Identification of functional modules in a PPI network by clique percolation clustering

Exploring disease comorbidity in a module-module interaction network

Understanding disease comorbidity contributes to improved quality of life in patients who are suffering from multiple diseases. Therefore, to better explore comorbid diseases, the clarification of associations between diseases based on biological functions is essential. In our study, we propose a method for identifying disease comorbidity in a module-based network, named the module-module interaction (MMI) network, which represents how biological functions influence each other. To construct the MMI network, we detected gene modules - sets of genes that have a higher probability of taking part in specific functions - and established a link between these modules. Subsequently, we constructed disease-related networks in the MMI network to understand inherent disease mechanisms and calculated comorbidity scores of disease pairs using Gene Ontology (GO) terms. Our results show that we can obtain further information on disease mechanisms by considering interactions between functional modules instead of between genes. In addition, we verified that predicted comorbid relationships of disease pairs based on the MMI network are more significant than those based on the protein-protein interaction (PPI) network. This study can be useful to elucidate the mechanisms underlying comorbidities for further study, which will provide a broader insight into the pathogenesis of diseases.

A three-phase method for identifying functionally related protein groups in weighted PPI networks

Identifying significant protein groups is of great importance for further understanding protein functions. This paper introduces a novel three-phase heuristic method for identifying such groups in weighted PPI networks. In the first phase a variable neighborhood search (VNS) algorithm is applied on a weighted PPI network, in order to support protein complexes by adding a minimum number of new PPIs. In the second phase proteins from different complexes are merged into larger protein groups. In the third phase these groups are expanded by a number of 2-level neighbor proteins, favoring proteins that have higher average gene co-expression with the base group proteins. Experimental results show that: (i) the proposed VNS algorithm outperforms the existing approach described in literature and (ii) the above-mentioned three-phase method identifies protein groups with very high statistical significance.

A network analysis revealed the essential and common downstream proteins related to inguinal hernia

Although more than 1 in 4 men develop symptomatic inguinal hernia during their lifetime, the molecular mechanism behind inguinal hernia remains unknown. Here, we explored the protein-protein interaction network built on known inguinal hernia-causative genes to identify essential and common downstream proteins for inguinal hernia formation. We discovered that PIK3R1, PTPN11, TGFBR1, CDC42, SOS1, and KRAS were the most essential inguinal hernia-causative proteins and UBC, GRB2, CTNNB1, HSP90AA1, CBL, PLCG1, and CRK were listed as the most commonly-involved downstream proteins. In addition, the transmembrane receptor protein tyrosine kinase signaling pathway was the most frequently found inguinal hernia-related pathway. Our in silico approach was able to uncover a novel molecular mechanism underlying inguinal hernia formation by identifying inguinal hernia-related essential proteins and potential common downstream proteins of inguinal hernia-causative proteins.

Discovery and characterization of functional modules associated with body weight in broilers

Aim of the present study was to investigate whether body weight (BW) in broilers is associated with functional modular genes. To this end, first a GWAS for BW was conducted using 6,598 broilers and the high density SNP array. The next step was to search for positional candidate genes and QTLs within strong LD genomic regions around the significant SNPs. Using all positional candidate genes, a network was then constructed and community structure analysis was performed. Finally, functional enrichment analysis was applied to infer the functional relevance of modular genes. A total number of 645 positional candidate genes were identified in strong LD genomic regions around 11 genome-wide significant markers. 428 of the positional candidate genes were located within growth related QTLs. Community structure analysis detected 5 modules while functional enrichment analysis showed that 52 modular genes participated in developmental processes such as skeletal system development. An additional number of 14 modular genes (GABRG1, NGF, APOBEC2, STAT5B, STAT3, SMAD4, MED1, CACNB1, SLAIN2, LEMD2, ZC3H18, TMEM132D, FRYL and SGCB) were also identified as related to body weight. Taken together, current results suggested a total number of 66 genes as most plausible functional candidates for the trait examined.

