Empirical Bayes model comparisons for differential methylation analysis

Mingxiang Teng, Yadong Wang, Yunlong Liu, Seongho Kim, Curt Balch, Kenneth Nephew, Lang Li

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

Abstract

A number of empirical Bayes models were developed to investigate the differential methylation analysis. However, it is not clear which empirical Bayes model performs best in differential methylation analysis. In this paper, five empirical Bayes models were constructed and applied to the differential methylation aualysis of A2780 cells between before and after 1, 3, and 5 round of cisplatin treatment. The log-normal model with the background variance showed the lowest minimized negative log-likelihood. It inferred increasing number of differentially methylated loci from 1 to 3 to 5 rounds of cisplatin treatment on the A2780 cells, which was consistent to cisplatin resistant IC50 data. Among differentially methylated loci selected from each empirical model, three time dependent methylation patterns were defined: stochastic hypomethylation, stochastic hypermethylation, and random methylation. If the empirical Bayes model of the DNA methylation assumed log-normal distribution, both stochastically hypomethylated loci and stochastically hypermethylated loci were enriched with a number of transcription factor binding sites. Almost no TFBS enrichment was observed if the gamma distribution was assumed in the empirical Bayes model. In summary, by comparing the performances of the differential methylation analysis and the TFBS enrichment analysis, log-normal distribution is a better statistical assumption than the gamma distribution in the empirical Bayes model.

Original languageEnglish
Title of host publicationProceedings - IEEE International Workshop on Genomic Signal Processing and Statistics
Pages9-12
Number of pages4
StatePublished - 2011
Event2011 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS'11 - San Antonio, TX, United States
Duration: Dec 4 2011Dec 6 2011

Other

Other2011 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS'11
CountryUnited States
CitySan Antonio, TX
Period12/4/1112/6/11

Fingerprint

Methylation
Cisplatin
Normal Distribution
Normal distribution
DNA Methylation
Inhibitory Concentration 50
Transcription factors
Transcription Factors
Binding sites
Binding Sites

Keywords

  • Differential methylation analysis
  • Empirical Bayes model
  • Transcription factor binding site enrichment

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology (miscellaneous)
  • Computational Theory and Mathematics
  • Signal Processing
  • Biomedical Engineering

Cite this

Teng, M., Wang, Y., Liu, Y., Kim, S., Balch, C., Nephew, K., & Li, L. (2011). Empirical Bayes model comparisons for differential methylation analysis. In Proceedings - IEEE International Workshop on Genomic Signal Processing and Statistics (pp. 9-12). [6169428]

Empirical Bayes model comparisons for differential methylation analysis. / Teng, Mingxiang; Wang, Yadong; Liu, Yunlong; Kim, Seongho; Balch, Curt; Nephew, Kenneth; Li, Lang.

Proceedings - IEEE International Workshop on Genomic Signal Processing and Statistics. 2011. p. 9-12 6169428.

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

Teng, M, Wang, Y, Liu, Y, Kim, S, Balch, C, Nephew, K & Li, L 2011, Empirical Bayes model comparisons for differential methylation analysis. in Proceedings - IEEE International Workshop on Genomic Signal Processing and Statistics., 6169428, pp. 9-12, 2011 IEEE International Workshop on Genomic Signal Processing and Statistics, GENSIPS'11, San Antonio, TX, United States, 12/4/11.
Teng M, Wang Y, Liu Y, Kim S, Balch C, Nephew K et al. Empirical Bayes model comparisons for differential methylation analysis. In Proceedings - IEEE International Workshop on Genomic Signal Processing and Statistics. 2011. p. 9-12. 6169428
Teng, Mingxiang ; Wang, Yadong ; Liu, Yunlong ; Kim, Seongho ; Balch, Curt ; Nephew, Kenneth ; Li, Lang. / Empirical Bayes model comparisons for differential methylation analysis. Proceedings - IEEE International Workshop on Genomic Signal Processing and Statistics. 2011. pp. 9-12
@inproceedings{5ef6c1edbadb4da1bc13eb2961f73785,
title = "Empirical Bayes model comparisons for differential methylation analysis",
abstract = "A number of empirical Bayes models were developed to investigate the differential methylation analysis. However, it is not clear which empirical Bayes model performs best in differential methylation analysis. In this paper, five empirical Bayes models were constructed and applied to the differential methylation aualysis of A2780 cells between before and after 1, 3, and 5 round of cisplatin treatment. The log-normal model with the background variance showed the lowest minimized negative log-likelihood. It inferred increasing number of differentially methylated loci from 1 to 3 to 5 rounds of cisplatin treatment on the A2780 cells, which was consistent to cisplatin resistant IC50 data. Among differentially methylated loci selected from each empirical model, three time dependent methylation patterns were defined: stochastic hypomethylation, stochastic hypermethylation, and random methylation. If the empirical Bayes model of the DNA methylation assumed log-normal distribution, both stochastically hypomethylated loci and stochastically hypermethylated loci were enriched with a number of transcription factor binding sites. Almost no TFBS enrichment was observed if the gamma distribution was assumed in the empirical Bayes model. In summary, by comparing the performances of the differential methylation analysis and the TFBS enrichment analysis, log-normal distribution is a better statistical assumption than the gamma distribution in the empirical Bayes model.",
keywords = "Differential methylation analysis, Empirical Bayes model, Transcription factor binding site enrichment",
author = "Mingxiang Teng and Yadong Wang and Yunlong Liu and Seongho Kim and Curt Balch and Kenneth Nephew and Lang Li",
year = "2011",
language = "English",
isbn = "9781467304900",
pages = "9--12",
booktitle = "Proceedings - IEEE International Workshop on Genomic Signal Processing and Statistics",

