A computational framework to integrate high-throughput '-omics' datasets for the identification of potential mechanistic links

Helle Krogh Pedersen, Sofia K Forslund, Valborg Gudmundsdottir, Anders Østergaard Petersen, Falk Hildebrand, Tuulia Hyötyläinen, Trine Nielsen, Torben Hansen, Peer Bork, S Dusko Ehrlich, Søren Brunak, Matej Oresic, Oluf Pedersen*, Henrik Bjørn Nielsen

*Corresponding author for this work

    Research output: Contribution to journalJournal articleResearchpeer-review

    Abstract

    We recently presented a three-pronged association study that integrated human intestinal microbiome data derived from shotgun-based sequencing with untargeted serum metabolome data and measures of host physiology. Metabolome and microbiome data are high dimensional, posing a major challenge for data integration. Here, we present a step-by-step computational protocol that details and discusses the dimensionality-reduction techniques used and methods for subsequent integration and interpretation of such heterogeneous types of data. Dimensionality reduction was achieved through a combination of data normalization approaches, binning of co-abundant genes and metabolites, and integration of prior biological knowledge. The use of prior knowledge to overcome functional redundancy across microbiome species is one central advance of our method over available alternative approaches. Applying this framework, other investigators can integrate various '-omics' readouts with variables of host physiology or any other phenotype of interest (e.g., connecting host and microbiome readouts to disease severity or treatment outcome in a clinical cohort) in a three-pronged association analysis to identify potential mechanistic links to be tested in experimental settings. Although we originally developed the framework for a human metabolome-microbiome study, it is generalizable to other organisms and environmental metagenomes, as well as to studies including other -omics domains such as transcriptomics and proteomics. The provided R code runs in ~1 h on a standard PC.
    Original languageEnglish
    JournalNature Protocols
    Volume13
    Pages (from-to)2781-2800
    Number of pages20
    ISSN1750-2799
    DOIs
    Publication statusPublished - 2018

    Cite this

    Pedersen, Helle Krogh ; Forslund, Sofia K ; Gudmundsdottir, Valborg ; Petersen, Anders Østergaard ; Hildebrand, Falk ; Hyötyläinen, Tuulia ; Nielsen, Trine ; Hansen, Torben ; Bork, Peer ; Ehrlich, S Dusko ; Brunak, Søren ; Oresic, Matej ; Pedersen, Oluf ; Nielsen, Henrik Bjørn. / A computational framework to integrate high-throughput '-omics' datasets for the identification of potential mechanistic links. In: Nature Protocols. 2018 ; Vol. 13. pp. 2781-2800.
    @article{54e286e77af04eaca2186ddedb7d699f,
    title = "A computational framework to integrate high-throughput '-omics' datasets for the identification of potential mechanistic links",
    abstract = "We recently presented a three-pronged association study that integrated human intestinal microbiome data derived from shotgun-based sequencing with untargeted serum metabolome data and measures of host physiology. Metabolome and microbiome data are high dimensional, posing a major challenge for data integration. Here, we present a step-by-step computational protocol that details and discusses the dimensionality-reduction techniques used and methods for subsequent integration and interpretation of such heterogeneous types of data. Dimensionality reduction was achieved through a combination of data normalization approaches, binning of co-abundant genes and metabolites, and integration of prior biological knowledge. The use of prior knowledge to overcome functional redundancy across microbiome species is one central advance of our method over available alternative approaches. Applying this framework, other investigators can integrate various '-omics' readouts with variables of host physiology or any other phenotype of interest (e.g., connecting host and microbiome readouts to disease severity or treatment outcome in a clinical cohort) in a three-pronged association analysis to identify potential mechanistic links to be tested in experimental settings. Although we originally developed the framework for a human metabolome-microbiome study, it is generalizable to other organisms and environmental metagenomes, as well as to studies including other -omics domains such as transcriptomics and proteomics. The provided R code runs in ~1 h on a standard PC.",
    author = "Pedersen, {Helle Krogh} and Forslund, {Sofia K} and Valborg Gudmundsdottir and Petersen, {Anders {\O}stergaard} and Falk Hildebrand and Tuulia Hy{\"o}tyl{\"a}inen and Trine Nielsen and Torben Hansen and Peer Bork and Ehrlich, {S Dusko} and S{\o}ren Brunak and Matej Oresic and Oluf Pedersen and Nielsen, {Henrik Bj{\o}rn}",
    year = "2018",
    doi = "10.1038/s41596-018-0064-z",
    language = "English",
    volume = "13",
    pages = "2781--2800",
    journal = "Nature Protocols (Online)",
    issn = "1750-2799",
    publisher = "Nature Publishing Group",

