Identification of gene interactions associated with disease from gene expression data using synergy networks

BackgroundAnalysis of microarray data has been used for the inference of gene-gene interactions. If, however, the aim is the discovery of disease-related biological mechanisms, then the criterion for defining such interactions must be specifically linked to disease.ResultsHere we present a computational methodology that jointly analyzes two sets of microarray data, one in the presence and one in the absence of a disease, identifying gene pairs whose correlation with disease is due to cooperative, rather than independent, contributions of genes, using the recently developed information theoretic measure of synergy. High levels of synergy in gene pairs indicates possible membership of the two genes in a shared pathway and leads to a graphical representation of inferred gene-gene interactions associated with disease, in the form of a "synergy network." We apply this technique on a set of publicly available prostate cancer expression data and successfully validate our results, confirming that they cannot be due to pure chance and providing a biological explanation for gene pairs with exceptionally high synergy.ConclusionThus, synergy networks provide a computational methodology helpful for deriving "disease interactomes" from biological data. When coupled with additional biological knowledge, they can also be helpful for deciphering biological mechanisms responsible for disease.

[1]  M Vasei,et al.  Frequent high-level expression of the immunotherapeutic target Ep-CAM in colon, stomach, prostate and lung cancers , 2006, British Journal of Cancer.

[2]  P. Vogt,et al.  A role of the kinase mTOR in cellular transformation induced by the oncoproteins P3k and Akt. , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[3]  J. Cheville,et al.  AGR2, an androgen‐inducible secretory protein overexpressed in prostate cancer , 2005, Genes, chromosomes & cancer.

[4]  Adam A. Margolin,et al.  Reverse engineering of regulatory networks in human B cells , 2005, Nature Genetics.

[5]  M. Daly,et al.  PGC-1α-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes , 2003, Nature Genetics.

[6]  Peter H. A. Sneath,et al.  Numerical Taxonomy: The Principles and Practice of Numerical Classification , 1973 .

[7]  A. Pendergast Stress and death: breaking up the c-Abl/14-3-3 complex in apoptosis , 2005, Nature Cell Biology.

[8]  E. Lander,et al.  Gene expression correlates of clinical prostate cancer behavior. , 2002, Cancer cell.

[9]  Christian A. Rees,et al.  Systematic variation in gene expression patterns in human cancer cell lines , 2000, Nature Genetics.

[10]  Y. Benjamini,et al.  Resampling-based false discovery rate controlling multiple test procedures for correlated test statistics , 1999 .

[11]  I S Kohane,et al.  Mutual information relevance networks: functional genomic clustering using pairwise entropy measurements. , 1999, Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing.

[12]  Nir Friedman,et al.  Inferring Cellular Networks Using Probabilistic Graphical Models , 2004, Science.

[13]  H Kishino,et al.  Correspondence analysis of genes and tissue types and finding genetic links from microarray data. , 2000, Genome informatics. Workshop on Genome Informatics.

[14]  Louis Ragolia,et al.  Elevated L-PGDS activity contributes to PMA-induced apoptosis concomitant with downregulation of PI3-K. , 2003, American journal of physiology. Cell physiology.

[15]  A. Hasman,et al.  Probabilistic reasoning in intelligent systems: Networks of plausible inference , 1991 .

[16]  J. Ramalho,et al.  Oxidative stress upregulates ubiquitin proteasome pathway in retinal endothelial cells. , 2006, Molecular vision.

[17]  Jeffrey A. Magee,et al.  Expression profiling reveals hepsin overexpression in prostate cancer. , 2001, Cancer research.

[18]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[19]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[20]  Bart De Moor,et al.  Predicting the prognosis of breast cancer by integrating clinical and microarray data with Bayesian networks , 2006, ISMB.

[21]  John T. Wei,et al.  Integrative molecular concept modeling of prostate cancer progression , 2007, Nature Genetics.

[22]  T. Kinzy,et al.  The Translation Elongation Factor eEF1B Plays a Role in the Oxidative Stress Response Pathway , 2004, RNA biology.

[23]  David M. Miller,et al.  Computational inference of the molecular logic for synaptic connectivity in C. elegans , 2006, ISMB.

[24]  T. DeWeese,et al.  Loss-of-function of Nkx3.1 promotes increased oxidative damage in prostate carcinogenesis. , 2005, Cancer research.

[25]  T. Barrette,et al.  Mining for regulatory programs in the cancer transcriptome , 2005, Nature Genetics.

[26]  D. Koller,et al.  A module map showing conditional activity of expression modules in cancer , 2004, Nature Genetics.

[27]  D. di Bernardo,et al.  How to infer gene networks from expression profiles , 2007, Molecular systems biology.

[28]  Pablo Tamayo,et al.  Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles , 2005, Proceedings of the National Academy of Sciences of the United States of America.

[29]  Evangelia Papadimitriou,et al.  Hydrogen Peroxide Stimulates Proliferation and Migration of Human Prostate Cancer Cells through Activation of Activator Protein-1 and Up-regulation of the Heparin Affin Regulatory Peptide Gene* , 2005, Journal of Biological Chemistry.

[30]  R. de Caterina,et al.  Inhibition of Major Histocompatibility Complex Class II Gene Transcription by Nitric Oxide and Antioxidants* , 2002, The Journal of Biological Chemistry.

[31]  Inge Jonassen,et al.  ERG upregulation and related ETS transcription factors in prostate cancer. , 2007, International journal of oncology.

[32]  Debashis Ghosh,et al.  alpha-Methylacyl coenzyme A racemase as a tissue biomarker for prostate cancer. , 2002, JAMA.

[33]  Rafael A Irizarry,et al.  Exploration, normalization, and summaries of high density oligonucleotide array probe level data. , 2003, Biostatistics.

[34]  E. Farias,et al.  Cellular retinol-binding protein-I inhibits PI3K/Akt signaling through a retinoic acid receptor-dependent mechanism that regulates p85–p110 heterodimerization , 2005, Oncogene.

[35]  B. Suffoletto,et al.  Prostate-specific membrane antigen: a novel folate hydrolase in human prostatic carcinoma cells. , 1996, Clinical cancer research : an official journal of the American Association for Cancer Research.

[36]  M. Gleave,et al.  Protection of androgen‐dependent human prostate cancer cells from oxidative stress‐induced DNA damage by overexpression of clusterin and its modulation by androgen , 2004, The Prostate.

[37]  Dimitris Anastassiou,et al.  Inference of Disease-Related Molecular Logic from Systems-Based Microarray Analysis , 2006, PLoS Comput. Biol..

[38]  Y. Benjamini,et al.  Controlling the false discovery rate: a practical and powerful approach to multiple testing , 1995 .

[39]  D. Anastassiou Computational analysis of the synergy among multiple interacting genes , 2007, Molecular systems biology.

[40]  A. Chinnaiyan,et al.  Integrative analysis of the cancer transcriptome , 2005, Nature Genetics.

[41]  J. Welsh,et al.  Analysis of gene expression identifies candidate markers and pharmacological targets in prostate cancer. , 2001, Cancer research.

[42]  F. Torti,et al.  Ferritin and the response to oxidative stress. , 2001, Biochemical Journal.

[43]  Manel Esteller,et al.  Hypermethylation-associated Inactivation of the Cellular Retinol-Binding-Protein 1 Gene in Human Cancer. , 2002, Cancer research.

[44]  S. Lowe,et al.  Survival signalling by Akt and eIF4E in oncogenesis and cancer therapy , 2004, Nature.

[45]  Korbinian Strimmer,et al.  An empirical Bayes approach to inferring large-scale gene association networks , 2005, Bioinform..