Learning chromatin states with factorized information criteria

Michiaki Hamada, Yukiteru Ono, Ryohei Fujimaki, Kiyoshi Asai

    研究成果: Article

    5 引用 (Scopus)

    抄録

    Motivation: Recent studies have suggested that both the genome and the genome with epigenetic modifications, the so-called epigenome, play important roles in various biological functions, such as transcription and DNA replication, repair, and recombination. It is well known that specific combinations of histone modifications (e.g. methylations and acetylations) of nucleosomes induce chromatin states that correspond to specific functions of chromatin. Although the advent of next-generation sequencing (NGS) technologies enables measurement of epigenetic information for entire genomes at high-resolution, the variety of chromatin states has not been completely characterized. Results: In this study, we propose a method to estimate the chromatin states indicated by genomewide chromatin marks identified by NGS technologies. The proposed method automatically estimates the number of chromatin states and characterize each state on the basis of a hidden Markov model (HMM) in combination with a recently proposed model selection technique, factorized information criteria. The method is expected to provide an unbiased model because it relies on only two adjustable parameters and avoids heuristic procedures as much as possible. Computational experiments with simulated datasets show that our method automatically learns an appropriate model, even in cases where methods that rely on Bayesian information criteria fail to learn the model structures. In addition, we comprehensively compare our method to ChromHMM on three real datasets and show that our method estimates more chromatin states than ChromHMM for those datasets.

    元の言語English
    ページ(範囲)2426-2433
    ページ数8
    ジャーナルBioinformatics
    31
    発行部数15
    DOI
    出版物ステータスPublished - 2015

    Fingerprint

    Information Criterion
    Chromatin
    Genes
    Learning
    Acetylation
    Methylation
    Transcription
    Hidden Markov models
    Genome
    Model structures
    DNA
    Repair
    Sequencing
    Epigenomics
    Estimate
    Histone Code
    Bayesian Information Criterion
    DNA Replication
    Technology
    Nucleosomes

    ASJC Scopus subject areas

    • Biochemistry
    • Molecular Biology
    • Computational Theory and Mathematics
    • Computer Science Applications
    • Computational Mathematics
    • Statistics and Probability

    これを引用

    Learning chromatin states with factorized information criteria. / Hamada, Michiaki; Ono, Yukiteru; Fujimaki, Ryohei; Asai, Kiyoshi.

    :: Bioinformatics, 巻 31, 番号 15, 2015, p. 2426-2433.

    研究成果: Article

    Hamada, Michiaki ; Ono, Yukiteru ; Fujimaki, Ryohei ; Asai, Kiyoshi. / Learning chromatin states with factorized information criteria. :: Bioinformatics. 2015 ; 巻 31, 番号 15. pp. 2426-2433.
    @article{d615e10dd9fd419e9165e04063a02482,
    title = "Learning chromatin states with factorized information criteria",
    abstract = "Motivation: Recent studies have suggested that both the genome and the genome with epigenetic modifications, the so-called epigenome, play important roles in various biological functions, such as transcription and DNA replication, repair, and recombination. It is well known that specific combinations of histone modifications (e.g. methylations and acetylations) of nucleosomes induce chromatin states that correspond to specific functions of chromatin. Although the advent of next-generation sequencing (NGS) technologies enables measurement of epigenetic information for entire genomes at high-resolution, the variety of chromatin states has not been completely characterized. Results: In this study, we propose a method to estimate the chromatin states indicated by genomewide chromatin marks identified by NGS technologies. The proposed method automatically estimates the number of chromatin states and characterize each state on the basis of a hidden Markov model (HMM) in combination with a recently proposed model selection technique, factorized information criteria. The method is expected to provide an unbiased model because it relies on only two adjustable parameters and avoids heuristic procedures as much as possible. Computational experiments with simulated datasets show that our method automatically learns an appropriate model, even in cases where methods that rely on Bayesian information criteria fail to learn the model structures. In addition, we comprehensively compare our method to ChromHMM on three real datasets and show that our method estimates more chromatin states than ChromHMM for those datasets.",
    author = "Michiaki Hamada and Yukiteru Ono and Ryohei Fujimaki and Kiyoshi Asai",
    year = "2015",
    doi = "10.1093/bioinformatics/btv163",
    language = "English",
    volume = "31",
    pages = "2426--2433",
    journal = "Bioinformatics",
    issn = "1367-4803",
    publisher = "Oxford University Press",
    number = "15",

    }

    TY - JOUR

    T1 - Learning chromatin states with factorized information criteria

    AU - Hamada, Michiaki

    AU - Ono, Yukiteru

    AU - Fujimaki, Ryohei

    AU - Asai, Kiyoshi

    PY - 2015

    Y1 - 2015

    N2 - Motivation: Recent studies have suggested that both the genome and the genome with epigenetic modifications, the so-called epigenome, play important roles in various biological functions, such as transcription and DNA replication, repair, and recombination. It is well known that specific combinations of histone modifications (e.g. methylations and acetylations) of nucleosomes induce chromatin states that correspond to specific functions of chromatin. Although the advent of next-generation sequencing (NGS) technologies enables measurement of epigenetic information for entire genomes at high-resolution, the variety of chromatin states has not been completely characterized. Results: In this study, we propose a method to estimate the chromatin states indicated by genomewide chromatin marks identified by NGS technologies. The proposed method automatically estimates the number of chromatin states and characterize each state on the basis of a hidden Markov model (HMM) in combination with a recently proposed model selection technique, factorized information criteria. The method is expected to provide an unbiased model because it relies on only two adjustable parameters and avoids heuristic procedures as much as possible. Computational experiments with simulated datasets show that our method automatically learns an appropriate model, even in cases where methods that rely on Bayesian information criteria fail to learn the model structures. In addition, we comprehensively compare our method to ChromHMM on three real datasets and show that our method estimates more chromatin states than ChromHMM for those datasets.

    AB - Motivation: Recent studies have suggested that both the genome and the genome with epigenetic modifications, the so-called epigenome, play important roles in various biological functions, such as transcription and DNA replication, repair, and recombination. It is well known that specific combinations of histone modifications (e.g. methylations and acetylations) of nucleosomes induce chromatin states that correspond to specific functions of chromatin. Although the advent of next-generation sequencing (NGS) technologies enables measurement of epigenetic information for entire genomes at high-resolution, the variety of chromatin states has not been completely characterized. Results: In this study, we propose a method to estimate the chromatin states indicated by genomewide chromatin marks identified by NGS technologies. The proposed method automatically estimates the number of chromatin states and characterize each state on the basis of a hidden Markov model (HMM) in combination with a recently proposed model selection technique, factorized information criteria. The method is expected to provide an unbiased model because it relies on only two adjustable parameters and avoids heuristic procedures as much as possible. Computational experiments with simulated datasets show that our method automatically learns an appropriate model, even in cases where methods that rely on Bayesian information criteria fail to learn the model structures. In addition, we comprehensively compare our method to ChromHMM on three real datasets and show that our method estimates more chromatin states than ChromHMM for those datasets.

    UR - http://www.scopus.com/inward/record.url?scp=84943619261&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84943619261&partnerID=8YFLogxK

    U2 - 10.1093/bioinformatics/btv163

    DO - 10.1093/bioinformatics/btv163

    M3 - Article

    C2 - 25810430

    AN - SCOPUS:84943619261

    VL - 31

    SP - 2426

    EP - 2433

    JO - Bioinformatics

    JF - Bioinformatics

    SN - 1367-4803

    IS - 15

    ER -