CASS: A distributed network clustering algorithm based on structure similarity for large-scale network

As the size of networks increases, it is becoming important to analyze large-scale network data. A network clustering algorithm is useful for analysis of network data. Conventional network clustering algorithms in a single machine environment rather than a parallel machine environment are actively being researched. However, these algorithms cannot analyze large-scale network data because of memory size issues. As a solution, we propose a network clustering algorithm for large-scale network data analysis using Apache Spark by changing the paradigm of the conventional clustering algorithm to improve its efficiency in the Apache Spark environment. We also apply optimization approaches such as Bloom filter and shuffle selection to reduce memory usage and execution time. By evaluating our proposed algorithm based on an average normalized cut, we confirmed that the algorithm can analyze diverse large-scale network datasets such as biological, co-authorship, internet topology and social networks. Experimental results show that the proposed algorithm can develop more accurate clusters than comparative algorithms with less memory usage. Furthermore, we confirm the proposed optimization approaches and the scalability of the proposed algorithm. In addition, we validate that clusters found from the proposed algorithm can represent biologically meaningful functions.

A Comprehensive In Silico Method to Study the QSTR of the Aconitine Alkaloids for Designing Novel Drugs

A combined in silico method was developed to predict potential protein targets that are involved in cardiotoxicity induced by aconitine alkaloids and to study the quantitative structure⁻toxicity relationship (QSTR) of these compounds. For the prediction research, a Protein-Protein Interaction (PPI) network was built from the extraction of useful information about protein interactions connected with aconitine cardiotoxicity, based on nearly a decade of literature and the STRING database. The software Cytoscape and the PharmMapper server were utilized to screen for essential proteins in the constructed network. The Calcium-Calmodulin-Dependent Protein Kinase II alpha (CAMK2A) and gamma (CAMK2G) were identified as potential targets. To obtain a deeper insight on the relationship between the toxicity and the structure of aconitine alkaloids, the present study utilized QSAR models built in Sybyl software that possess internal robustness and external high predictions. The molecular dynamics simulation carried out here have demonstrated that aconitine alkaloids possess binding stability for the receptor CAMK2G. In conclusion, this comprehensive method will serve as a tool for following a structural modification of the aconitine alkaloids and lead to a better insight into the cardiotoxicity induced by the compounds that have similar structures to its derivatives.

Detection of core-periphery structure in networks based on 3-tuple motifs

Detecting mesoscale structure, such as community structure, is of vital importance for analyzing complex networks. Recently, a new mesoscale structure, core-periphery (CP) structure, has been identified in many real-world systems. In this paper, we propose an effective algorithm for detecting CP structure based on a 3-tuple motif. In this algorithm, we first define a 3-tuple motif in terms of the patterns of edges as well as the property of nodes, and then a motif adjacency matrix is constructed based on the 3-tuple motif. Finally, the problem is converted to find a cluster that minimizes the smallest motif conductance. Our algorithm works well in different CP structures: including single or multiple CP structure, and local or global CP structures. Results on the synthetic and the empirical networks validate the high performance of our method.

A unified method of detecting core-periphery structure and community structure in networks

The core-periphery structure and the community structure are two typical meso-scale structures in complex networks. Although community detection has been extensively investigated from different perspectives, the definition and the detection of the core-periphery structure have not received much attention. Furthermore, the detection problems of the core-periphery and community structure were separately investigated. In this paper, we develop a unified framework to simultaneously detect the core-periphery structure and community structure in complex networks. Moreover, there are several extra advantages of our algorithm: our method can detect not only single but also multiple pairs of core-periphery structures; the overlapping nodes belonging to different communities can be identified; different scales of core-periphery structures can be detected by adjusting the size of the core. The good performance of the method has been validated on synthetic and real complex networks. So, we provide a basic framework to detect the two typical meso-scale structures: the core-periphery structure and the community structure.

Functional diversity of topological modules in human protein-protein interaction networks

A large-scale molecular interaction network of protein-protein interactions (PPIs) enables the automatic detection of molecular functional modules through a computational approach. However, the functional modules that are typically detected by topological community detection algorithms may be diverse in functional homogeneity and are empirically considered to be default functional modules. Thus, a significant challenge that has been described but not elucidated is investigating the relationship between topological modules and functional modules. We systematically investigated this issue by initially using seven widely used community detection algorithms to partition the PPI network into communities. Four homogeneity measures were subsequently implemented to evaluate the functional homogeneity of protein community. We determined that a significant portion of topological modules with heterogeneous functionality exists and should be further investigated; moreover, these findings indicated that topologically based functional module detection approaches must be reconsidered. Furthermore, we found that the functional homogeneity of topological modules is positively correlated with their edge densities, degree of association with diseases and general Gene Ontology (GO) terms. Thus, topologically based module detection approaches should be used with caution in the identification of functional modules with high homogeneity

