Network Biology Resources

An annotated but not complete catalog of publications on disease models and computational model building and testing. Please send suggestions for updates to

**These present some of the most significant approaches and/or findings in the network field.

A: Studies that use network approaches to identify novel biological insights which are validated genetically/experimentally

**Bandyopadhyay et al. Rewiring of Genetic Networks in Response to DNA Damage. Science 2010.
Study of dynamic changes in genetic interactions in response to DNA damage. Kinases, phosphatases and transcription factors change in a way that would not be discovered by static profiling. They identify DNA repair pathways and find new roles for three new genes, which they also validate.

Basso et al. Reverse engineering of regulatory networks in human B cells. Nat Genet 2005
ARACNe method for reverse engineering networks. Validations performed showing accuracy of network.

Basso et al. Integrated biochemical and computational approach identifies BCL6 direct target genes controlling multiple pathways in normal germinal-center B cells. Blood 2009.
Discovered and validated a canonical target signature for BCL6 using ARACNe and other approaches.

Butte and Kohane. Creation and implications of a phenome-genome network. Nat Biotechnol. 2006
Shows how gene measurements and phenotypes can be extracted from the public repository of microarray data, and related for the discovery of genes associated across conditions.

**Carro et al. The transcriptional network for mesenchymal transformation of brain tumours. Nature 2010.
They built a glioma-specific expression network and identified a module that activates mesenchymal genes in glioma. They identified two central genes which they validated by showing that modulation of the genes induces stem cell differentiation to mesenchymal lineages. Also showed expression changes of the genes in tumors and predicted clinical outcome.

**Chen et al. Variations in DNA elucidate molecular networks that cause disease. Nature 2008
eQTL and coexpression modules are studied and the MEMN is identified as a causal network for obesity. Three genes are validated as previously unknown diabetes genes.

Emilsson et al. Genetics of gene expression and its effect on disease. Nature 2008.
eQTL and eSNPs are analyzed in blood and adipose in a large Icelandic cohort. A thorough characterization of cis and trans signals for gene expression. They also identify an obesity module that overlaps with the MEMN in mouse.

**Heinig et al. A trans-acting locus regulates an anti-viral expression network and type 1 diabetes risk. Nature 2011
Networks were built from seven rat tissues. An interferon-driven inflammatory network enriched for virus response genes was regulated by chromosome 15 in several tissues. They identified Ebi2 as regulating the network. The expression of several network genes correlated with type 1 diabetes risk in humans. They also found SNPs near the human EBI2 gene that affected the expression of EBI2.

Hong et al. Dissecting Early Differentially Expressed Genes in a Mixture of Differentiating Embryonic Stem Cells. PLOS Comp Biol 2009
They apply an algorithm to identify genes that regulate early differentiation of stem cells. Smarcad1 is identified as a novel gene controlling self-renewal and differentiation and they validate their findings successfully.

Lefebvre et al. A Human B Cell Interactome Identifies MYB and FOXM1 as Master Regulators of Proliferation in Germinal Centers. Mol Syst Biol 2010
Discovered and validated master regulators (MYB and FOXM1) of Germinal Center reaction in mature human B cells.

Musunura K et al. From noncoding variant to phenotype via SORT1 at the 1p13 cholesterol locus.  Nature 2010
eSNPs/causal networks leveraged to elucidate SORT1 as key driver gene at chromosome 1p13 locus

Palomero et al. NOTCH1 directly regulates MYC expression and controls oncogenic cell growth, Proc Natl Acad Sci 2008.
Discovered and validated NOTCH1 as a regulator of MYC metabolism control targets in T-ALL, using ARACNe.

