Detecting hierarchical modularity in biological networks

Compression-based inference of network motif sets

Physical and functional constraints on biological networks lead to complex topological patterns across multiple scales in their organization. A particular type of higher-order network feature that has received considerable interest is network motifs, defined as statistically regular subgraphs. These may implement fundamental logical and computational circuits and are referred to as "building blocks of complex networks". Their well-defined structures and small sizes also enable the testing of their functions in synthetic and natural biological experiments. Here, we develop a framework for motif mining based on lossless network compression using subgraph contractions. This provides an alternative definition of motif significance which allows us to compare different motifs and select the collectively most significant set of motifs as well as other prominent network features in terms of their combined compression of the network. Our approach inherently accounts for multiple testing and correlations between subgraphs and does not rely on a priori specification of an appropriate null model. It thus overcomes common problems in hypothesis testing-based motif analysis and guarantees robust statistical inference. We validate our methodology on numerical data and then apply it on synaptic-resolution biological neural networks, as a medium for comparative connectomics, by evaluating their respective compressibility and characterize their inferred circuit motifs.

vCSF Danger-associated Molecular Patterns After Traumatic and Nontraumatic Acute Brain Injury: A Prospective Study

BACKGROUND

Danger-associated molecular patterns (DAMPs) may be implicated in the pathophysiological pathways associated with an unfavorable outcome after acute brain injury (ABI).

METHODS

We collected samples of ventricular cerebrospinal fluid (vCSF) for 5 days in 50 consecutive patients at risk of intracranial hypertension after traumatic and nontraumatic ABI. Differences in vCSF protein expression over time were evaluated using linear models and selected for functional network analysis using the PANTHER and STRING databases. The primary exposure of interest was the type of brain injury (traumatic vs. nontraumatic), and the primary outcome was the vCSF expression of DAMPs. Secondary exposures of interest included the occurrence of intracranial pressure ≥20 or ≥ 30 mm Hg during the 5 days post-ABI, intensive care unit (ICU) mortality, and neurological outcome (assessed using the Glasgow Outcome Score) at 3 months post-ICU discharge. Secondary outcomes included associations of these exposures with the vCSF expression of DAMPs.

RESULTS

A network of 6 DAMPs ( DAMP_trauma ; protein-protein interaction [PPI] P =0.04) was differentially expressed in patients with ABI of traumatic origin compared with those with nontraumatic ABI. ABI patients with intracranial pressure ≥30 mm Hg differentially expressed a set of 38 DAMPS ( DAMP_ICP30 ; PPI P < 0.001). Proteins in DAMP_ICP30 are involved in cellular proteolysis, complement pathway activation, and post-translational modifications. There were no relationships between DAMP expression and ICU mortality or unfavorable versus favorable outcomes.

CONCLUSIONS

Specific patterns of vCSF DAMP expression differentiated between traumatic and nontraumatic types of ABI and were associated with increased episodes of severe intracranial hypertension.

Defining hierarchical protein interaction networks from spectral analysis of bacterial proteomes

Cellular behaviors emerge from layers of molecular interactions: proteins interact to form complexes, pathways, and phenotypes. We show that hierarchical networks of protein interactions can be defined from the statistical pattern of proteome variation measured across thousands of diverse bacteria and that these networks reflect the emergence of complex bacterial phenotypes. Our results are validated through gene-set enrichment analysis and comparison to existing experimentally derived databases. We demonstrate the biological utility of our approach by creating a model of motility in Pseudomonas aeruginosa and using it to identify a protein that affects pilus-mediated motility. Our method, SCALES (Spectral Correlation Analysis of Layered Evolutionary Signals), may be useful for interrogating genotype-phenotype relationships in bacteria.

Computational Network Inference for Bacterial Interactomics

Since the large-scale experimental characterization of protein–protein interactions (PPIs) is not possible for all species, several computational PPI prediction methods have been developed that harness existing data from other species. While PPI network prediction has been extensively used in eukaryotes, microbial network inference has lagged behind. However, bacterial interactomes can be built using the same principles and techniques; in fact, several methods are better suited to bacterial genomes. These predicted networks allow systems-level analyses in species that lack experimental interaction data. This review describes the current network inference and analysis techniques and summarizes the use of computationally-predicted microbial interactomes to date.