The essential and downstream common proteins of amyotrophic lateral sclerosis: A protein-protein interaction network analysis

Amyotrophic Lateral Sclerosis (ALS) is a devastative neurodegenerative disease characterized by selective loss of motoneurons. While several breakthroughs have been made in identifying ALS genetic defects, the detailed molecular mechanisms are still unclear. These genetic defects involve in numerous biological processes, which converge to a common destiny: motoneuron degeneration. In addition, the common comorbid Frontotemporal Dementia (FTD) further complicates the investigation of ALS etiology. In this study, we aimed to explore the protein-protein interaction network built on known ALS-causative genes to identify essential proteins and common downstream proteins between classical ALS and ALS+FTD (classical ALS + ALS/FTD) groups. The results suggest that classical ALS and ALS+FTD share similar essential protein set (VCP, FUS, TDP-43 and hnRNPA1) but have distinctive functional enrichment profiles. Thus, disruptions to these essential proteins might cause motoneuron susceptible to cellular stresses and eventually vulnerable to proteinopathies. Moreover, we identified a common downstream protein, ubiquitin-C, extensively interconnected with ALS-causative proteins (22 out of 24) which was not linked to ALS previously. Our in silico approach provides the computational background for identifying ALS therapeutic targets, and points out the potential downstream common ground of ALS-causative mutations.

Pathogenicity Genes in Ustilaginoidea virens Revealed by a Predicted Protein-Protein Interaction Network

Rice false smut, caused by Ustilaginoidea virens, produces significant losses in rice yield and grain quality and has recently emerged as one of the most important rice diseases worldwide. Despite its importance in rice production, relatively few studies have been conducted to illustrate the complex interactome and the pathogenicity gene interactions. Here a protein-protein interaction network of U. virens was built through two well-recognized approaches, interolog- and domain-domain interaction-based methods. A total of 20 217 interactions associated with 3305 proteins were predicted after strict filtering. The reliability of the network was assessed computationally and experimentally. The topology of the interactome network revealed highly connected proteins. A pathogenicity-related subnetwork involving up-regulated genes during early U. virens infection was also constructed, and many novel pathogenicity proteins were predicted in the subnetwork. In addition, we built an interspecies PPI network between U. virens and Oryza sativa, providing new insights for molecular interactions of this host-pathogen pathosystem. A web-based publicly available interactive database based on these interaction networks has also been released. In summary, a proteome-scale map of the PPI network was described for U. virens, which will provide new perspectives for finely dissecting interactions of genes related to its pathogenicity.

Network Modules of the Cross-Species Genotype-Phenotype Map Reflect the Clinical Severity of Human Diseases

Recent advances in genome sequencing techniques have improved our understanding of the genotype-phenotype relationship between genetic variants and human diseases. However, genetic variations uncovered from patient populations do not provide enough information to understand the mechanisms underlying the progression and clinical severity of human diseases. Moreover, building a high-resolution genotype-phenotype map is difficult due to the diverse genetic backgrounds of the human population. We built a cross-species genotype-phenotype map to explain the clinical severity of human genetic diseases. We developed a data-integrative framework to investigate network modules composed of human diseases mapped with gene essentiality measured from a model organism. Essential and nonessential genes connect diseases of different types which form clusters in the human disease network. In a large patient population study, we found that disease classes enriched with essential genes tended to show a higher mortality rate than disease classes enriched with nonessential genes. Moreover, high disease mortality rates are explained by the multiple comorbid relationships and the high pleiotropy of disease genes found in the essential gene-enriched diseases. Our results reveal that the genotype-phenotype map of a model organism can facilitate the identification of human disease-gene associations and predict human disease progression.

