PPI-IRO: a two-stage method for protein-protein interaction extraction based on interaction relation ontology.

Automatic extraction of protein-protein interaction (PPI) information from scientific literature is important for building PPI databases, studying biological networks and discovering new biological knowledge through automatic hypothesis generation. In this paper, we present a new method for PPI extraction based on a mixture of logistic models. The method automatically clusters interaction words (words that describe the interactions of protein pairs) into groups with similar grammatical properties. Logistic models are fitted for each cluster of interaction words. Directionality of interactions is an essential piece of information for many protein interactions and important for building directed biological networks. Most of current PPI extraction methods do not extract the directional information of interactions. This is in part due to the lack of specific corpora with directionality information annotated. We introduce a new corpus, PICAD, for evaluating PPI extraction tools that includes directional annotation. The corpus is available at http://stat.fsu.edu/~jinfeng/resources/PICAD.txt. In addition, we propose an ensemble approach using logistic regression, Bayesian Networks, and SVM for identifying PPIs. We show that using an ensemble of classifiers allows us to capture different features in the text and report an F-measure of 75.7% using our new corpus.

Protein-protein interaction (PPI) plays an extremely important role in almost all life activities. The study of PPI has always been an important issue in the field of biomedicine. Extracting PPI information from the literature can provide important references for related research. In order to build an automated PPI extraction system, labeled corpora are required. However, annotated corpora are very limited, and annotating corpora is a time-consuming, labor-intensive, and costly task. The amount of unlabeled data is huge and easy to obtain, so it is of great significance to apply semi-supervised learning to PPI extraction. Existing semi-supervised methods cannot make good use of the information in unlabeled data. Therefore, a semi-supervised PPI extraction method based on knowledge distillation and virtual adversarial training is proposed in this work, which enhances the modeling ability of the target model for text features through knowledge distillation, and conducts semi-supervised learning through virtual adversarial training to better utilize a large number of unlabeled data. Experiments on public datasets for PPI extraction show that the proposed method can achieve competitive performance while using only a small amount of labeled data.

The state-of-the-art methods for protein-protein interaction (PPI) extraction are primarily based on kernel methods, and their performances strongly depend on the handcraft features. In this paper, we tackle PPI extraction by using convolutional neural networks (CNN) and propose a shortest dependency path based CNN (sdpCNN) model. The proposed method (1) only takes the sdp and word embedding as input and (2) could avoid bias from feature selection by using CNN. We performed experiments on standard Aimed and BioInfer datasets, and the experimental results demonstrated that our approach outperformed state-of-the-art kernel based methods. In particular, by tracking the sdpCNN model, we find that sdpCNN could extract key features automatically and it is verified that pretrained word embedding is crucial in PPI task.

Automatically extracting protein–protein interactions (PPIs) from biomedical literature provides additional support for precision medicine efforts. This paper proposes a novel memory network-based model (MNM) for PPI extraction, which leverages prior knowledge about protein–protein pairs with memory networks. The proposed MNM captures important context clues related to knowledge representations learned from knowledge bases. Both entity embeddings and relation embeddings of prior knowledge are effective in improving the PPI extraction model, leading to a new state-of-the-art performance on the BioCreative VI PPI dataset. The paper also shows that multiple computational layers over an external memory are superior to long short-term memory networks with the local memories.Database URL: http://www.biocreative.org/tasks/biocreative-vi/track-4/

Signal flow direction is one of the most important features of the protein-protein interactions in signaling networks. However, almost all the outcomes of current high-throughout techniques for protein-protein interactions mapping are usually supposed to be non-directional. Based on the pairwise interaction domains, here we defined a novel parameter protein interaction directional score and then used it to predict the direction of signal flow between proteins in proteome-wide signaling networks. Using 5-fold cross-validation, our approach obtained a satisfied performance with the accuracy 89.79%, coverage 48.08%, and error ratio 16.91%. As an application, we established an integrated human directional protein interaction network, including 2,237 proteins and 5,530 interactions, and inferred a large amount of novel signaling pathways. Directional protein interaction network was strongly supported by the known signaling pathways literature (with the 87.5% accuracy) and further analyses on the biological annotation, subcellular localization, and network topology property. Thus, this study provided an effective method to define the upstream/downstream relations of interacting protein pairs and a powerful tool to unravel the unknown signaling pathways.

We are developing an information extraction system for life science literature. We are currently focusing on PubMed abstracts and trying to extract named entities and their relationships, especially protein names and protein-protein interactions. We are adopting methods including natural language processing, machine learning, and text processing. But we are not developing a new tagging or parsing technique. Developing a new tagger or a new parser specialized in life science literature is a very complex job. And it is not easy to get a good result by tuning an existing parser or by training it without a sufficient corpus. These all are another research topics and we are trying to extract information, not to develop something to help the extracting job or else. In this paper, we introduce our method to use an existing full parser without training or tuning. After tagging sentences and extracting proteins, we make sentences simple by substituting some words like named entities, nouns into one word. Then parsing errors are reduced and parsing precision is increased by this sentence simplification. We parse the simplified sentences syntactically with an existing syntactic parser and extract protein-protein interactions from its results. We show the effects of sentence simplification and syntactic parsing.