Network Theoretical Approach to Explore Factors Affecting Signal Propagation and Stability in Dementia’s Protein-Protein Interaction Network

Dementia—a syndrome affecting human cognition—is a major public health concern given to its rising prevalence worldwide. Though multiple research studies have analyzed disorders such as Alzheimer’s disease and Frontotemporal dementia using a systems biology approach, a similar approach to dementia syndrome as a whole is required. In this study, we try to find the high-impact core regulating processes and factors involved in dementia’s protein–protein interaction network. We also explore various aspects related to its stability and signal propagation. Using gene interaction databases such as STRING and GeneMANIA, a principal dementia network (PDN) consisting of 881 genes and 59,085 interactions was achieved. It was assortative in nature with hierarchical, scale-free topology enriched in various gene ontology (GO) categories and KEGG pathways, such as negative and positive regulation of apoptotic processes, macroautophagy, aging, response to drug, protein binding, etc. Using a clustering algorithm (Louvain method of modularity maximization) iteratively, we found a number of communities at different levels of hierarchy in PDN consisting of 95 “motif-localized hubs”, out of which, 7 were present at deepest level and hence were key regulators (KRs) of PDN (HSP90AA1, HSP90AB1, EGFR, FYN, JUN, CELF2 and CTNNA3). In order to explore aspects of network’s resilience, a knockout (of motif-localized hubs) experiment was carried out. It changed the network’s topology from a hierarchal scale-free topology to scale-free, where independent clusters exhibited greater control. Additionally, network experiments on interaction of druggable genome and motif-localized hubs were carried out where UBC, EGFR, APP, CTNNB1, NTRK1, FN1, HSP90AA1, MDM2, VCP, CTNNA1 and GRB2 were identified as hubs in the resultant network (RN). We finally concluded that stability and resilience of PDN highly relies on motif-localized hubs (especially those present at deeper levels), making them important therapeutic intervention candidates. HSP90AA1, involved in heat shock response (and its master regulator, i.e., HSF1), and EGFR are most important genes in pathology of dementia apart from KRs, given their presence as KRs as well as hubs in RN.

Extended graphical lasso for multiple interaction networks for high dimensional omics data

There has been a spate of interest in association networks in biological and medical research, for example, genetic interaction networks. In this paper, we propose a novel method, the extended joint hub graphical lasso (EDOHA), to estimate multiple related interaction networks for high dimensional omics data across multiple distinct classes. To be specific, we construct a convex penalized log likelihood optimization problem and solve it with an alternating direction method of multipliers (ADMM) algorithm. The proposed method can also be adapted to estimate interaction networks for high dimensional compositional data such as microbial interaction networks. The performance of the proposed method in the simulated studies shows that EDOHA has remarkable advantages in recognizing class-specific hubs than the existing comparable methods. We also present three applications of real datasets. Biological interpretations of our results confirm those of previous studies and offer a more comprehensive understanding of the underlying mechanism in disease.

Quantifying the compressibility of complex networks

Many complex networks depend upon biological entities for their preservation. Such entities, from human cognition to evolution, must first encode and then replicate those networks under marked resource constraints. Networks that survive are those that are amenable to constrained encoding-or, in other words, are compressible. But how compressible is a network? And what features make one network more compressible than another? Here, we answer these questions by modeling networks as information sources before compressing them using rate-distortion theory. Each network yields a unique rate-distortion curve, which specifies the minimal amount of information that remains at a given scale of description. A natural definition then emerges for the compressibility of a network: the amount of information that can be removed via compression, averaged across all scales. Analyzing an array of real and model networks, we demonstrate that compressibility increases with two common network properties: transitivity (or clustering) and degree heterogeneity. These results indicate that hierarchical organization-which is characterized by modular structure and heterogeneous degrees-facilitates compression in complex networks. Generally, our framework sheds light on the interplay between a network's structure and its capacity to be compressed, enabling investigations into the role of compression in shaping real-world networks.