**Ravasi et al.  An atlas of combinatorial transcriptional regulation in mouse and man.  Cell 2010
They build a global atlas of physical interactions between transcription factors. Highly connected transcription factors are broadly expressed across tissues. They show that TF combinations are important for cell fate and identify SMAD3/FLI1 as important for immunity development. They validate the interaction of those two TFs.
Schadt et al. Mapping the Genetic Architecture of Gene Expression in Human Liver. Plos Biol 2008
Studied eSNPs and transcript data in liver and identified a number of eSNP hotspots, i.e. affecting several expression traits. They used GWAS and previous mouse network data to prioritize the genes. They selected a SNP for type 1 diabetes that was near both ERBB1 and RPS26 and built networks around those two genes. The RPS26 subnetwork was enriched for T1D-associated genes, suggesting that RPS26 rather than ERBB1 was affected by the SNP. This is a good way to infer which genes in the vicinity of a SNP that are is most likely to be associated with the effect on disease.

**Suthram et al. Network-Based Elucidation of Human Disease Similarities Reveals Common Functional Modules Enriched for Pluripotent Drug Targets. PLoS Computational Biology. 2010
Studying the common protein-interaction networks across 54 diseases, finding the common core networks contain known drug targets, and the drugs that hit these targets are already known to be efficacious against more diseases.

Teslovich et al.  Biological, clinical and population relevance of 95 loci for blood lipids.  Nature 2010.
Transcriptional networks and eSNP/causality approach used to help identify genes underlying genetic loci driving blood lipids.

**Wang et al. Genome-wide identification of post-translational modulators of transcription factor activity in human B cell. Nat Biotechnol 2009.
Discovered and validated multiple novel modulators of MYC activity (experimentally validated modulators include STK38, HDAC1, MEF2B, and BHLHB2).

Workman et al. Systems Approach to Mapping DNA Damage Response Pathways. Science 2006.
They measured genomewide binding locations for 30 damage-related transcription factors (TFs) after exposure of yeast to DNA damage. A number of damage-specific binding motifs are identified. They validated their findings by identifying interactions for which the target changed expression in wild-type cells in response to DNA damage but was nonresponsive in cells lacking the TF.

**Yang et al. Validation of candidate causal genes for obesity that affect shared metabolic pathways and networks. Nat Genet 2009.
They take a causal network for obesity, from which a number of genes have previously been validated. Here they validate additional genes as important for obesity phenotypes.

Yang et al. Identification and validation of genes affecting aortic lesions in mice. JCI 2010
They used the BxH cross to identifiy 300 genes causal for aortic lesion in liver and adipose tissue. There was an overlap with the MEMN network (Chen et al.). They selected genes with eQTL overlapping clinical QTLs for aortic lesion. C3ar1 was of particular interest because it was in the MEMN and was suggested as a causal gene for obesity and insulin resistance. It was successfully validated.

**Zhang et al. Integrative Modeling Defines the Nova Splicing-Regulatory Network and Its Combinatorial Controls. Science 2010.
Used Bayesian networks across several data sets to study the targets of expression regulators. By integrating splice data, as well as RNA-binding and motif data they identified Nova as regulating 700 alternative splicing events. They also validated some interactions with Nova.

Zhao et al. The N-Myc-DLL3 cascade is suppressed by the ubiquitin ligase Huwe1 to inhibit proliferation and promote neurogenesis in the developing brain. Dev. Cell 2009.
Discovered HUWE1 as a major MINDy predicted modulator of N-MYC and DLL3 (ARACNe inferred target of N-MYC) as the key effector of HUWE1 induced accumulation of cyclin D1 in brain, causing morphogenesis defects in the embryo and cancer. Complete experimental validation of the findings

Zhidong et al.   Integrating siRNA and protein-protein interaction data to identify Edg5 as a type 2 diabetes gene.  Genome Research 2009.
Approach developed to integrate siRNA and protein-protein interaction data to provide a more meaningful context within which canonical pathways operate. Edg5 identified and validated as a novel type 2 diabetes gene. 

Zhong et al., Liver and Adipose Expression Associated SNPs Are Enriched for Association to Type 2 Diabetes. Plos Genetics 2010
They leveraged GWAS data by intersecting T2D SNPs that did not reach genome-wide significance with eSNP data from liver and adipose tissue and found an enrichment of T2D-associated SNPs in the intersect. They further intersected GWAS data with a causal T2D network and got even higher enrichment of T2D SNPs. They also validated malic enzyme 1. This paper shows the power of using networks to leverage global genomic data and to prioritize genes.

