Protein Interaction Information Research Articles

Identification of Genes Involved in Breast Cancer Metastasis by Integrating Protein–Protein Interaction Information with Expression Data

The selection of relevant genes for breast cancer metastasis is critical for the treatment and prognosis of cancer patients. Although much effort has been devoted to the gene selection procedures by use of different statistical analysis methods or computational techniques, the interpretation of the variables in the resulting survival models has been limited so far. This article proposes a new Random Forest (RF)-based algorithm to identify important variables highly related with breast cancer metastasis, which is based on the important scores of two variable selection algorithms, including the mean decrease Gini (MDG) criteria of Random Forest and the GeneRank algorithm with protein-protein interaction (PPI) information. The new gene selection algorithm can be called PPIRF. The improved prediction accuracy fully illustrated the reliability and high interpretability of gene list selected by the PPIRF approach.

Relevance search for predicting lncRNA–protein interactions based on heterogeneous network

lncRNA plays important roles in many biological and pathological processes. lncRNA–protein interaction is the most common way of lncRNA performing their functions. Thus, predicting lncRNA–protein interaction is very significant to understand the nature of lncRNA. Earlier methods to predict RNA–protein interaction always adopt classification algorithms using features extracted from RNA and protein themselves. But their performance is not good enough in lncRNA–protein interaction prediction because of lncRNA’s low conservation. In this paper, we try to use information implicit in the topologies of biological network associated with lncRNA to solve this problem. Firstly we construct a heterogeneous lncRNA–protein network which incorporate protein interaction information. We use an algorithm called HeteSim which can evaluate relevance between heterogeneous objects to perform relevance search in the lncRNA and protein heterogeneous network. The relevance search results are used to help the identification of lncRNA–protein interactions.

Predicting pathogenic single nucleotide variants through a comprehensive analysis on multiple level features

Benefiting from the high-throughput sequencing technologies, many single nucleotide variants (SNVs) among individuals have been detected. SNVs in gene code regions were known to possibly disrupt protein functions. For this, many efforts were devoted to sort deleterious SNVs from benign ones. In general, features in the past studies can be categorized into codon level, peptide level and protein level. While those at peptide level were in widespread use, few works have carried out a comprehensive analysis by combining three levels information.In the present work, we incorporated both codon and protein level information with peptide level information to predict disease-related SNVs. Taking the advantage of combinatory multiple level features, our method exhibited competitive performance against seven well-known classifiers. Additionally, by incorporating selective pressure score and protein–protein interaction (PPI) information, we found that the functional important proteins were protected through a pressure-resistant mechanism during the evolution. Although critical proteins were obviously related with more deleterious SNVs, these pathogenic SNVs were tend to under higher selective pressures comparing to the benign variants. These results support the ongoing researches about relation between genotype and phenotype.

Mining for genes related to choroidal neovascularization based on the shortest path algorithm and protein interaction information

Background: Choroidal neovascularization (CNV) is a serious eye disease that may cause visual loss, especially for older people. Many factors have been proven to induce this disease including age, gender, obesity, and so on. However, until now, we have had limited knowledge on CNV's pathogenic mechanism. Discovering the genes that underlie this disease and performing extensive studies on them can help us to understand how CNV occurs and design effective treatments.Methods: In this study, we designed a computational method to identify novel CNV-related genes in a large protein network constructed using the protein–protein interaction information in STRING. The candidate genes were first extracted from the shortest paths connecting any two known CNV-related genes and then filtered by a permutation test and using knowledge of their linkages to known CNV-related genes. ResultsA list of putative CNV-related candidate genes was accessed by our method. These genes are deemed to have strong relationships with CNV. ConclusionsExtensive analyses of several of the putative genes such as ANK1, ITGA4, CD44 and others indicate that they are related to specific biological processes involved in CNV, implying they may be novel CNV-related genes.General significance: The newfound putative CNV-related genes may provide new insights into CNV and help design more effective treatments. This article is part of a Special Issue entitled “System Genetics” Guest Editor: Dr. Yudong Cai and Dr. Tao Huang.

Exploring novel mechanistic insights in Alzheimer's disease by assessing reliability of protein interactions.