Shannon entropy approach reveals relevant genes in Alzheimer's disease

Alzheimer’s disease (AD) is the most common type of dementia and affects millions of people worldwide. Since complex diseases are often the result of combinations of gene interactions, microarray data and gene co-expression analysis can provide tools for addressing complexity. Our study aimed to find groups of interacting genes that are relevant in the development of AD. In this perspective, we implemented a method proposed in a previous work to detect gene communities linked to AD. Our strategy combined co-expression network analysis with the study of Shannon entropy of the betweenness. We analyzed the publicly available GSE1297 dataset, achieved from the GEO database in NCBI, containing hippocampal gene expression of 9 control and 22 AD human subjects. Co-expressed genes were clustered into different communities. Two communities of interest (composed by 72 and 39 genes) were found by calculating the correlation coefficient between communities and clinical features. The detected communities resulted stable, replicated on two independent datasets and mostly enriched in pathways closely associated with neuro-degenative diseases. A comparison between our findings and other module detection techniques showed that the detected communities were more related to AD phenotype. Lastly, the hub genes within the two communities of interest were identified by means of a centrality analysis and a bootstrap procedure. The communities of the hub genes presented even stronger correlation with clinical features. These findings and further explorations on the detected genes could shed light on the genetic aspects related with physiological aspects of Alzheimer’s disease.

Emergence of Hierarchical Modularity in Evolving Networks Uncovered by Phylogenomic Analysis

Networks describe how parts associate with each other to form integrated systems which often have modular and hierarchical structure. In biology, network growth involves two processes, one that unifies and the other that diversifies. Here, we propose a biphasic (bow-tie) theory of module emergence. In the first phase, parts are at first weakly linked and associate variously. As they diversify, they compete with each other and are often selected for performance. The emerging interactions constrain their structure and associations. This causes parts to self-organize into modules with tight linkage. In the second phase, variants of the modules diversify and become new parts for a new generative cycle of higher level organization. The paradigm predicts the rise of hierarchical modularity in evolving networks at different timescales and complexity levels. Remarkably, phylogenomic analyses uncover this emergence in the rewiring of metabolomic and transcriptome-informed metabolic networks, the nanosecond dynamics of proteins, and evolving networks of metabolism, elementary functionomes, and protein domain organization.

A complex network approach reveals a pivotal substructure of genes linked to schizophrenia

Research on brain disorders with a strong genetic component and complex heritability, such as schizophrenia, has led to the development of brain transcriptomics. This field seeks to gain a deeper understanding of gene expression, a key factor in exploring further research issues. Our study focused on how genes are associated amongst each other. In this perspective, we have developed a novel data-driven strategy for characterizing genetic modules, i.e., clusters of strongly interacting genes. The aim was to uncover a pivotal community of genes linked to a target gene for schizophrenia. Our approach combined network topological properties with information theory to highlight the presence of a pivotal community, for a specific gene, and to simultaneously assess the information content of partitions with the Shannon’s entropy based on betweenness. We analyzed the publicly available BrainCloud dataset containing post-mortem gene expression data and focused on the Dopamine D2 receptor, encoded by the DRD2 gene. We used four different community detection algorithms to evaluate the consistence of our approach. A pivotal DRD2 community emerged for all the procedures applied, with a considerable reduction in size, compared to the initial network. The stability of the results was confirmed by a Dice index ≥80% within a range of tested parameters. The detected community was also the most informative, as it represented an optimization of the Shannon entropy. Lastly, we verified the strength of connection of the DRD2 community, which was stronger than any other randomly selected community and even more so than the Weighted Gene Co-expression Network Analysis module, commonly considered the standard approach for such studies. This finding substantiates the conclusion that the detected community represents a more connected and informative cluster of genes for the DRD2 community, and therefore better elucidates the behavior of this module of strongly related DRD2 genes. Because this gene plays a relevant role in Schizophrenia, this finding of a more specific DRD2 community will improve the understanding of the genetic factors related with this disorder.

Patients experiencing statin-induced myalgia exhibit a unique program of skeletal muscle gene expression following statin re-challenge

