Investigating cellular network heterogeneity and modularity in cancer: a network entropy and unbalanced motif approach

BackgroundCancer is increasingly recognized as a cellular system phenomenon that is attributed to the accumulation of genetic or epigenetic alterations leading to the perturbation of the molecular network architecture. Elucidation of network properties that can characterize tumor initiation and progression, or pinpoint the molecular targets related to the drug sensitivity or resistance, is therefore of critical importance for providing systems-level insights into tumorigenesis and clinical outcome in the molecularly targeted cancer therapy.ResultsIn this study, we developed a network-based framework to quantitatively examine cellular network heterogeneity and modularity in cancer. Specifically, we constructed gene co-expressed protein interaction networks derived from large-scale RNA-Seq data across 8 cancer types generated in The Cancer Genome Atlas (TCGA) project. We performed gene network entropy and balanced versus unbalanced motif analysis to investigate cellular network heterogeneity and modularity in tumor versus normal tissues, different stages of progression, and drug resistant versus sensitive cancer cell lines. We found that tumorigenesis could be characterized by a significant increase of gene network entropy in all of the 8 cancer types. The ratio of the balanced motifs in normal tissues is higher than that of tumors, while the ratio of unbalanced motifs in tumors is higher than that of normal tissues in all of the 8 cancer types. Furthermore, we showed that network entropy could be used to characterize tumor progression and anticancer drug responses. For example, we found that kinase inhibitor resistant cancer cell lines had higher entropy compared to that of sensitive cell lines using the integrative analysis of microarray gene expression and drug pharmacological data collected from the Genomics of Drug Sensitivity in Cancer database. In addition, we provided potential network-level evidence that smoking might increase cancer cellular network heterogeneity and further contribute to tyrosine kinase inhibitor (e.g., gefitinib) resistance.ConclusionIn summary, we demonstrated that network properties such as network entropy and unbalanced motifs associated with tumor initiation, progression, and anticancer drug responses, suggesting new potential network-based prognostic and predictive measure in cancer.

[1]  T. Hubbard,et al.  A census of human cancer genes , 2004, Nature Reviews Cancer.

[2]  T. Hampton,et al.  The Cancer Genome Atlas , 2020, Indian Journal of Medical and Paediatric Oncology.

[3]  Evan Bolton,et al.  Database resources of the National Center for Biotechnology Information , 2017, Nucleic Acids Res..

[4]  Gregory D. Schuler,et al.  Database resources of the National Center for Biotechnology Information: update , 2004, Nucleic acids research.

[5]  Simone Severini,et al.  Intra-Tumour Signalling Entropy Determines Clinical Outcome in Breast and Lung Cancer , 2015, PLoS Comput. Biol..

[6]  Zhongming Zhao,et al.  Studying tumorigenesis through network evolution and somatic mutational perturbations in the cancer interactome. , 2014, Molecular biology and evolution.

[7]  Benjamin J. Raphael,et al.  Mutational landscape and significance across 12 major cancer types , 2013, Nature.

[8]  W. Lim,et al.  Defining Network Topologies that Can Achieve Biochemical Adaptation , 2009, Cell.

[9]  G. Bianconi,et al.  Differential network entropy reveals cancer system hallmarks , 2012, Scientific Reports.

[10]  C. Waddington,et al.  The strategy of the genes , 1957 .

[11]  C. Altafini,et al.  Computing global structural balance in large-scale signed social networks , 2011, Proceedings of the National Academy of Sciences.

[12]  S. Ramaswamy,et al.  Systematic identification of genomic markers of drug sensitivity in cancer cells , 2012, Nature.

[13]  Zhongming Zhao,et al.  ccmGDB: a database for cancer cell metabolism genes , 2015, Nucleic Acids Res..

[14]  D. Pe’er,et al.  Principles and Strategies for Developing Network Models in Cancer , 2011, Cell.

[15]  R. Gillies,et al.  Evolutionary dynamics of carcinogenesis and why targeted therapy does not work , 2012, Nature Reviews Cancer.

[16]  Peilin Jia,et al.  VarWalker: Personalized Mutation Network Analysis of Putative Cancer Genes from Next-Generation Sequencing Data , 2014, PLoS Comput. Biol..

[17]  Hyeong Jun An,et al.  Estimating the size of the human interactome , 2008, Proceedings of the National Academy of Sciences.