A novel protein complex identification algorithm based on Connected Affinity Clique Extension (CACE)

A novel algorithm based on Connected Affinity Clique Extension (CACE) for mining overlapping functional modules in protein interaction network is proposed in this paper. In this approach, the value of protein connected affinity which is inferred from protein complexes is interpreted as the reliability and possibility of interaction. The protein interaction network is constructed as a weighted graph, and the weight is dependent on the connected affinity coefficient. The experimental results of our CACE in two test data sets show that the CACE can detect the functional modules much more effectively and accurately when compared with other state-of-art algorithms CPM and IPC-MCE.

Leveraging gene co-expression networks to pinpoint the regulation of complex traits and disease, with a focus on cardiovascular traits

Over the past decade, the number of genome-scale transcriptional datasets in publicly available databases has climbed to nearly one million, providing an unprecedented opportunity for extensive analyses of gene co-expression networks. In systems-genetic studies of complex diseases researchers increasingly focus on groups of highly interconnected genes within complex transcriptional networks (referred to as clusters, modules or subnetworks) to uncover specific molecular processes that can inform functional disease mechanisms and pathological pathways. Here, we outline the basic paradigms underlying gene co-expression network analysis and critically review the most commonly used computational methods. Finally, we discuss specific applications of network-based approaches to the study of cardiovascular traits, which highlight the power of integrated analyses of networks, genetic and gene-regulation data to elucidate the complex mechanisms underlying cardiovascular disease.

Navigating traditional chinese medicine network pharmacology and computational tools

The concept of "network target" has ushered in a new era in the field of traditional Chinese medicine (TCM). As a new research approach, network pharmacology is based on the analysis of network models and systems biology. Taking advantage of advancements in systems biology, a high degree of integration data analysis strategy and interpretable visualization provides deeper insights into the underlying mechanisms of TCM theories, including the principles of herb combination, biological foundations of herb or herbal formulae action, and molecular basis of TCM syndromes. In this study, we review several recent developments in TCM network pharmacology research and discuss their potential for bridging the gap between traditional and modern medicine. We briefly summarize the two main functional applications of TCM network models: understanding/uncovering and predicting/discovering. In particular, we focus on how TCM network pharmacology research is conducted and highlight different computational tools, such as network-based and machine learning algorithms, and sources that have been proposed and applied to the different steps involved in the research process. To make network pharmacology research commonplace, some basic network definitions and analysis methods are presented.

Complex disease interventions from a network model for type 2 diabetes

There is accumulating evidence that the proteins encoded by the genes associated with a common disorder interact with each other, participate in similar pathways and share GO terms. It has been anticipated that the functional modules in a disease related functional linkage network are informative to reveal significant metabolic processes and disease's associations with other complex disorders. In the current study, Type 2 diabetes associated functional linkage network (T2DFN) containing 2770 proteins and 15041 linkages was constructed. The functional modules in this network were scored and evaluated in terms of shared pathways, co-localization, co-expression and associations with similar diseases. The assembly of top scoring overlapping members in the functional modules revealed that, along with the well known biological pathways, circadian rhythm, diverse actions of nuclear receptors in steroid and retinoic acid metabolisms have significant occurrence in the pathophysiology of the disease. The disease's association with other metabolic and neuromuscular disorders was established through shared proteins. Nuclear receptor NRIP1 has a pivotal role in lipid and carbohydrate metabolism, indicating the need to investigate subsequent effects of NRIP1 on Type 2 diabetes. Our study also revealed that CREB binding protein (CREBBP) and cardiotrophin-1 (CTF1) have suggestive roles in linking Type 2 diabetes and neuromuscular diseases.

Spotlight: assembly of protein complexes by integrating graph clustering methods

As is generally assumed, clusters in protein-protein interaction (PPI) networks perform specific, crucial functions in biological systems. Various network community detection methods have been developed to exploit PPI networks in order to identify protein complexes and functional modules. Due to the potential role of various regulatory modes in biological networks, a single method may just apply a single graph property and neglect communities highlighted by other network properties. This work presents a novel integration method to capture protein modules/protein complexes by multiple network features detected by different algorithms. The integration method is further implemented in a web-based platform with a highly effective interactive network analyzer. Conventionally adopted methods with different perspectives on network community detection (e.g., CPM, FastGreedy, HUNTER, MCL, LE, SpinGlass, and WalkTrap) are also executed simultaneously. Analytical results indicate that the proposed method performs better than the conventional ones. The proposed approach can capture the transcription and RNA splicing machineries from the yeast protein network. Meanwhile, proteins that are highly associated with each other, yet not described in both machineries are also identified. In sum, a protein that is closely connected to components of a known module or a complex in the network view implies the functional association among them. Importantly, our method can detect these unique network features, thus facilitating efforts to discover unknown components of functional modules/protein complexes.