Statins, the 3-hydroxy-3-methyl-glutaryl (HMG)-CoA reductase inhibitors, are widely prescribed for treatment of hypercholesterolemia. Although statins are generally well tolerated, up to ten percent of statin-treated patients experience myalgia symptoms, defined as muscle pain without elevated creatinine phosphokinase (CPK) levels. Myalgia is the most frequent reason for discontinuation of statin therapy. The mechanisms underlying statin myalgia are not clearly understood. To elucidate changes in gene expression associated with statin myalgia, we compared profiles of gene expression in skeletal muscle biopsies from patients with statin myalgia who were undergoing statin re-challenge (cases) versus those of statin-tolerant controls. A robust separation of case and control cohorts was revealed by Principal Component Analysis of differentially expressed genes (DEGs). To identify putative gene expression and metabolic pathways that may be perturbed in skeletal muscles of patients with statin myalgia, we subjected DEGs to Ingenuity Pathways (IPA) and DAVID (Database for Annotation, Visualization and Integrated Discovery) analyses. The most prominent pathways altered by statins included cellular stress, apoptosis, cell senescence and DNA repair (TP53, BARD1, Mre11 and RAD51); activation of pro-inflammatory immune response (CXCL12, CST5, POU2F1); protein catabolism, cholesterol biosynthesis, protein prenylation and RAS-GTPase activation (FDFT1, LSS, TP53, UBD, ATF2, H-ras). Based on these data we tentatively conclude that persistent myalgia in response to statins may emanate from cellular stress underpinned by mechanisms of post-inflammatory repair and regeneration. We also posit that this subset of individuals is genetically predisposed to eliciting altered statin metabolism and/or increased end-organ susceptibility that lead to a range of statin-induced myopathies. This mechanistic scenario is further bolstered by the discovery that a number of single nucleotide polymorphisms (e.g., SLCO1B1, SLCO2B1 and RYR2) associated with statin myalgia and myositis were observed with increased frequency among patients with statin myalgia.

Comparison of module detection algorithms in protein networks and investigation of the biological meaning of predicted modules

Background

It is generally acknowledged that a functional understanding of a biological system can only be obtained by an understanding of the collective of molecular interactions in form of biological networks. Protein networks are one particular network type of special importance, because proteins form the functional base units of every biological cell. On a mesoscopic level of protein networks, modules are of significant importance because these building blocks may be the next elementary functional level above individual proteins allowing to gain insight into fundamental organizational principles of biological cells.

Results

In this paper, we provide a comparative analysis of five popular and four novel module detection algorithms. We study these module prediction methods for simulated benchmark networks as well as 10 biological protein interaction networks (PINs). A particular focus of our analysis is placed on the biological meaning of the predicted modules by utilizing the Gene Ontology (GO) database as gold standard for the definition of biological processes. Furthermore, we investigate the robustness of the results by perturbing the PINs simulating in this way our incomplete knowledge of protein networks.

Conclusions

Overall, our study reveals that there is a large heterogeneity among the different module prediction algorithms if one zooms-in the biological level of biological processes in the form of GO terms and all methods are severely affected by a slight perturbation of the networks. However, we also find pathways that are enriched in multiple modules, which could provide important information about the hierarchical organization of the system.

A combination of gene expression ranking and co-expression network analysis increases discovery rate in large-scale mutant screens for novel Arabidopsis thaliana abiotic stress genes

As challenges to food security increase, the demand for lead genes for improving crop production is growing. However, genetic screens of plant mutants typically yield very low frequencies of desired phenotypes. Here, we present a powerful computational approach for selecting candidate genes for screening insertion mutants. We combined ranking of Arabidopsis thaliana regulatory genes according to their expression in response to multiple abiotic stresses (Multiple Stress [MST] score), with stress-responsive RNA co-expression network analysis to select candidate multiple stress regulatory (MSTR) genes. Screening of 62 T-DNA insertion mutants defective in candidate MSTR genes, for abiotic stress germination phenotypes yielded a remarkable hit rate of up to 62%; this gene discovery rate is 48-fold greater than that of other large-scale insertional mutant screens. Moreover, the MST score of these genes could be used to prioritize them for screening. To evaluate the contribution of the co-expression analysis, we screened 64 additional mutant lines of MST-scored genes that did not appear in the RNA co-expression network. The screening of these MST-scored genes yielded a gene discovery rate of 36%, which is much higher than that of classic mutant screens but not as high as when picking candidate genes from the co-expression network. The MSTR co-expression network that we created, AraSTressRegNet is publicly available at http://netbio.bgu.ac.il/arnet. This systems biology-based screening approach combining gene ranking and network analysis could be generally applicable to enhancing identification of genes regulating additional processes in plants and other organisms provided that suitable transcriptome data are available.