Protein interaction networks are widely used in computational biology as a graphical means of representing higher-level systemic functions in a computable form. Although, many algorithms exist that seamlessly collect and measure protein interaction information in network models, they often do not provide novel mechanistic insights using quantitative criteria. Measuring information content and knowledge representation in network models about disease mechanisms becomes crucial particularly when exploring new target candidates in a well-defined functional context of a potential disease mechanism. To this end, we have developed a knowledge-based scoring approach that uses literature-derived protein interaction features to quantify protein interaction confidence. Thereby, we introduce the novel concept of knowledge cliffs, regions of the interaction network where a significant gap between high scoring and low scoring interactions is observed, representing a divide between established and emerging knowledge on disease mechanism. To show the application of this approach, we constructed and assessed reliability of a protein-protein interaction model specific to Alzheimer’s disease, which led to screening, and prioritization of four novel protein candidates. Evaluation of the identified candidates showed that two of them are already followed in clinical trials for testing potential AD drugs.

Discovery of new candidate genes related to brain development using protein interaction information.

Human brain development is a dramatic process composed of a series of complex and fine-tuned spatiotemporal gene expressions. A good comprehension of this process can assist us in developing the potential of our brain. However, we have only limited knowledge about the genes and gene functions that are involved in this biological process. Therefore, a substantial demand remains to discover new brain development-related genes and identify their biological functions. In this study, we aimed to discover new brain-development related genes by building a computational method. We referred to a series of computational methods used to discover new disease-related genes and developed a similar method. In this method, the shortest path algorithm was executed on a weighted graph that was constructed using protein-protein interactions. New candidate genes fell on at least one of the shortest paths connecting two known genes that are related to brain development. A randomization test was then adopted to filter positive discoveries. Of the final identified genes, several have been reported to be associated with brain development, indicating the effectiveness of the method, whereas several of the others may have potential roles in brain development.

Combining affinity proteomics and network context to identify new phosphatase substrates and adapters in growth pathways.

Protein phosphorylation homoeostasis is tightly controlled and pathological conditions are caused by subtle alterations of the cell phosphorylation profile. Altered levels of kinase activities have already been associated to specific diseases. Less is known about the impact of phosphatases, the enzymes that down-regulate phosphorylation by removing the phosphate groups. This is partly due to our poor understanding of the phosphatase-substrate network. Much of phosphatase substrate specificity is not based on intrinsic enzyme specificity with the catalytic pocket recognizing the sequence/structure context of the phosphorylated residue. In addition many phosphatase catalytic subunits do not form a stable complex with their substrates. This makes the inference and validation of phosphatase substrates a non-trivial task. Here, we present a novel approach that builds on the observation that much of phosphatase substrate selection is based on the network of physical interactions linking the phosphatase to the substrate. We first used affinity proteomics coupled to quantitative mass spectrometry to saturate the interactome of eight phosphatases whose down regulations was shown to affect the activation of the RAS-PI3K pathway. By integrating information from functional siRNA with protein interaction information, we develop a strategy that aims at inferring phosphatase physiological substrates. Graph analysis is used to identify protein scaffolds that may link the catalytic subunits to their substrates. By this approach we rediscover several previously described phosphatase substrate interactions and characterize two new protein scaffolds that promote the dephosphorylation of PTPN11 and ERK by DUSP18 and DUSP26, respectively.

PhenoNet: identification of key networks associated with disease phenotype.

At the core of transcriptome analyses of cancer is a challenge to detect molecular differences affiliated with disease phenotypes. This approach has led to remarkable progress in identifying molecular signatures and in stratifying patients into clinical groups. Yet, despite this progress, many of the identified signatures are not robust enough to be clinically used and not consistent enough to provide a follow-up on molecular mechanisms. To address these issues, we introduce PhenoNet, a novel algorithm for the identification of pathways and networks associated with different phenotypes. PhenoNet uses two types of input data: gene expression data (RMA, RPKM, FPKM, etc.) and phenotypic information, and integrates these data with curated pathways and protein-protein interaction information. Comprehensive iterations across all possible pathways and subnetworks result in the identification of key pathways or subnetworks that distinguish between the two phenotypes. Matlab code is available upon request. Supplementary data are available at Bioinformatics online.