B: Studies that use network models to generate biological insights but do not provide independent validation.

Breitkreutz et al. A Global Protein Kinase and Phosphatase Interaction Network in Yeast. Science 2010.
They use mass spectrometry to analyze protein complexes in yeast and build a kinase-phosphatase network. They find new interactions for Cdc14 and Torc1.

Cadeiras et al. Drawing Networks of Rejection - A Systems Biological Approach to the Identification of Candidate Genes in Heart Transplantation, J.Cell. and Mol. Medicine 2010.
Validated targets of two TFs in cardiac allograft rejection but did not validate effect.

Crawford et al. The Diasporin Pathway: a tumor progression-related transcriptional network that predicts breast cancer survival. Clin Exp Metastasis 2008.
They studied extracellular matrix genes in breast cancer metastasis. They mapped eQTLs in inbred mice and found three eQTL and seven specific genes in the loci. They did cell line and in vivo validations showing that those seven genes indeed modulated tumor progression and induced a gene expression signature that could be used to predict survival in a human breast cancer cohort. The seven genes were identified through eQTL analyses (no network) although they built an interaction network used to highlight hubs. The validations were however mainly based on the eQTL and not on the network analyses.

Inouye et al. Metabonomic, transcriptomic, and genomic variation of a population cohort. Molecular Systems Biology 2010.
They used metabonomic, transcriptomic and genomic data from a large population cohort to build a network. A lipid-leukocyte module was identified that was highly associated with 80 out of 134 metabolites measured. They also inferred directed links using the genomic data and showed that metabolic status affect the coherency of the network. Although there are no specific validations, this is a nice description of the expression-metabolite interactions highlighting important novel module genes affecting metabolism.

Jerby et al. Computational reconstruction of tissue-specific metabolic models: application to human liver metabolism. Mol Systems Biol 2010.
The paper builds on a previously published generic stoichiometric model for metabolism. They integrate literature-based, transcriptomic, proteomic, metabolomic and phenotypic data to extend the generic model to a hepatic tissue-specific. They show that this model is better at predicting metabolic fluxes than the generic one.

Kirouac et al. Dynamic interaction networks in a hierarchically organized tissue. Mol syst Biol 2010.
They integrated literature-based data with expression data to characterize intercellular networks in stem cells. They identified a circuit of secreted factors that affect stem cell differentiation and identified a few key nodes. No really unexpected genes but they performed nice validations and also built a coherent model.

Lage et al., A human phenome-interactome network of protein complexes implicated in genetic disorders. Comp Biol 2007
Mapped genes encoding disease-associated proteins in a phenome-interactome network. They combined known protein-protein interactions and calculated similarities to infer links in the network. They used a Bayesian predictor that correctly predicted several known protein-disease relations, and they also claim to have identified a number of novel links that might be used to identify candidate genes for a range of diseases. No genetic or experimental validations.

Li et al. Identification of high-quality cancer prognostic markers and metastasis network modules. Nature Communications 2010.
They aimed at finding predictive signatures for cancer patients at low risk for metastases. They used tumor gene microarrays from breast cancer samples and applied a novel algorithm that successfully identified gene signatures ER+/ER+ subtypes. The signatures were combined into networks and they found that the genes from the signatures interacted with known tumor genes. Thus, they formed expression-based gene signature but nothing novel revealed by the network approach they used.

Lipshtat et al. Functions of Bifans in Context of Multiple Regulatory Motifs in Signaling Networks. Biophysical Journal 2008.
They studied bifan motifs, i.e. two nodes affecting two targets in concert. Differential equations were used to investigate synchronization and feedback loops among the motifs. Interesting to add feedback data, which is often missing in network studies but they only use simulations.

Ma, H et al. COSINE: Condition-SpecIfic sub-NEtwork identification using a global optimization method. Bioinformatics 2011
They implement a new method, COSINE, that analyzes both differential expression and gene-pair correlation. The addition of edge information is useful and they highlight a number of pathways in prostate cancer and obesity. The identified subnetworks that appear more biologically meaningful than those identified from other methods they used for comparison. However, no validations.