Elements of the cellular metabolic structure

A large number of studies have demonstrated the existence of metabolic covalent modifications in different molecular structures, which are able to store biochemical information that is not encoded by DNA. Some of these covalent mark patterns can be transmitted across generations (epigenetic changes). Recently, the emergence of Hopfield-like attractor dynamics has been observed in self-organized enzymatic networks, which have the capacity to store functional catalytic patterns that can be correctly recovered by specific input stimuli. Hopfield-like metabolic dynamics are stable and can be maintained as a long-term biochemical memory. In addition, specific molecular information can be transferred from the functional dynamics of the metabolic networks to the enzymatic activity involved in covalent post-translational modulation, so that determined functional memory can be embedded in multiple stable molecular marks. The metabolic dynamics governed by Hopfield-type attractors (functional processes), as well as the enzymatic covalent modifications of specific molecules (structural dynamic processes) seem to represent the two stages of the dynamical memory of cellular metabolism (metabolic memory). Epigenetic processes appear to be the structural manifestation of this cellular metabolic memory. Here, a new framework for molecular information storage in the cell is presented, which is characterized by two functionally and molecularly interrelated systems: a dynamic, flexible and adaptive system (metabolic memory) and an essentially conservative system (genetic memory). The molecular information of both systems seems to coordinate the physiological development of the whole cell.

The biological networks in studying cell signal transduction complexity: The examples of sperm capacitation and of endocannabinoid system

Cellular signal transduction is a complex phenomenon, which plays a central role in cell surviving and adaptation. The great amount of molecular data to date present in literature, together with the adoption of high throughput technologies, on the one hand, made available to scientists an enormous quantity of information, on the other hand, failed to provide a parallel increase in the understanding of biological events. In this context, a new discipline arose, the systems biology, aimed to manage the information with a computational modeling-based approach. In particular, the use of biological networks has allowed the making of huge progress in this field.

Here we discuss two possible application of the use of biological networks to explore cell signaling: the study of the architecture of signaling systems that cooperate in determining the acquisition of a complex cellular function (as it is the case of the process of activation of spermatozoa) and the organization of a single specific signaling systems expressed by different cells in different tissues (i.e. the endocannabinoid system). In both the cases we have found that the networks follow a scale free and small world topology, likely due to the evolutionary advantage of robustness against random damages, fastness and specific of information processing, and easy navigability.

The search for organizing principles as a cure against reductionism in systems medicine

Biological complexity has forced scientists to develop highly reductive approaches, with an ever-increasing degree of specialization. As a consequence, research projects have become fragmented, and their results strongly dependent on the experimental context. The general research question, that originally motivated these projects, appears to have been forgotten in many highly specialized research programmes. We here investigate the prospects for use of an old regulative ideal from systems theory to describe the organization of cellular systems 'in general' by identifying key concepts, challenges and strategies to pursue the search for organizing principles. We argue that there is no tension between the complexity of biological systems and the search for organizing principles. On the contrary, it is the complexity of organisms and the current level of techniques and knowledge that urge us to renew the search for organizing principles in order to meet the challenges that are arise from reductive approaches in systems medicine. Reductive approaches, as important and inevitable as they are, should be complemented by an integrative strategy that de-contextualizes through abstractions, and thereby generalizes results.

Hierarchical modular structure identification with its applications in gene coexpression networks

Network module (community) structure has been a hot research topic in recent years. Many methods have been proposed for module detection and identification. Hierarchical structure of modules is shown to exist in many networks such as biological networks and social networks. Compared to the partitional module identification methods, less research is done on the inference of hierarchical modular structure. In this paper, we propose a method for constructing the hierarchical modular structure based on the stochastic block model. Statistical tests are applied to test the hierarchical relations between different modules. We give both artificial networks and real data examples to illustrate the performance of our approach. Application of the proposed method to yeast gene coexpression network shows that it does have a hierarchical modular structure with the modules on different levels corresponding to different gene functions.

Characterizing protein domain associations by Small-molecule ligand binding

BACKGROUND

Protein domains are evolutionarily conserved building blocks for protein structure and function, which are conventionally identified based on protein sequence or structure similarity. Small molecule binding domains are of great importance for the recognition of small molecules in biological systems and drug development. Many small molecules, including drugs, have been increasingly identified to bind to multiple targets, leading to promiscuous interactions with protein domains. Thus, a large scale characterization of the protein domains and their associations with respect to small-molecule binding is of particular interest to system biology research, drug target identification, as well as drug repurposing.

METHODS

We compiled a collection of 13,822 physical interactions of small molecules and protein domains derived from the Protein Data Bank (PDB) structures. Based on the chemical similarity of these small molecules, we characterized pairwise associations of the protein domains and further investigated their global associations from a network point of view.

RESULTS

We found that protein domains, despite lack of similarity in sequence and structure, were comprehensively associated through binding the same or similar small-molecule ligands. Moreover, we identified modules in the domain network that consisted of closely related protein domains by sharing similar biochemical mechanisms, being involved in relevant biological pathways, or being regulated by the same cognate cofactors.