Screening of Hub Genes and Pathways in Colorectal Cancer with Microarray Technology

Here we intend to identify key genes and pathways in the pathogenesis of colorectal cancer (CRC) through analyzing microarray data with bioinformatic tools. The gene expression profile dataset GSE23878 was downloaded from Gene Expression Omnibus and differentially expressed genes (DEGs) were screened out using Student's t-test. GO function and KEGG pathway enrichment analyses were performed for these DEGs with the DAVID online tool. Interaction network was constructed among the over-represented pathways based on the protein-protein interactions within the pathways. Besides, the protein interaction information obtained from HPRD database were applied to constructed protein-protein interaction networks among the DEGs and hub genes and function module were screened out. A total of 2,296 DEGs were obtained and they were enriched in 34 pathways. An interaction network was constructed among 32 pathways, in which p53 signaling pathway acted as the hub pathway as it showed the highest node degree. The protein-protein interaction network comprised 1,481 interaction relationships among 332 genes which included 40 DEGs. Further analysis revealed that theses DEGs formed 7 function modules and many genes, such as PDGFRB, MET, FZD2, CCND1, PRKCB, ARHGEF6, JUP, WNT2, WNT5A and WNT11 were key genes in the networks. The DEGs and disturbed biological functions uncovered in present study may play important roles in the development of CRC and can contribute to the understanding on molecular mechanisms of CRC. Further these DEGs we obtained can be acted as potential biomarkers for diagnosis and therapy of CRC.

Identifying the potential extracellular electron transfer pathways from a c-type cytochrome network.

Extracellular electron transfer (EET) is the key feature of some bacteria, such as Geobacter sulfurreducens and Shewanella oneidensis. Via EET processes, these bacteria can grow on electrode surfaces and make current output of microbial fuel cells. c-Type cytochromes can be used as carriers to transfer electrons, which play an important role in EET processes. Typically, from the inner (cytoplasmic) membrane through the periplasm to the outer membrane, they could form EET pathways. Recent studies suggest that a group of c-type cytochromes could form a network which extended the well-known EET pathways. We obtained the protein interaction information for all 41 c-type cytochromes in Shewanella oneidensis MR-1, constructed a large-scale protein interaction network, and studied its structural characteristics and functional significance. Centrality analysis has identified the top 10 key proteins of the network, and 7 of them are associated with electricity production in the bacteria, which suggests that the ability of Shewanella oneidensis MR-1 to produce electricity might be derived from the unique structure of the c-type cytochrome network. By modularity analysis, we obtained 5 modules from the network. The subcellular localization study has shown that the proteins in these modules all have diversiform cellular compartments, which reflects their potential to form EET pathways. In particular, combination of protein subcellular localization and operon analysis, the well-known and new candidate EET pathways are obtained from the Mtr-like module, indicating that potential EET pathways could be obtained from such a c-type cytochrome network.

Towards a Protein–Protein Interaction information extraction system: Recognizing named entities

The majority of biological functions of any living being are related to Protein–Protein Interactions (PPI). PPI discoveries are reported in form of research publications whose volume grows day after day. Consequently, automatic PPI information extraction systems are a pressing need for biologists. In this paper we are mainly concerned with the named entity detection module of PPIES (the PPI information extraction system we are implementing) which recognizes twelve entity types relevant in PPI context. It is composed of two sub-modules: a dictionary look-up with extensive normalization and acronym detection, and a Conditional Random Field classifier. The dictionary look-up module has been tested with Interaction Method Task (IMT), and it improves by approximately 10% the current solutions that do not use Machine Learning (ML). The second module has been used to create a classifier using the Joint Workshop on Natural Language Processing in Biomedicine and its Applications (JNLPBA’04) data set. It does not use any external resources, or complex or ad hoc post-processing, and obtains 77.25%, 75.04% and 76.13 for precision, recall, and F1-measure, respectively, improving all previous results obtained for this data set.

Functional Clustering of Immunoglobulin Superfamily Proteins with Protein–Protein Interaction Information Calibrated Hidden Markov Model Sequence Profiles

Secreted and cell-surface-localized members of the immunoglobulin superfamily (IgSF) play central roles in regulating adaptive and innate immune responses and are prime targets for the development of protein-based therapeutics. An essential activity of the ectodomains of these proteins is the specific recognition of cognate ligands, which are often other members of the IgSF. In this work, we provide functional insight for this important class of proteins through the development of a clustering algorithm that groups together extracellular domains of the IgSF with similar binding preferences. Information from hidden Markov model-based sequence profiles and domain architecture is calibrated against manually curated protein interaction data to define functional families of IgSF proteins. The method is able to assign 82% of the 477 extracellular IgSF protein to a functional family, while the rest are either single proteins with unique function or proteins that could not be assigned with the current technology. The functional clustering of IgSF proteins generates hypotheses regarding the identification of new cognate receptor–ligand pairs and reduces the pool of possible interacting partners to a manageable level for experimental validation.

