%0 Journal Article %1 Narayanan2010Simultaneous %A Narayanan, Manikandan %A Vetta, Adrian %A Schadt, Eric E. %A Zhu, Jun %D 2010 %I Public Library of Science %J PLoS Comput Biol %K clustering data-mining gene-expression integrated-modelling %N 4 %P e1000742+ %R 10.1371/journal.pcbi.1000742 %T Simultaneous Clustering of Multiple Gene Expression and Physical Interaction Datasets %U http://dx.doi.org/10.1371/journal.pcbi.1000742 %V 6 %X Many genome-wide datasets are routinely generated to study different aspects of biological systems, but integrating them to obtain a coherent view of the underlying biology remains a challenge. We propose simultaneous clustering of multiple networks as a framework to integrate large-scale datasets on the interactions among and activities of cellular components. Specifically, we develop an algorithm JointCluster that finds sets of genes that cluster well in multiple networks of interest, such as coexpression networks summarizing correlations among the expression profiles of genes and physical networks describing protein-protein and protein-DNA interactions among genes or gene-products. Our algorithm provides an efficient solution to a well-defined problem of jointly clustering networks, using techniques that permit certain theoretical guarantees on the quality of the detected clustering relative to the optimal clustering. These guarantees coupled with an effective scaling heuristic and the flexibility to handle multiple heterogeneous networks make our method JointCluster an advance over earlier approaches. Simulation results showed JointCluster to be more robust than alternate methods in recovering clusters implanted in networks with high false positive rates. In systematic evaluation of JointCluster and some earlier approaches for combined analysis of the yeast physical network and two gene expression datasets under glucose and ethanol growth conditions, JointCluster discovers clusters that are more consistently enriched for various reference classes capturing different aspects of yeast biology or yield better coverage of the analysed genes. These robust clusters, which are supported across multiple genomic datasets and diverse reference classes, agree with known biology of yeast under these growth conditions, elucidate the genetic control of coordinated transcription, and enable functional predictions for a number of uncharacterized genes. The generation of high-dimensional datasets in the biological sciences has become routine (protein interaction, gene expression, and DNA/RNA sequence data, to name a few), stretching our ability to derive novel biological insights from them, with even less effort focused on integrating these disparate datasets available in the public domain. Hence a most pressing problem in the life sciences today is the development of algorithms to combine large-scale data on different biological dimensions to maximize our understanding of living systems. We present an algorithm for simultaneously clustering multiple biological networks to identify coherent sets of genes (clusters) underlying cellular processes. The algorithm allows theoretical guarantees on the quality of the detected clusters relative to the optimal clusters that are computationally infeasible to find, and could be applied to coexpression, protein interaction, protein-DNA networks, and other network types. When combining multiple physical and gene expression based networks in yeast, the clusters we identify are consistently enriched for reference classes capturing diverse aspects of biology, yield good coverage of the analysed genes, and highlight novel members in well-studied cellular processes.

@article{Narayanan2010Simultaneous, abstract = {Many genome-wide datasets are routinely generated to study different aspects of biological systems, but integrating them to obtain a coherent view of the underlying biology remains a challenge. We propose simultaneous clustering of multiple networks as a framework to integrate large-scale datasets on the interactions among and activities of cellular components. Specifically, we develop an algorithm {JointCluster} that finds sets of genes that cluster well in multiple networks of interest, such as coexpression networks summarizing correlations among the expression profiles of genes and physical networks describing protein-protein and {protein-DNA} interactions among genes or gene-products. Our algorithm provides an efficient solution to a well-defined problem of jointly clustering networks, using techniques that permit certain theoretical guarantees on the quality of the detected clustering relative to the optimal clustering. These guarantees coupled with an effective scaling heuristic and the flexibility to handle multiple heterogeneous networks make our method {JointCluster} an advance over earlier approaches. Simulation results showed {JointCluster} to be more robust than alternate methods in recovering clusters implanted in networks with high false positive rates. In systematic evaluation of {JointCluster} and some earlier approaches for combined analysis of the yeast physical network and two gene expression datasets under glucose and ethanol growth conditions, {JointCluster} discovers clusters that are more consistently enriched for various reference classes capturing different aspects of yeast biology or yield better coverage of the analysed genes. These robust clusters, which are supported across multiple genomic datasets and diverse reference classes, agree with known biology of yeast under these growth conditions, elucidate the genetic control of coordinated transcription, and enable functional predictions for a number of uncharacterized genes. The generation of high-dimensional datasets in the biological sciences has become routine (protein interaction, gene expression, and {DNA}/{RNA} sequence data, to name a few), stretching our ability to derive novel biological insights from them, with even less effort focused on integrating these disparate datasets available in the public domain. Hence a most pressing problem in the life sciences today is the development of algorithms to combine large-scale data on different biological dimensions to maximize our understanding of living systems. We present an algorithm for simultaneously clustering multiple biological networks to identify coherent sets of genes (clusters) underlying cellular processes. The algorithm allows theoretical guarantees on the quality of the detected clusters relative to the optimal clusters that are computationally infeasible to find, and could be applied to coexpression, protein interaction, {protein-DNA} networks, and other network types. When combining multiple physical and gene expression based networks in yeast, the clusters we identify are consistently enriched for reference classes capturing diverse aspects of biology, yield good coverage of the analysed genes, and highlight novel members in well-studied cellular processes.}, added-at = {2018-12-02T16:09:07.000+0100}, author = {Narayanan, Manikandan and Vetta, Adrian and Schadt, Eric E. and Zhu, Jun}, biburl = {https://www.bibsonomy.org/bibtex/288ddbd7aa519f3feb18def4c72150ed4/karthikraman}, citeulike-article-id = {7024472}, citeulike-linkout-0 = {http://dx.doi.org/10.1371/journal.pcbi.1000742}, citeulike-linkout-1 = {http://view.ncbi.nlm.nih.gov/pubmed/20419151}, citeulike-linkout-2 = {http://www.hubmed.org/display.cgi?uids=20419151}, day = 15, doi = {10.1371/journal.pcbi.1000742}, interhash = {ccdc220c6cf5abc0636c77c2c2e452e2}, intrahash = {88ddbd7aa519f3feb18def4c72150ed4}, issn = {1553-7358}, journal = {PLoS Comput Biol}, keywords = {clustering data-mining gene-expression integrated-modelling}, month = apr, number = 4, pages = {e1000742+}, pmid = {20419151}, posted-at = {2010-04-16 09:31:19}, priority = {5}, publisher = {Public Library of Science}, timestamp = {2018-12-02T16:09:07.000+0100}, title = {Simultaneous Clustering of Multiple Gene Expression and Physical Interaction Datasets}, url = {http://dx.doi.org/10.1371/journal.pcbi.1000742}, volume = 6, year = 2010 }