}

TY - GEN

T1 - Empirical Bayes model comparisons for differential methylation analysis

AU - Teng, Mingxiang

AU - Wang, Yadong

AU - Liu, Yunlong

AU - Kim, Seongho

AU - Balch, Curt

AU - Nephew, Kenneth

AU - Li, Lang

PY - 2011

Y1 - 2011

N2 - A number of empirical Bayes models were developed to investigate the differential methylation analysis. However, it is not clear which empirical Bayes model performs best in differential methylation analysis. In this paper, five empirical Bayes models were constructed and applied to the differential methylation aualysis of A2780 cells between before and after 1, 3, and 5 round of cisplatin treatment. The log-normal model with the background variance showed the lowest minimized negative log-likelihood. It inferred increasing number of differentially methylated loci from 1 to 3 to 5 rounds of cisplatin treatment on the A2780 cells, which was consistent to cisplatin resistant IC50 data. Among differentially methylated loci selected from each empirical model, three time dependent methylation patterns were defined: stochastic hypomethylation, stochastic hypermethylation, and random methylation. If the empirical Bayes model of the DNA methylation assumed log-normal distribution, both stochastically hypomethylated loci and stochastically hypermethylated loci were enriched with a number of transcription factor binding sites. Almost no TFBS enrichment was observed if the gamma distribution was assumed in the empirical Bayes model. In summary, by comparing the performances of the differential methylation analysis and the TFBS enrichment analysis, log-normal distribution is a better statistical assumption than the gamma distribution in the empirical Bayes model.

AB - A number of empirical Bayes models were developed to investigate the differential methylation analysis. However, it is not clear which empirical Bayes model performs best in differential methylation analysis. In this paper, five empirical Bayes models were constructed and applied to the differential methylation aualysis of A2780 cells between before and after 1, 3, and 5 round of cisplatin treatment. The log-normal model with the background variance showed the lowest minimized negative log-likelihood. It inferred increasing number of differentially methylated loci from 1 to 3 to 5 rounds of cisplatin treatment on the A2780 cells, which was consistent to cisplatin resistant IC50 data. Among differentially methylated loci selected from each empirical model, three time dependent methylation patterns were defined: stochastic hypomethylation, stochastic hypermethylation, and random methylation. If the empirical Bayes model of the DNA methylation assumed log-normal distribution, both stochastically hypomethylated loci and stochastically hypermethylated loci were enriched with a number of transcription factor binding sites. Almost no TFBS enrichment was observed if the gamma distribution was assumed in the empirical Bayes model. In summary, by comparing the performances of the differential methylation analysis and the TFBS enrichment analysis, log-normal distribution is a better statistical assumption than the gamma distribution in the empirical Bayes model.

KW - Differential methylation analysis

KW - Empirical Bayes model

KW - Transcription factor binding site enrichment

UR - http://www.scopus.com/inward/record.url?scp=84863664794&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863664794&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781467304900

SP - 9

EP - 12

BT - Proceedings - IEEE International Workshop on Genomic Signal Processing and Statistics

ER -