Comparative study of three commonly used continuous deterministic methods for modeling gene regulation networks

Martin T Swain, Johannes J Mandel, Werner Dubitzky

    Research output: Contribution to journalArticle

    17 Citations (Scopus)

    Abstract

    A gene-regulatory network (GRN) refers to DNA segments that interact through their RNA and protein products and thereby govern the rates at which genes are transcribed. Creating accurate dynamic models of GRNs is gaining importance in biomedical research and development. To improve our understanding of continuous deterministic modeling methods employed to construct dynamic GRN models, we have carried out a comprehensive comparative study of three commonly used systems of ordinary differential equations: The S-system (SS), artificial neural networks (ANNs), and the general rate law of transcription (GRLOT) method. These were thoroughly evaluated in terms of their ability to replicate the reference models' regulatory structure and dynamic gene expression behavior under varying conditions.While the ANN and GRLOT methods appeared to produce robust models even when the model parameters deviated considerably from those of the reference models, SS-based models exhibited a notable loss of performance even when the parameters of the reverse-engineered models corresponded closely to those of the reference models: this is due to the high number of power terms in the SS-method, and the manner in which they are combined. In cross-method reverse-engineering experiments the different characteristics, biases and idiosynchracies of the methods were revealed. Based on limited training data, with only one experimental condition, all methods produced dynamic models that were able to reproduce the training data accurately. However, an accurate reproduction of regulatory network features was only possible with training data originating from multiple experiments under varying conditions.The studied GRN modeling methods produced dynamic GRN models exhibiting marked differences in their ability to replicate the reference models' structure and behavior. Our results suggest that care should be taking when a method is chosen for a particular application. In particular, reliance on only a single method might unduly bias the results.
    LanguageEnglish
    JournalBMC Bioinformatics
    Volume11
    Issue number459
    DOIs
    Publication statusPublished - 14 Sep 2010

    Fingerprint

    Gene Regulation
    Gene Regulatory Networks
    Gene expression
    Comparative Study
    Gene Regulatory Network
    Reference Model
    Modeling
    S-system
    Genes
    Modeling Method
    Transcription
    Network Model
    Artificial Neural Network
    Dynamic Model
    Dynamic models
    Neural networks
    Network Modeling
    Regulatory Networks
    Reverse Engineering
    Reverse engineering

    Keywords

    • Bioinformatics
    • computational biology

    Cite this

    Swain, Martin T ; Mandel, Johannes J ; Dubitzky, Werner. / Comparative study of three commonly used continuous deterministic methods for modeling gene regulation networks. In: BMC Bioinformatics. 2010 ; Vol. 11, No. 459.
    @article{8a57974a45e44bda8d98db6985d403d5,
    title = "Comparative study of three commonly used continuous deterministic methods for modeling gene regulation networks",
    abstract = "A gene-regulatory network (GRN) refers to DNA segments that interact through their RNA and protein products and thereby govern the rates at which genes are transcribed. Creating accurate dynamic models of GRNs is gaining importance in biomedical research and development. To improve our understanding of continuous deterministic modeling methods employed to construct dynamic GRN models, we have carried out a comprehensive comparative study of three commonly used systems of ordinary differential equations: The S-system (SS), artificial neural networks (ANNs), and the general rate law of transcription (GRLOT) method. These were thoroughly evaluated in terms of their ability to replicate the reference models' regulatory structure and dynamic gene expression behavior under varying conditions.While the ANN and GRLOT methods appeared to produce robust models even when the model parameters deviated considerably from those of the reference models, SS-based models exhibited a notable loss of performance even when the parameters of the reverse-engineered models corresponded closely to those of the reference models: this is due to the high number of power terms in the SS-method, and the manner in which they are combined. In cross-method reverse-engineering experiments the different characteristics, biases and idiosynchracies of the methods were revealed. Based on limited training data, with only one experimental condition, all methods produced dynamic models that were able to reproduce the training data accurately. However, an accurate reproduction of regulatory network features was only possible with training data originating from multiple experiments under varying conditions.The studied GRN modeling methods produced dynamic GRN models exhibiting marked differences in their ability to replicate the reference models' structure and behavior. Our results suggest that care should be taking when a method is chosen for a particular application. In particular, reliance on only a single method might unduly bias the results.",
    keywords = "Bioinformatics, computational biology",
    author = "Swain, {Martin T} and Mandel, {Johannes J} and Werner Dubitzky",
    year = "2010",
    month = "9",
    day = "14",
    doi = "10.1186/1471-2105-11-459",
    language = "English",
    volume = "11",
    journal = "BMC Bioinformatics",
    issn = "1471-2105",
    publisher = "BioMed Central",
    number = "459",

    }

    Comparative study of three commonly used continuous deterministic methods for modeling gene regulation networks. / Swain, Martin T; Mandel, Johannes J; Dubitzky, Werner.