[18]  Daniel V. Veres,et al.  Cancer stem cells display extremely large evolvability: alternating plastic and rigid networks as a potential Mechanism: network models, novel therapeutic target strategies, and the contributions of hypoxia, inflammation and cellular senescence. , 2013, Seminars in cancer biology.

[19]  Li Ding,et al.  Genomic Landscape of Non-Small Cell Lung Cancer in Smokers and Never-Smokers , 2012, Cell.

[20]  Steven J. M. Jones,et al.  Genomic Classification of Cutaneous Melanoma , 2015, Cell.

[21]  Carlo C. Maley,et al.  Cancer in Light of Experimental Evolution , 2012, Current Biology.

[22]  Zhongming Zhao,et al.  Systematic Prioritization of Druggable Mutations in ∼5000 Genomes Across 16 Cancer Types Using a Structural Genomics-based Approach* , 2015, Molecular & Cellular Proteomics.

[23]  Yang Xiang,et al.  Weighted Frequent Gene Co-expression Network Mining to Identify Genes Involved in Genome Stability , 2012, PLoS Comput. Biol..

[24]  Peng Qiu,et al.  TCGA-Assembler: open-source software for retrieving and processing TCGA data , 2014, Nature Methods.

[25]  Peter Kuhn,et al.  Entropy, complexity, and Markov diagrams for random walk cancer models , 2014, Scientific Reports.

[26]  N. Dubrawsky.  Cancer statistics , 1989, CA: a cancer journal for clinicians.

[27]  Kornelia Polyak,et al.  Heterogeneity in breast cancer. , 2011, The Journal of clinical investigation.

[28]  Arun K. Ramani,et al.  How complete are current yeast and human protein-interaction networks? , 2006, Genome Biology.

[29]  Min Hwan Kim,et al.  Impact of cigarette smoking on response to epidermal growth factor receptor (EGFR)-tyrosine kinase inhibitors in lung adenocarcinoma with activating EGFR mutations. , 2014, Lung cancer.

[30]  Gary D Bader,et al.  Pathway and network analysis of cancer genomes , 2015, Nature Methods.

[31]  Simone Severini,et al.  Increased entropy of signal transduction in the cancer metastasis phenotype , 2010, BMC Systems Biology.

[32]  Chuang Liu,et al.  A Gene Gravity Model for the Evolution of Cancer Genomes: A Study of 3,000 Cancer Genomes across 9 Cancer Types , 2015, PLoS Comput. Biol..

[33]  Rune Linding,et al.  Navigating cancer network attractors for tumor-specific therapy , 2012, Nature Biotechnology.

[34]  Yanhui Hu,et al.  Integrating protein-protein interaction networks with phenotypes reveals signs of interactions , 2013, Nature Methods.

[35]  C. Klein.  Selection and adaptation during metastatic cancer progression , 2013, Nature.

[36]  Sridhar Ramaswamy,et al.  Genomics of Drug Sensitivity in Cancer (GDSC): a resource for therapeutic biomarker discovery in cancer cells , 2012, Nucleic Acids Res..

[37]  Shane J. Neph,et al.  Developmental Fate and Cellular Maturity Encoded in Human Regulatory DNA Landscapes , 2013, Cell.

[38]  T. Goldkorn,et al.  Cigarette Smoke Induces Aberrant EGF Receptor Activation That Mediates Lung Cancer Development and Resistance to Tyrosine Kinase Inhibitors , 2012, Molecular Cancer Therapeutics.

[39]  C. Swanton,et al.  Cancer chromosomal instability: therapeutic and diagnostic challenges , 2012, EMBO reports.

[40]  Peilin Jia,et al.  Patterns and processes of somatic mutations in nine major cancers , 2014, BMC Medical Genomics.

[41]  F. Harary,et al.  STRUCTURAL BALANCE: A GENERALIZATION OF HEIDER'S THEORY1 , 1977 .

[42]  A. Jemal,et al.  Cancer statistics, 2014 , 2014, CA: a cancer journal for clinicians.

[43]  M. Stratton,et al.  Abstract 2206: Genomics of Drug Sensitivity in Cancer (GDSC): A resource for therapeutic biomarker discovery in cancer cells. , 2013 .

[44]  N. McGranahan,et al.  The causes and consequences of genetic heterogeneity in cancer evolution , 2013, Nature.

[45]  Zhongming Zhao,et al.  Quantitative network mapping of the human kinome interactome reveals new clues for rational kinase inhibitor discovery and individualized cancer therapy , 2014, Oncotarget.