Identification of biological regulatory networks from Process Hitting models

Understanding the complex interplay between proteins in physiological contexts poses a great challenge, since a striking number of human proteins still remain without solid functional annotation. To gain insight into the function of proteins encoded by human chromosome 21 (Hsa21) and to identify regulatory networks potentially relevant to Down syndrome (DS), 206 annotated open reading frames (ORFs) on Hsa21 were used as basis for amplification and cloning of ORFs and for collection of available full-length cDNA clones. This work resulted in the largest library of Hsa21 ORF clones available to date, covering 81% of the 206 annotated ORFs. A high success rate was reached in production of recombinant proteins (80%), ensuring appropriate downstream functional genomics analyses. For high-throughput determination of subcellular protein localizations, 89 constructs encoding N-terminally tagged proteins were introduced into HEK293T cells on transfected cell arrays. Localization information could be obtained for 52 out of the 89 Hsa21 proteins (58%). Of these, 28 subcellular localizations were described for the first time (published in Hu et al. 2006). All Hsa21 localization information has been entered into the public Gene Ontology (GO) database. For the identification of protein-protein interactions (PPIs), 53 constructs encoding non-autoactivating baits were screened against a prey set of 5,640 human proteins using a high- throughput mating array-based yeast two-hybrid (Y2H) system. Altogether, this mating array identified 56 new interactions for 24 different Hsa21 proteins. Twenty-three of the new interactions could be verified in an independent cotransformation Y2H set-up (56% of 41 tested). Cellular colocalization experiments in COS-1 cells and pull-down experiments confirmed five of six tested protein pairs (83%). All previously known protein interaction data for Hsa21 proteins were retrieved from available public sources. This resulted in a set of 684 new and known PPIs for 108 Hsa21 proteins, representing the largest Hsa21-centered interaction network to date. A proteome-wide interaction network was then generated by computational searches for indirect protein interactors in available data sets. Analysis of transcription factors showed a compelling inter-connection of thirteen Hsa21-encoded transcription factors. Detailed functional analysis of sub-networks involving NRIP1, UBE2G2, PCP4, MCM3AP and C21orf127 revealed novel functional properties of these Hsa21-encoded proteins. 35 Hsa21 proteins were found connected to nine signaling pathways, some of which have been previously implicated in the pathogenesis of DS, while others appear in this context for the first time. Further expansion of the Hsa21 interaction network will support the identification of new protein functions and regulatory networks. The data presented here shows that a Hsa21 interaction network is a useful resource for elucidating the biological contexts of new PPIs, providing an essential tool for the systems biology of Down syndrome.

Attaining a detailed understanding of the various biological networks in an organism lies at the core of the emerging discipline of systems biology. A precise description of the relationships formed between genes, mRNA molecules, and proteins is a necessary step toward a complete description of the dynamic behavior of an organism at the cellular level, and toward intelligent, efficient, and directed modification of an organism. The importance of understanding such regulatory, signaling, and interaction networks has fueled the development of numerous in silico inference algorithms, as well as new experimental techniques and a growing collection of public databases. The Software Environment for BIological Network Inference (SEBINI) has been created to provide an interactive environment for the deployment, evaluation, and improvement of algorithms used to reconstruct the structure of biological regulatory and interaction networks. SEBINI can be used to analyze high-throughput gene expression, protein abundance, or protein activation data via a suite of state-of-the-art network inference algorithms. It also allows algorithm developers to compare and train network inference methods on artificial networks and simulated gene expression perturbation data. SEBINI can therefore be used by software developers wishing to evaluate, refine, or combine inference techniques, as well as by bioinformaticians analyzing experimental data. Networks inferred from the SEBINI software platform can be further analyzed using the Collective Analysis of Biological Interaction Networks (CABIN) tool, which is an exploratory data analysis software that enables integration and analysis of protein-protein interaction and gene-to-gene regulatory evidence obtained from multiple sources. The collection of edges in a public database, along with the confidence held in each edge (if available), can be fed into CABIN as one "evidence network," using the Cytoscape SIF file format. Using CABIN, one may increase the confidence in individual edges in a network inferred by an algorithm in SEBINI, as well as extend such a network by combining it with species-specific or generic information, e.g., known protein-protein interactions or target genes identified for known transcription factors. Thus, the combined SEBINI-CABIN toolkit aids in the more accurate reconstruction of biological networks, with less effort, in less time.A demonstration web site for SEBINI can be accessed from https://www.emsl.pnl.gov/SEBINI/RootServlet . Source code and PostgreSQL database schema are available under open source license. ronald.taylor@pnl.gov. For commercial use, some algorithms included in SEBINI require licensing from the original developers. CABIN can be downloaded from http://www.sysbio.org/dataresources/cabin.stm . mudita.singhal@pnl.gov.