    In: BMC Bioinformatics, Vol. 11, No. 459, 14.09.2010.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Comparative study of three commonly used continuous deterministic methods for modeling gene regulation networks

    AU - Swain, Martin T

    AU - Mandel, Johannes J

    AU - Dubitzky, Werner

    PY - 2010/9/14

    Y1 - 2010/9/14

    N2 - A gene-regulatory network (GRN) refers to DNA segments that interact through their RNA and protein products and thereby govern the rates at which genes are transcribed. Creating accurate dynamic models of GRNs is gaining importance in biomedical research and development. To improve our understanding of continuous deterministic modeling methods employed to construct dynamic GRN models, we have carried out a comprehensive comparative study of three commonly used systems of ordinary differential equations: The S-system (SS), artificial neural networks (ANNs), and the general rate law of transcription (GRLOT) method. These were thoroughly evaluated in terms of their ability to replicate the reference models' regulatory structure and dynamic gene expression behavior under varying conditions.While the ANN and GRLOT methods appeared to produce robust models even when the model parameters deviated considerably from those of the reference models, SS-based models exhibited a notable loss of performance even when the parameters of the reverse-engineered models corresponded closely to those of the reference models: this is due to the high number of power terms in the SS-method, and the manner in which they are combined. In cross-method reverse-engineering experiments the different characteristics, biases and idiosynchracies of the methods were revealed. Based on limited training data, with only one experimental condition, all methods produced dynamic models that were able to reproduce the training data accurately. However, an accurate reproduction of regulatory network features was only possible with training data originating from multiple experiments under varying conditions.The studied GRN modeling methods produced dynamic GRN models exhibiting marked differences in their ability to replicate the reference models' structure and behavior. Our results suggest that care should be taking when a method is chosen for a particular application. In particular, reliance on only a single method might unduly bias the results.

    AB - A gene-regulatory network (GRN) refers to DNA segments that interact through their RNA and protein products and thereby govern the rates at which genes are transcribed. Creating accurate dynamic models of GRNs is gaining importance in biomedical research and development. To improve our understanding of continuous deterministic modeling methods employed to construct dynamic GRN models, we have carried out a comprehensive comparative study of three commonly used systems of ordinary differential equations: The S-system (SS), artificial neural networks (ANNs), and the general rate law of transcription (GRLOT) method. These were thoroughly evaluated in terms of their ability to replicate the reference models' regulatory structure and dynamic gene expression behavior under varying conditions.While the ANN and GRLOT methods appeared to produce robust models even when the model parameters deviated considerably from those of the reference models, SS-based models exhibited a notable loss of performance even when the parameters of the reverse-engineered models corresponded closely to those of the reference models: this is due to the high number of power terms in the SS-method, and the manner in which they are combined. In cross-method reverse-engineering experiments the different characteristics, biases and idiosynchracies of the methods were revealed. Based on limited training data, with only one experimental condition, all methods produced dynamic models that were able to reproduce the training data accurately. However, an accurate reproduction of regulatory network features was only possible with training data originating from multiple experiments under varying conditions.The studied GRN modeling methods produced dynamic GRN models exhibiting marked differences in their ability to replicate the reference models' structure and behavior. Our results suggest that care should be taking when a method is chosen for a particular application. In particular, reliance on only a single method might unduly bias the results.

    KW - Bioinformatics

    KW - computational biology

    U2 - 10.1186/1471-2105-11-459

    DO - 10.1186/1471-2105-11-459

    M3 - Article

    VL - 11

    JO - BMC Bioinformatics

    T2 - BMC Bioinformatics

    JF - BMC Bioinformatics

    SN - 1471-2105

    IS - 459

    ER -