Identifying Genes Relevant to Specific Biological Conditions in Time Course Microarray Experiments

Microarrays have been useful in understanding various biological processes by allowing the simultaneous study of the expression of thousands of genes. However, the analysis of microarray data is a challenging task. One of the key problems in microarray analysis is the classification of unknown expression profiles. Specifically, the often large number of non-informative genes on the microarray adversely affects the performance and efficiency of classification algorithms. Furthermore, the skewed ratio of sample to variable poses a risk of overfitting. Thus, in this context, feature selection methods become crucial to select relevant genes and, hence, improve classification accuracy. In this study, we investigated feature selection methods based on gene expression profiles and protein interactions. We found that in our setup, the addition of protein interaction information did not contribute to any significant improvement of the classification results. Furthermore, we developed a novel feature selection method that relies exclusively on observed gene expression changes in microarray experiments, which we call “relative Signal-to-Noise ratio” (rSNR). More precisely, the rSNR ranks genes based on their specificity to an experimental condition, by comparing intrinsic variation, i.e. variation in gene expression within an experimental condition, with extrinsic variation, i.e. variation in gene expression across experimental conditions. Genes with low variation within an experimental condition of interest and high variation across experimental conditions are ranked higher, and help in improving classification accuracy. We compared different feature selection methods on two time-series microarray datasets and one static microarray dataset. We found that the rSNR performed generally better than the other methods.

PPIExtractor: A Protein Interaction Extraction and Visualization System for Biomedical Literature

Protein-protein interactions (PPIs) play a key role in various aspects of the structural and functional organization of the cell. Knowledge about them unveils the molecular mechanisms of biological processes. However, the amount of biomedical literature regarding protein interactions is increasing rapidly and it is difficult for interaction database curators to detect and curate protein interaction information manually. In this paper, we present a PPI extraction system, termed PPIExtractor, which automatically extracts PPIs from biomedical text and visualizes them. Given a Medline record dataset, PPIExtractor first applies Feature Coupling Generalization (FCG) to tag protein names in text, next uses the extended semantic similarity-based method to normalize them, then combines feature-based, convolution tree and graph kernels to extract PPIs, and finally visualizes the PPI network. Experimental evaluations show that PPIExtractor can achieve state-of-the-art performance on a DIP subset with respect to comparable evaluations.

A new approach for prediction of tumor sensitivity to targeted drugs based on functional data

BackgroundThe success of targeted anti-cancer drugs are frequently hindered by the lack of knowledge of the individual pathway of the patient and the extreme data requirements on the estimation of the personalized genetic network of the patient’s tumor. The prediction of tumor sensitivity to targeted drugs remains a major challenge in the design of optimal therapeutic strategies. The current sensitivity prediction approaches are primarily based on genetic characterizations of the tumor sample. We propose a novel sensitivity prediction approach based on functional perturbation data that incorporates the drug protein interaction information and sensitivities to a training set of drugs with known targets.ResultsWe illustrate the high prediction accuracy of our framework on synthetic data generated from the Kyoto Encyclopedia of Genes and Genomes (KEGG) and an experimental dataset of four canine osteosarcoma tumor cultures following application of 60 targeted small-molecule drugs. We achieve a low leave one out cross validation error of <10% for the canine osteosarcoma tumor cultures using a drug screen consisting of 60 targeted drugs.ConclusionsThe proposed framework provides a unique input-output based methodology to model a cancer pathway and predict the effectiveness of targeted anti-cancer drugs. This framework can be developed as a viable approach for personalized cancer therapy.

SLiMScape: a protein short linear motif analysis plugin for Cytoscape

BackgroundComputational protein short linear motif discovery can use protein interaction information to search for motifs among proteins which share a common interactor. Cytoscape provides a visual interface for protein networks but there is no streamlined way to rapidly visualize motifs in a network of proteins, or to integrate computational discovery with such visualizations.ResultsWe present SLiMScape, a Cytoscape plugin, which enables both de novo motif discovery and searches for instances of known motifs. Data is presented using Cytoscape’s visualization features thus providing an intuitive interface for interpreting results. The distribution of discovered or user-defined motifs may be selectively displayed and the distribution of protein domains may be viewed simultaneously. To facilitate this SLiMScape automatically retrieves domains for each protein.ConclusionSLiMScape provides a platform for performing short linear motif analyses of protein interaction networks by integrating motif discovery and search tools in a network visualization environment. This significantly aids in the discovery of novel short linear motifs and in visualizing the distribution of known motifs.