CONCLUSIONS

A novel protein domain relationship was identified in the context of small-molecule binding, which is complementary to those identified by traditional sequence-based or structure-based approaches. The protein domain network constructed in the present study provides a novel perspective for chemogenomic study and network pharmacology, as well as target identification for drug repurposing.

Functional modules, structural topology, and optimal activity in metabolic networks

Modular organization in biological networks has been suggested as a natural mechanism by which a cell coordinates its metabolic strategies for evolving and responding to environmental perturbations. To understand how this occurs, there is a need for developing computational schemes that contribute to integration of genomic-scale information and assist investigators in formulating biological hypotheses in a quantitative and systematic fashion. In this work, we combined metabolome data and constraint-based modeling to elucidate the relationships among structural modules, functional organization, and the optimal metabolic phenotype of Rhizobium etli, a bacterium that fixes nitrogen in symbiosis with Phaseolus vulgaris. To experimentally characterize the metabolic phenotype of this microorganism, we obtained the metabolic profile of 220 metabolites at two physiological stages: under free-living conditions, and during nitrogen fixation with P. vulgaris. By integrating these data into a constraint-based model, we built a refined computational platform with the capability to survey the metabolic activity underlying nitrogen fixation in R. etli. Topological analysis of the metabolic reconstruction led us to identify modular structures with functional activities. Consistent with modular activity in metabolism, we found that most of the metabolites experimentally detected in each module simultaneously increased their relative abundances during nitrogen fixation. In this work, we explore the relationships among topology, biological function, and optimal activity in the metabolism of R. etli through an integrative analysis based on modeling and metabolome data. Our findings suggest that the metabolic activity during nitrogen fixation is supported by interacting structural modules that correlate with three functional classifications: nucleic acids, peptides, and lipids. More fundamentally, we supply evidence that such modular organization during functional nitrogen fixation is a robust property under different environmental conditions.

Biological networks are an inherent concept in systems biology that is useful in elucidating how biological entities—as metabolites or proteins—work together in supporting specific phenotypes in microorganisms. Notably, topological analyses carried out over these networks have shown that modular organization is a ubiquitous property at different levels of biological organization, in such a way that modular organization may serve as an organizing principle governing the metabolic activity in microorganisms. With the aim of elucidating the relationship among functional modules, network topology, and optimal metabolic activity, here we present an integrative study that combines computational modeling and metabolome data for evaluation of the metabolic activity of the soil bacterium Rhizobium etli during symbiotic nitrogen fixation with Phaseolus vulgaris. As a result, we supply experimental and computational evidence supporting the concept that the optimal metabolic activity during this biological process is guided by modular structures in the metabolic network of R. etli. Even more fundamentally, we suggest that these biochemical modules interact among each other to ensure an optimal phenotype during nitrogen fixation. Finally, through the in silico analysis on the genome scale metabolic reconstruction for R.etli, we give some examples that suggest that these modular structures supporting nitrogen fixation are robust to external physiological conditions.

Cluster-based assessment of protein-protein interaction confidence

Background

Protein-protein interaction networks are key to a systems-level understanding of cellular biology. However, interaction data can contain a considerable fraction of false positives. Several methods have been proposed to assess the confidence of individual interactions. Most of them require the integration of additional data like protein expression and interaction homology information. While being certainly useful, such additional data are not always available and may introduce additional bias and ambiguity.

Results

We propose a novel, network topology based interaction confidence assessment method called CAPPIC (cluster-based assessment of protein-protein interaction confidence). It exploits the network’s inherent modular architecture for assessing the confidence of individual interactions. Our method determines algorithmic parameters intrinsically and does not require any parameter input or reference sets for confidence scoring.

Conclusions

On the basis of five yeast and two human physical interactome maps inferred using different techniques, we show that CAPPIC reliably assesses interaction confidence and its performance compares well to other approaches that are also based on network topology. The confidence score correlates with the agreement in localization and biological process annotations of interacting proteins. Moreover, it corroborates experimental evidence of physical interactions. Our method is not limited to physical interactome maps as we exemplify with a large yeast genetic interaction network. An implementation of CAPPIC is available at http://intscore.molgen.mpg.de.