    }

    Pedersen, HK, Forslund, SK, Gudmundsdottir, V, Petersen, AØ, Hildebrand, F, Hyötyläinen, T, Nielsen, T, Hansen, T, Bork, P, Ehrlich, SD, Brunak, S, Oresic, M, Pedersen, O & Nielsen, HB 2018, 'A computational framework to integrate high-throughput '-omics' datasets for the identification of potential mechanistic links', Nature Protocols, vol. 13, pp. 2781-2800. https://doi.org/10.1038/s41596-018-0064-z

    A computational framework to integrate high-throughput '-omics' datasets for the identification of potential mechanistic links. / Pedersen, Helle Krogh; Forslund, Sofia K; Gudmundsdottir, Valborg; Petersen, Anders Østergaard; Hildebrand, Falk; Hyötyläinen, Tuulia; Nielsen, Trine; Hansen, Torben; Bork, Peer; Ehrlich, S Dusko; Brunak, Søren; Oresic, Matej; Pedersen, Oluf; Nielsen, Henrik Bjørn.

    In: Nature Protocols, Vol. 13, 2018, p. 2781-2800.

    Research output: Contribution to journalJournal articleResearchpeer-review

    TY - JOUR

    T1 - A computational framework to integrate high-throughput '-omics' datasets for the identification of potential mechanistic links

    AU - Pedersen, Helle Krogh

    AU - Forslund, Sofia K

    AU - Gudmundsdottir, Valborg

    AU - Petersen, Anders Østergaard

    AU - Hildebrand, Falk

    AU - Hyötyläinen, Tuulia

    AU - Nielsen, Trine

    AU - Hansen, Torben

    AU - Bork, Peer

    AU - Ehrlich, S Dusko

    AU - Brunak, Søren

    AU - Oresic, Matej

    AU - Pedersen, Oluf

    AU - Nielsen, Henrik Bjørn

    PY - 2018

    Y1 - 2018

    N2 - We recently presented a three-pronged association study that integrated human intestinal microbiome data derived from shotgun-based sequencing with untargeted serum metabolome data and measures of host physiology. Metabolome and microbiome data are high dimensional, posing a major challenge for data integration. Here, we present a step-by-step computational protocol that details and discusses the dimensionality-reduction techniques used and methods for subsequent integration and interpretation of such heterogeneous types of data. Dimensionality reduction was achieved through a combination of data normalization approaches, binning of co-abundant genes and metabolites, and integration of prior biological knowledge. The use of prior knowledge to overcome functional redundancy across microbiome species is one central advance of our method over available alternative approaches. Applying this framework, other investigators can integrate various '-omics' readouts with variables of host physiology or any other phenotype of interest (e.g., connecting host and microbiome readouts to disease severity or treatment outcome in a clinical cohort) in a three-pronged association analysis to identify potential mechanistic links to be tested in experimental settings. Although we originally developed the framework for a human metabolome-microbiome study, it is generalizable to other organisms and environmental metagenomes, as well as to studies including other -omics domains such as transcriptomics and proteomics. The provided R code runs in ~1 h on a standard PC.

    AB - We recently presented a three-pronged association study that integrated human intestinal microbiome data derived from shotgun-based sequencing with untargeted serum metabolome data and measures of host physiology. Metabolome and microbiome data are high dimensional, posing a major challenge for data integration. Here, we present a step-by-step computational protocol that details and discusses the dimensionality-reduction techniques used and methods for subsequent integration and interpretation of such heterogeneous types of data. Dimensionality reduction was achieved through a combination of data normalization approaches, binning of co-abundant genes and metabolites, and integration of prior biological knowledge. The use of prior knowledge to overcome functional redundancy across microbiome species is one central advance of our method over available alternative approaches. Applying this framework, other investigators can integrate various '-omics' readouts with variables of host physiology or any other phenotype of interest (e.g., connecting host and microbiome readouts to disease severity or treatment outcome in a clinical cohort) in a three-pronged association analysis to identify potential mechanistic links to be tested in experimental settings. Although we originally developed the framework for a human metabolome-microbiome study, it is generalizable to other organisms and environmental metagenomes, as well as to studies including other -omics domains such as transcriptomics and proteomics. The provided R code runs in ~1 h on a standard PC.

    U2 - 10.1038/s41596-018-0064-z

    DO - 10.1038/s41596-018-0064-z

    M3 - Journal article

    VL - 13

    SP - 2781

    EP - 2800

    JO - Nature Protocols (Online)

    JF - Nature Protocols (Online)

    SN - 1750-2799

    ER -