AVAILABILITY

Spotlight is freely accessible at http://hub.iis.sinica.edu.tw/spotlight. Video clips for a quick view of usage are available in the website online help page.

An efficient heuristic method for active feature acquisition and its application to protein-protein interaction prediction

BACKGROUND

Machine learning approaches for classification learn the pattern of the feature space of different classes, or learn a boundary that separates the feature space into different classes. The features of the data instances are usually available, and it is only the class-labels of the instances that are unavailable. For example, to classify text documents into different topic categories, the words in the documents are features and they are readily available, whereas the topic is what is predicted. However, in some domains obtaining features may be resource-intensive because of which not all features may be available. An example is that of protein-protein interaction prediction, where not only are the labels ('interacting' or 'non-interacting') unavailable, but so are some of the features. It may be possible to obtain at least some of the missing features by carrying out a few experiments as permitted by the available resources. If only a few experiments can be carried out to acquire missing features, which proteins should be studied and which features of those proteins should be determined? From the perspective of machine learning for PPI prediction, it would be desirable that those features be acquired which when used in training the classifier, the accuracy of the classifier is improved the most. That is, the utility of the feature-acquisition is measured in terms of how much acquired features contribute to improving the accuracy of the classifier. Active feature acquisition (AFA) is a strategy to preselect such instance-feature combinations (i.e. protein and experiment combinations) for maximum utility. The goal of AFA is the creation of optimal training set that would result in the best classifier, and not in determining the best classification model itself.

RESULTS

We present a heuristic method for active feature acquisition to calculate the utility of acquiring a missing feature. This heuristic takes into account the change in belief of the classification model induced by the acquisition of the feature under consideration. As compared to random selection of proteins on which the experiments are performed and the type of experiment that is performed, the heuristic method reduces the number of experiments to as few as 40%. Most notable characteristic of this method is that it does not require re-training of the classification model on every possible combination of instance, feature and feature-value tuples. For this reason, our method is far less computationally expensive as compared with previous AFA strategies.

CONCLUSIONS

The results show that our heuristic method for AFA creates an optimal training set with far less features acquired as compared to random acquisition. This shows the value of active feature acquisition to aid in protein-protein interaction prediction where feature acquisition is costly. Compared to previous methods, the proposed method reduces computational cost while also achieving a better F-score. The proposed method is valuable as it presents a direction to AFA with a far lesser computational expense by removing the need for the first time, of training a classifier for every combination of instance, feature and feature-value tuples which would be impractical for several domains.

ModuleSearch: finding functional modules in a protein-protein interaction network

Many biological processes are performed by a group of proteins rather than by individual proteins. Proteins involved in the same biological process often form a densely connected sub-graph in a protein-protein interaction network. Therefore, finding a dense sub-graph provides useful information to predict the function or protein complex of uncharacterised proteins in the sub-graph. We developed a heuristic algorithm that finds functional modules in a protein-protein interaction network and visualises the modules. The algorithm has been implemented in a platform-independent, standalone program called ModuleSearch. In an interaction network of yeast proteins, ModuleSearch found 366 overlapping modules. Of the modules, 71% have a function shared by more than half the proteins in the module and 58% have a function shared by all proteins in the module. Comparison of ModuleSearch with other programs shows that ModuleSearch finds more sub-graphs than most other programs, yet a higher proportion of the sub-graphs correspond to known functional modules. ModuleSearch and sample data are freely available to academics at http://bclab.inha.ac.kr/ModuleSearch.