Uncovering Arabidopsis membrane protein interactome enriched in transporters using mating-based split ubiquitin assays and classification models

High-throughput data are a double-edged sword; for the benefit of large amount of data, there is an associated cost of noise. To increase reliability and scalability of high-throughput protein interaction data generation, we tested the efficacy of classification to enrich potential protein-protein interactions. We applied this method to identify interactions among Arabidopsis membrane proteins enriched in transporters. We validated our method with multiple retests. Classification improved the quality of the ensuing interaction network and was effective in reducing the search space and increasing true positive rate. The final network of 541 interactions among 239 proteins (of which 179 are transporters) is the first protein interaction network enriched in membrane transporters reported for any organism. This network has similar topological attributes to other published protein interaction networks. It also extends and fills gaps in currently available biological networks in plants and allows building a number of hypotheses about processes and mechanisms involving signal-transduction and transport systems.

Network methods for describing sample relationships in genomic datasets: application to Huntington's disease

BACKGROUND

Genomic datasets generated by new technologies are increasingly prevalent in disparate areas of biological research. While many studies have sought to characterize relationships among genomic features, commensurate efforts to characterize relationships among biological samples have been less common. Consequently, the full extent of sample variation in genomic studies is often under-appreciated, complicating downstream analytical tasks such as gene co-expression network analysis.

RESULTS

Here we demonstrate the use of network methods for characterizing sample relationships in microarray data generated from human brain tissue. We describe an approach for identifying outlying samples that does not depend on the choice or use of clustering algorithms. We introduce a battery of measures for quantifying the consistency and integrity of sample relationships, which can be compared across disparate studies, technology platforms, and biological systems. Among these measures, we provide evidence that the correlation between the connectivity and the clustering coefficient (two important network concepts) is a sensitive indicator of homogeneity among biological samples. We also show that this measure, which we refer to as cor(K,C), can distinguish biologically meaningful relationships among subgroups of samples. Specifically, we find that cor(K,C) reveals the profound effect of Huntington's disease on samples from the caudate nucleus relative to other brain regions. Furthermore, we find that this effect is concentrated in specific modules of genes that are naturally co-expressed in human caudate nucleus, highlighting a new strategy for exploring the effects of disease on sets of genes.

CONCLUSIONS

These results underscore the importance of systematically exploring sample relationships in large genomic datasets before seeking to analyze genomic feature activity. We introduce a standardized platform for this purpose using freely available R software that has been designed to enable iterative and interactive exploration of sample networks.

Revisiting the variation of clustering coefficient of biological networks suggests new modular structure

Background

A central idea in biology is the hierarchical organization of cellular processes. A commonly used method to identify the hierarchical modular organization of network relies on detecting a global signature known as variation of clustering coefficient (so-called modularity scaling). Although several studies have suggested other possible origins of this signature, it is still widely used nowadays to identify hierarchical modularity, especially in the analysis of biological networks. Therefore, a further and systematical investigation of this signature for different types of biological networks is necessary.

Results

We analyzed a variety of biological networks and found that the commonly used signature of hierarchical modularity is actually the reflection of spoke-like topology, suggesting a different view of network architecture. We proved that the existence of super-hubs is the origin that the clustering coefficient of a node follows a particular scaling law with degree k in metabolic networks. To study the modularity of biological networks, we systematically investigated the relationship between repulsion of hubs and variation of clustering coefficient. We provided direct evidences for repulsion between hubs being the underlying origin of the variation of clustering coefficient, and found that for biological networks having no anti-correlation between hubs, such as gene co-expression network, the clustering coefficient doesn’t show dependence of degree.

Conclusions

Here we have shown that the variation of clustering coefficient is neither sufficient nor exclusive for a network to be hierarchical. Our results suggest the existence of spoke-like modules as opposed to “deterministic model” of hierarchical modularity, and suggest the need to reconsider the organizational principle of biological hierarchy.

Detecting breakdown points in metabolic networks

BACKGROUND

A complex network of biochemical reactions present in an organism generates various biological moieties necessary for its survival. It is seen that biological systems are robust to genetic and environmental changes at all levels of organization. Functions of various organisms are sustained against mutational changes by using alternative pathways. It is also seen that if any one of the paths for production of the same metabolite is hampered, an alternate path tries to overcome this defect and helps in combating the damage.