The Protein Interaction Landscape of the Human CMGC Kinase Group

Cellular information processing via reversible protein phosphorylation requires tight control of the localization, activity, and substrate specificity of protein kinases, which to a large extent is accomplished by complex formation with other proteins. Despite their critical role in cellular regulation and pathogenesis, protein interaction information is available for only a subset of the 518 human protein kinases. Here we present a global proteomic analysis of complexes of the human CMGC kinase group. In addition to subgroup-specific functional enrichment and modularity, the identified 652 high-confidence kinase-protein interactions provide a specific biochemical context for many poorly studied CMGC kinases. Furthermore, the analysis revealed a kinase-kinase subnetwork and candidate substrates for CMGC kinases. Finally, the presented interaction proteome uncovered a large set of interactions with proteins genetically linked to a range of human diseases, including cancer, suggesting additional routes for analyzing the role of CMGC kinases in controlling human disease pathways.

Co-expression network with protein–protein interaction and transcription regulation in malaria parasite Plasmodium falciparum

BackgroundMalaria continues to be one of the most severe global infectious diseases, as a major threat to human health and economic development. Network-based biological analysis is a promising approach to uncover key genes and biological processes from a network viewpoint, which could not be recognized from individual gene-based signatures. ResultsWe integrated gene co-expression profile with protein–protein interaction and transcriptional regulation information to construct a comprehensive gene co-expression network of Plasmodium falciparum. Based on this network, we identified 10 core modules by using ICE (Iterative Clique Enumeration) algorithm, which were essential for malaria parasite development in intraerythrocytic developmental cycle (IDC) stages. In each module, all genes were highly correlated probably due to co-regulation or formation of a protein complex. Some of these genes were recognized to be differentially coexpressed among three close-by IDC stages. The gene of prpf8 (PFD0265w) encoding pre-mRNA processing splicing factor 8 product was identified as DCGs (differentially co-expressed genes) among IDC stages, although this gene function was seldom reported in previous researches. Integrating the species-specific gene prediction and differential co-expression gene detection, we found some modules could perform species-specific functions according to some of genes in these modules were species-specific genes, like the module 10.Furthermore, in order to reveal the underlying mechanisms of the erythrocyte invasion by P. falciparum, Steiner Tree algorithm was employed to identify the invasion subnetwork from our gene co-expression network. The subnetwork-based analysis indicated that some important Plasmodium parasite specific genes could corporate with each other and be co-regulated during the parasite invasion process, which including a head-to-head gene pair of PfRH2a (PF13_0198) and PfRH2b (MAL13P1.176). ConclusionsThis study based on gene co-expression network could shed new insights on the mechanisms of pathogenesis, even virulence and P. falciparum development.

Network-Induced Classification Kernels for Gene Expression Profile Analysis

Computational classification of gene expression profiles into distinct disease phenotypes has been highly successful to date. Still, robustness, accuracy, and biological interpretation of the results have been limited, and it was suggested that use of protein interaction information jointly with the expression profiles can improve the results. Here, we study three aspects of this problem. First, we show that interactions are indeed relevant by showing that co-expressed genes tend to be closer in the network of interactions. Second, we show that the improved performance of one extant method utilizing expression and interactions is not really due to the biological information in the network, while in another method this is not the case. Finally, we develop a new kernel method--called NICK--that integrates network and expression data for SVM classification, and demonstrate that overall it achieves better results than extant methods while running two orders of magnitude faster.

Literature mining of host–pathogen interactions: comparing feature-based supervised learning and language-based approaches

In an infectious disease, the pathogen's strategy to enter the host organism and breach its immune defenses often involves interactions between the host and pathogen proteins. Currently, the experimental data on host-pathogen interactions (HPIs) are scattered across multiple databases, which are often specialized to target a specific disease or host organism. An accurate and efficient method for the automated extraction of HPIs from biomedical literature is crucial for creating a unified repository of HPI data. Here, we introduce and compare two new approaches to automatically detect whether the title or abstract of a PubMed publication contains HPI data, and extract the information about organisms and proteins involved in the interaction. The first approach is a feature-based supervised learning method using support vector machines (SVMs). The SVM models are trained on the features derived from the individual sentences. These features include names of the host/pathogen organisms and corresponding proteins or genes, keywords describing HPI-specific information, more general protein-protein interaction information, experimental methods and other statistical information. The language-based method employed a link grammar parser combined with semantic patterns derived from the training examples. The approaches have been trained and tested on manually curated HPI data. When compared to a naïve approach based on the existing protein-protein interaction literature mining method, our approaches demonstrated higher accuracy and recall in the classification task. The most accurate, feature-based, approach achieved 66-73% accuracy, depending on the test protocol.