A novel integrative network approach to understand the interplay between cardiovascular disease and other complex disorders

There is accumulating evidence that the proteins encoded by the genes associated with a common disorder interact with each other, participate in similar pathways and share GO terms. It has been anticipated that the functional modules in a disease related functional linkage network can be integrated with bibliomics to reveal association with other complex disorders. In this study, the cardiovascular disease functional linkage network (CFN) containing 1536 nodes and 3345 interactions was constructed using proteins encoded by 234 genes associated with the disease. Integration of CFN with bibliomics showed that 227 out of 566 functional modules are significantly associated with one or more diseases. Analysis of functional modules revealed the possible regulatory roles of SP1 and CXCL12 in the pathogenesis of cardiovascular disease (CVD) and modulation of their activities may be considered as potential therapeutic tools. The integration of CFN with bibliomics also indicated significant relations of CVD with other complex disorders. In a stratified map the members of 227 functional modules and 58 diseases in 15 disease classes were combined. In this map, leprosy, listeria monocytogenes, myasthenia, hemorrhagic diathesis and Protein S deficiency, which were not previously reported to be associated with CVD, showed significant associations. Several cancers arising from epithelial cells were also found to be linked to other diseases through hub proteins, VEGFA and PTGS2.

Recent advances in clustering methods for protein interaction networks

The increasing availability of large-scale protein-protein interaction data has made it possible to understand the basic components and organization of cell machinery from the network level. The arising challenge is how to analyze such complex interacting data to reveal the principles of cellular organization, processes and functions. Many studies have shown that clustering protein interaction network is an effective approach for identifying protein complexes or functional modules, which has become a major research topic in systems biology. In this review, recent advances in clustering methods for protein interaction networks will be presented in detail. The predictions of protein functions and interactions based on modules will be covered. Finally, the performance of different clustering methods will be compared and the directions for future research will be discussed.

Identifying protein complexes from interaction networks based on clique percolation and distance restriction

BACKGROUND

Identification of protein complexes in large interaction networks is crucial to understand principles of cellular organization and predict protein functions, which is one of the most important issues in the post-genomic era. Each protein might be subordinate multiple protein complexes in the real protein-protein interaction networks. Identifying overlapping protein complexes from protein-protein interaction networks is a considerable research topic.

RESULT

As an effective algorithm in identifying overlapping module structures, clique percolation method (CPM) has a wide range of application in social networks and biological networks. However, the recognition accuracy of algorithm CPM is lowly. Furthermore, algorithm CPM is unfit to identifying protein complexes with meso-scale when it applied in protein-protein interaction networks. In this paper, we propose a new topological model by extending the definition of k-clique community of algorithm CPM and introduced distance restriction, and develop a novel algorithm called CP-DR based on the new topological model for identifying protein complexes. In this new algorithm, the protein complex size is restricted by distance constraint to conquer the shortcomings of algorithm CPM. The algorithm CP-DR is applied to the protein interaction network of Sacchromyces cerevisiae and identifies many well known complexes.

CONCLUSION

The proposed algorithm CP-DR based on clique percolation and distance restriction makes it possible to identify dense subgraphs in protein interaction networks, a large number of which correspond to known protein complexes. Compared to algorithm CPM, algorithm CP-DR has more outstanding performance.

A hub-attachment based method to detect functional modules from confidence-scored protein interactions and expression profiles

Background

Many research results show that the biological systems are composed of functional modules. Members in the same module usually have common functions. This is useful information to understand how biological systems work. Therefore, detecting functional modules is an important research topic in the post-genome era. One of functional module detecting methods is to find dense regions in Protein-Protein Interaction (PPI) networks. Most of current methods neglect confidence-scores of interactions, and pay little attention on using gene expression data to improve their results.

Results

In this paper, we propose a novel hub-attachment based method to detect functional modules from confidence-scored protein interactions and expression profiles, and we name it HUNTER. Our method not only can extract functional modules from a weighted PPI network, but also use gene expression data as optional input to increase the quality of outcomes. Using HUNTER on yeast data, we found it can discover more novel components related with RNA polymerase complex than those existed methods from yeast interactome. And these new components show the close relationship with polymerase after functional analysis on Gene Ontology.

Conclusion

A C++ implementation of our prediction method, dataset and supplementary material are available at http://hub.iis.sinica.edu.tw/Hunter/. Our proposed HUNTER method has been applied on yeast data, and the empirical results show that our method can accurately identify functional modules. Such useful application derived from our algorithm can reconstruct the biological machinery, identify undiscovered components and decipher common sub-modules inside these complexes like RNA polymerases I, II, III.

