log-normal models were enriched with numerous TFBSs, we observed almost no TFBS-enriched sequences using gamma assumption models. Statistical and biological results suggest log-normal, rather than gamma, empirical Bayes model distribution to be a highly accurate and precise method for differential methylation microarray analysis. In addition, we presented one of the log-normal models for differential methylation analysis and tested its reproducibility by simulation study. We believe this research to be the first extensive comparison of statistical modeling for the analysis of differential DNA methylation, an important biological phenomenon that precisely regulates gene transcription.

UR - http://www.scopus.com/inward/record.url?scp=84866091554&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866091554&partnerID=8YFLogxK

U2 - 10.1155/2012/376706

DO - 10.1155/2012/376706

M3 - Article

C2 - 22956892

AN - SCOPUS:84866091554

VL - 2012

JO - Comparative and Functional Genomics

JF - Comparative and Functional Genomics

SN - 1531-6912

M1 - 376706

ER -