Mani et al. A systems biology approach to prediction of oncogenes and molecular perturbation targets in B-cell lymphomas. Mol Syst Biol 2008.
Built a network based on molecular interactions that become dysregulated in different tumors. They identified perturbation targets and predicted dysregulated interactions in lymphoma based on existing data. No novel genes or validations.

Narayanan et al. Simultaneous clustering of multiple gene expression and physical interaction datasets. PLoS Computational Biology 2010.
Novel network-based approach for combining gene

Pandey et al. An integrative multi-network and multi-classifier approach to predict genetic interactions. PLoS Comput Biol. 2010. 
Methodological network-based approach to identifying genetic interactions in yeast with some validations. Potentially applicable to higher organisms

Park et al. The impact of cellular networks on disease comorbidity. Mol Syst Biol 2009.
They integrated disease-gene and disease comorbidity data with cellular interaction data to build a network that can be used to infer new relationships between diseases and their mechanisms. Interesting approach but no validations.

Romanoski et al. Systems Genetics Analysis of Gene-by-Environment Interactions in Human Cells. Am J Human Gen 2010
They analyzed gene expression in response to oxidized phospholipids. They found a number of genes which were affected by loci both in cis and in trans in response to environmental perturbation. A few of the genes were validated as important mediators of cell stress.

Vaske et al. Inference of patient-specific pathway activities from multi-dimensional cancer genomics data using PARADIGM. Bioinformatics 2010.
Method for identifying pathway activity in patients.

Xiang et al. A systems biology approach to transcription factor binding site prediction. PLoS One 2010.
Method to recover novel binding sites for TFs based on network biology, with significant experimental validation of novel sites, both individually and in combination.

Yang et al. Systematic genetic and genomic analysis of cytochrome P450 enzyme activities in human liver. Genome Research 2010.
Among first papers to use an integrative genomics, network-based approach to elucidate the networks in which p450’s operate.  A number of putative causal relationships identified. expression and physical interaction datasets to provide a more integrated network view of systems of interest.  

Zhang et al.  A Bayesian partition method for detecting pleiotropic and epistatic eQTL modules.  PLoS Computational Biology 2010.
 Method developed to simultaneously detect subnetworks associated with traits of interest and the underlying genetic model driving that subnetwork, with interactions among genetic loci allowed.  Led to novel detections of interacting loci that had previously gone unreported.

Zoppoli et al. TimeDelay-ARACNE: Reverse engineering of gene networks from time-course data by an information theoretic approach. BMC Bioinformatics 2010.
They used ARACNE for time-series data and mapped the dependencies of two genes at different time points and can e.g. study feedback loops. It is compared with e.g. dynamic Bayesian modelling in existing data sets and performs well. No novel biological findings.


Bullmore et al. Complex brain networks: graph theoretical analysis of structural and functional systems. Nature Reviews Neuroscience 2009.
Review and theory paper on graph theory.

Cancer Target Discovery and Development Network. Towards patient-based cancer therapeutics. Nat Biotech 2010.
This white paper illustrates how network biology, chemical biology, and functional biology can transform patient care.

Derry et al. Developing Predictive Molecular Maps of Human Disease through Community-based Modeling. Nature Precedings 2011.
A commentary paper on the vision for the Sage Platform and Commons.

Dudley et al. Drug discovery in a multidimensional world: Systems, patterns, and networks.  Journal of Cardiovascular Translational Research 2010.

Friend, and Ideker.  POINT: Are we prepared for the future doctor visit? Nat Biotech 2011

Schadt EE et al. A network view of disease and compound screening. Nat Rev Drug Discov. 2009.

Schadt E. Molecular networks as sensors and drivers of common human diseases. Nature 2009.

Vidal et al. Interactome Networks and Human Disease. Cell 2011.
A review paper that among other things describe disease map and interactome data that Vidal and Barabasi have been conducting.


^ Back to top