Breaking the hierarchy--a new cluster selection mechanism for hierarchical clustering methods

Background

Hierarchical clustering methods like Ward's method have been used since decades to understand biological and chemical data sets. In order to get a partition of the data set, it is necessary to choose an optimal level of the hierarchy by a so-called level selection algorithm. In 2005, a new kind of hierarchical clustering method was introduced by Palla et al. that differs in two ways from Ward's method: it can be used on data on which no full similarity matrix is defined and it can produce overlapping clusters, i.e., allow for multiple membership of items in clusters. These features are optimal for biological and chemical data sets but until now no level selection algorithm has been published for this method.

Results

In this article we provide a general selection scheme, the level independent clustering selection method, called LInCS. With it, clusters can be selected from any level in quadratic time with respect to the number of clusters. Since hierarchically clustered data is not necessarily associated with a similarity measure, the selection is based on a graph theoretic notion of cohesive clusters. We present results of our method on two data sets, a set of drug like molecules and set of protein-protein interaction (PPI) data. In both cases the method provides a clustering with very good sensitivity and specificity values according to a given reference clustering. Moreover, we can show for the PPI data set that our graph theoretic cohesiveness measure indeed chooses biologically homogeneous clusters and disregards inhomogeneous ones in most cases. We finally discuss how the method can be generalized to other hierarchical clustering methods to allow for a level independent cluster selection.

Conclusion

Using our new cluster selection method together with the method by Palla et al. provides a new interesting clustering mechanism that allows to compute overlapping clusters, which is especially valuable for biological and chemical data sets.

Triangle network motifs predict complexes by complementing high-error interactomes with structural information

Background

A lot of high-throughput studies produce protein-protein interaction networks (PPINs) with many errors and missing information. Even for genome-wide approaches, there is often a low overlap between PPINs produced by different studies. Second-level neighbors separated by two protein-protein interactions (PPIs) were previously used for predicting protein function and finding complexes in high-error PPINs. We retrieve second level neighbors in PPINs, and complement these with structural domain-domain interactions (SDDIs) representing binding evidence on proteins, forming PPI-SDDI-PPI triangles.

Results

We find low overlap between PPINs, SDDIs and known complexes, all well below 10%. We evaluate the overlap of PPI-SDDI-PPI triangles with known complexes from Munich Information center for Protein Sequences (MIPS). PPI-SDDI-PPI triangles have ~20 times higher overlap with MIPS complexes than using second-level neighbors in PPINs without SDDIs. The biological interpretation for triangles is that a SDDI causes two proteins to be observed with common interaction partners in high-throughput experiments. The relatively few SDDIs overlapping with PPINs are part of highly connected SDDI components, and are more likely to be detected in experimental studies. We demonstrate the utility of PPI-SDDI-PPI triangles by reconstructing myosin-actin processes in the nucleus, cytoplasm, and cytoskeleton, which were not obvious in the original PPIN. Using other complementary datatypes in place of SDDIs to form triangles, such as PubMed co-occurrences or threading information, results in a similar ability to find protein complexes.

Conclusion

Given high-error PPINs with missing information, triangles of mixed datatypes are a promising direction for finding protein complexes. Integrating PPINs with SDDIs improves finding complexes. Structural SDDIs partially explain the high functional similarity of second-level neighbors in PPINs. We estimate that relatively little structural information would be sufficient for finding complexes involving most of the proteins and interactions in a typical PPIN.

Clustering algorithms for detecting functional modules in protein interaction networks

Protein-Protein Interaction (PPI) networks are believed to be important sources of information related to biological processes and complex metabolic functions of the cell. When studying the workings of a biological cell, it is useful to be able to detect known and predict still undiscovered protein complexes within the cell's PPI networks. Such predictions may be used as an inexpensive tool to direct biological experiments. The increasing amount of available PPI data necessitate a fast, accurate approach to biological complex identification. Because of its importance in the studies of protein interaction network, there are different models and algorithms in identifying functional modules in PPI networks. In this paper, we review some representative algorithms, focusing on the algorithms underlying the approaches and how the algorithms relate to each other. In particular, a comparison is given based on the property of the algorithms. Since the PPI network is noisy and still incomplete, some methods which consider other additional properties for preprocessing and purifying of PPI data are presented. We also give a discussion about the functional annotation and validation of protein complexes. Finally, new progress and future research directions are discussed from the computational viewpoint.