METHODOLOGY

Certain physical, chemical or genetic change in any of the precursor substrate of a biochemical reaction may damage the production of the ultimate product. We employ a quantitative approach for simulating this phenomena of causing a physical change in the biochemical reactions by performing external perturbations to 12 metabolic pathways under carbohydrate metabolism in Saccharomyces cerevisae as well as 14 metabolic pathways under carbohydrate metabolism in Homo sapiens. Here, we investigate the relationship between structure and degree of compatibility of metabolites against external perturbations, i.e., robustness. Robustness can also be further used to identify the extent to which a metabolic pathway can resist a mutation event. Biological networks with a certain connectivity distribution may be very resilient to a particular attack but not to another. The goal of this work is to determine the exact boundary of network breakdown due to both random and targeted attack, thereby analyzing its robustness. We also find that compared to various non-standard models, metabolic networks are exceptionally robust. Here, we report the use of a 'Resilience-based' score for enumerating the concept of 'network-breakdown'. We also use this approach for analyzing metabolite essentiality providing insight into cellular robustness that can be further used for future drug development.

RESULTS

We have investigated the behavior of metabolic pathways under carbohydrate metabolism in S. cerevisae and H. sapiens against random and targeted attack. Both random as well as targeted resilience were calculated by formulating a measure, that we termed as 'Resilience score'. Datasets of metabolites were collected for 12 metabolic pathways belonging to carbohydrate metabolism in S. cerevisae and 14 metabolic pathways belonging to carbohydrate metabolism in H. sapiens from Kyoto Encyclopedia for Genes and Genomes (KEGG).

Conserved and variable architecture of human white matter connectivity

Whole-brain network analysis of diffusion imaging tractography data is an important new tool for quantification of differential connectivity patterns across individuals and between groups. Here we investigate both the conservation of network architectural properties across methodological variation and the reproducibility of individual architecture across multiple scanning sessions. Diffusion spectrum imaging (DSI) and diffusion tensor imaging (DTI) data were both acquired in triplicate from a cohort of healthy young adults. Deterministic tractography was performed on each dataset and inter-regional connectivity matrices were then derived by applying each of three widely used whole-brain parcellation schemes over a range of spatial resolutions. Across acquisitions and preprocessing streams, anatomical brain networks were found to be sparsely connected, hierarchical, and assortative. They also displayed signatures of topo-physical interdependence such as Rentian scaling. Basic connectivity properties and several graph metrics consistently displayed high reproducibility and low variability in both DSI and DTI networks. The relative increased sensitivity of DSI to complex fiber configurations was evident in increased tract counts and network density compared with DTI. In combination, this pattern of results shows that network analysis of human white matter connectivity provides sensitive and temporally stable topological and physical estimates of individual cortical structure across multiple spatial scales.

The use of network analyses for elucidating mechanisms in cardiovascular disease

Systems biology offers the potential to provide new insights into our understanding of the pathogenesis of complex diseases such as atherosclerosis. It seeks to comprehend the system properties of the non-linear interactions of the multiple biomolecular components that characterize a living organism. An important component of this research approach is identifying the biological networks that connect the differing elements of a system and in the process describe the characteristics that define a shift in equilibrium from a healthy to a diseased state. The utility of this method becomes clear when applied to multifactorial diseases with complex etiologies such as inflammatory-related diseases, herein exemplified by cardiovascular disease. In this study, the application of network theory to systems biology is described in detail and an example is provided using data from a clinical biobank database of carotid endarterectomies from the Karolinska University Hospital (Biobank of Karolinska Endarterectomies, BiKE). Data from 47 microarrays were examined using a combination of Bioconductor modules and the Cytoscape resource with several associated plugins to analyze the transcriptomics data and create a combined gene association and correlation network of atherosclerosis. The methodology and workflow are described in detail, with a total of 43 genes found to be differentially expressed on a gender-specific basis, of which 15 were not directly linked to the sex chromosomes. In particular, the APOC1 gene was 2.1-fold down-regulated in plaques in women relative to men and was selected for further analysis based upon a purported role in cardiovascular disease. The resulting network was identified as a scale-free network that contained specific sub-networks related to immune function and lipid biosynthesis. These sub-networks link atherosclerotic-related genes to other genes that may not have previously known roles in disease etiology and only evidence small alterations, which are challenging to find by statistical and comparison-based methods. A number of Gene Ontology (GO), BioCarta and KEGG pathways involved in the atherosclerotic process were identified in the constructed sub-network, with 19 GO pathways related to APOC1 of which 'phospholipid efflux' evidenced the strongest association. The utility and functionality of network analysis and the different Cytoscape plugins employed are discussed. Lastly, the applications of these methods to cardiovascular disease are discussed with focus on the current limitations and future visions of this emerging field.