Identifying responsive functional modules from protein-protein interaction network

Proteins interact with each other within a cell, and those interactions give rise to the biological function and dynamical behavior of cellular systems. Generally, the protein interactions are temporal, spatial, or condition dependent in a specific cell, where only a small part of interactions usually take place under certain conditions. Recently, although a large amount of protein interaction data have been collected by high-throughput technologies, the interactions are recorded or summarized under various or different conditions and therefore cannot be directly used to identify signaling pathways or active networks, which are believed to work in specific cells under specific conditions. However, protein interactions activated under specific conditions may give hints to the biological process underlying corresponding phenotypes. In particular, responsive functional modules consist of protein interactions activated under specific conditions can provide insight into the mechanism underlying biological systems, e.g. protein interaction subnetworks found for certain diseases rather than normal conditions may help to discover potential biomarkers. From computational viewpoint, identifying responsive functional modules can be formulated as an optimization problem. Therefore, efficient computational methods for extracting responsive functional modules are strongly demanded due to the NP-hard nature of such a combinatorial problem. In this review, we first report recent advances in development of computational methods for extracting responsive functional modules or active pathways from protein interaction network and microarray data. Then from computational aspect, we discuss remaining obstacles and perspectives for this attractive and challenging topic in the area of systems biology.

Dynamic network topology changes in functional modules predict responses to oxidative stress in yeast

In response to environmental challenges, biological systems respond with dynamic adaptive changes in order to maintain the functionality of the system. Such adaptations may lead to cumulative stress over time, possibly leading to global failure of the system. When studying such systems responses, it is therefore important to understand them in system-wide and dynamic context. Here we hypothesize that dynamic changes in the topology of functional modules of integrated biological networks reflect their activity under specific environmental challenges. We introduce topological enrichment analysis of functional subnetworks (TEAFS), a method for the analysis of integrated molecular profile and interactome data, which we validated by comprehensive metabolomic analysis of dynamic yeast response under oxidative stress. TEAFS identified activation of multiple stress response related mechanisms, such as lipid metabolism and phospholipid biosynthesis. We identified, among others, a fatty acid elongase IFA38 as a hub protein which was absent at all time points under oxidative stress conditions. The deletion mutant of the IFA38 encoding gene is known for the accumulation of ceramides. By applying a comprehensive metabolomic analysis, we confirmed the increased concentrations over time of ceramides and palmitic acid, a precursor of de novo ceramide biosynthesis. Our results imply that the connectivity of the system is being dynamically modulated in response to oxidative stress, progressively leading to the accumulation of (lipo)toxic lipids such as ceramides. Studies of local network topology dynamics can be used to investigate as well as predict the activity of biological processes and the system's responses to environmental challenges and interventions.

Gene module level analysis: identification to networks and dynamics

Nature exhibits modular design in biological systems. Gene module level analysis is based on this module concept, aiming to understand biological network design and systems behavior in disease and development by emphasizing on modules of genes rather than individual genes. Module level analysis has been extensively applied in genome wide level analysis, exploring the organization of biological systems from identifying modules to reconstructing module networks and analyzing module dynamics. Such module level perspective provides a high level representation of the regulatory scenario and design of biological systems, promising to revolutionize our view of systems biology, genetic engineering as well as disease mechanisms and molecular medicine.

Discovering functions and revealing mechanisms at molecular level from biological networks

With the increasingly accumulated data from high-throughput technologies, study on biomolecular networks has become one of key focuses in systems biology and bioinformatics. In particular, various types of molecular networks (e.g., protein-protein interaction (PPI) network; gene regulatory network (GRN); metabolic network (MN); gene coexpression network (GCEN)) have been extensively investigated, and those studies demonstrate great potentials to discover basic functions and to reveal essential mechanisms for various biological phenomena, by understanding biological systems not at individual component level but at a system-wide level. Recent studies on networks have created very prolific researches on many aspects of living organisms. In this paper, we aim to review the recent developments on topics related to molecular networks in a comprehensive manner, with the special emphasis on the computational aspect. The contents of the survey cover global topological properties and local structural characteristics, network motifs, network comparison and query, detection of functional modules and network motifs, function prediction from network analysis, inferring molecular networks from biological data as well as representative databases and software tools.