Feature representation is essential to machine learning and text mining. In this paper, we present a feature coupling generalization (FCG) framework for generating new features from unlabeled data. It selects two special types of features, i.e., example-distinguishing features (EDFs) and class-distinguishing features (CDFs) from original feature set, and then generalizes EDFs into higher-level features based on their coupling degrees with CDFs in unlabeled data. The advantage is: EDFs with extreme sparsity in labeled data can be enriched by their co-occurrences with CDFs in unlabeled data so that the performance of these low-frequency features can be greatly boosted and new information from unlabeled can be incorporated. We apply this approach to three tasks in biomedical literature mining: gene named entity recognition (NER), protein-protein interaction extraction (PPIE), and text classification (TC) for gene ontology (GO) annotation. New features are generated from over 20 GB unlabeled PubMed abstracts. The experimental results on BioCreative 2, AIMED corpus, and TREC 2005 Genomics Track show that 1) FCG can utilize well the sparse features ignored by supervised learning. 2) It improves the performance of supervised baselines by 7.8 percent, 5.0 percent, and 5.8 percent, respectively, in the tree tasks. 3) Our methods achieve 89.1, 64.5 F-score, and 60.1 normalized utility on the three benchmark data sets.

Alzheimer's disease (AD) is the most prevalent neurological disorder worldwide, affecting approximately 24 million individuals. Despite more than a century of research on AD, its pathophysiology is still not fully understood. Recently, genetic studies of AD have focused on analyzing the general expression profile by employing high-throughput genomic techniques such as microarrays. Current research has leveraged bioinformatics advancements in genetic science to build upon previous efforts. Data from the GSE118553 dataset used in this investigation, and the analyses carried out using programs such as Limma and BioBase. Differentially expressed genes (DEGs) and differentially expressed microRNAs (DEmiRs) associated with AD identified in the studied areas of the brain. Target genes of the DEmiRs identified using the MultiMiR package. Gene ontology (GO) completed using the Enrichr website, and the protein-protein interaction (PPI) network for these genes drawn using STRING and Cytoscape software. The findings introduced DEGs including CTNNB1, PAK2, MAP2K1, PNPLA6, IGF1R, FOXL2, DKK3, LAMA4, PABPN1, and GDPD5, and DEmiRs linked to AD (miR-106A, miR-1826, miR-1253, miR-10B, miR-18B, miR-101-2, miR-761, miR-199A1, miR-379 and miR-668), (miR-720, miR-218-2, miR-25, miR-602, miR-1226, miR-548K, miR-H1, miR-410, miR-548F2, miR-181A2), (miR-1470, miR-651, miR-544, miR-1826, miR-195, miR-610, miR-599, miR-323, miR-587 and miR-340), and (miR-1282, miR-1914, miR-642, miR-1323, miR-373, miR-323, miR-1322, miR-612, miR-606 and miR-758) in cerebellum, frontal cortex, temporal cortex, and entorhinal cortex, respectively. The majority of the genes and miRNAs identified by our findings may be employed as biomarkers for prediction, diagnosis, or therapy response monitoring.

The aim of this study was to mine cartilage damage and regeneration-related biomarkers and identify the gene regulatory networks of cartilage damage. A gene expression data set (GSE129147) containing damaged and control samples collected from the knee of the same patients was employed. R package limma was used to identify differentially expressed genes (DEGs), and clusterProfiler was performed for the GO and KEGG functional enrichment analysis. Cytoscape plug-ins of CytoHubba and MCODE were applied to investigate protein-protein interaction (PPI) network, modules, and hub genes. We identified 422 DEGs that were involved in skeletal system development, bone development, ossification, mesenchyme development, mesenchymal cell differentiation, connective tissue development, osteoblast differentiation, and extracellular matrix. We dug out 30 hub genes, identified three PPI modules, and constructed a miRNA regulatory network for DEGs. The miRNAs of the DEGs were predicted by miRNet, and the miRNA-mRNA network displayed some important miRNAs such as miR-335-5p, miR-92a-3p, and miR-98-5p. Collectively, these results have the potential to clarify the mechanism of cartilage damage and to assist us in discovering the damage and repair-related biomarkers.

The Impact of Gene Expression Regulation on Evolution of Extracellular Signaling Pathways

Exploring gene regulatory interaction networks and predicting therapeutic molecules among major depressive disorder and bipolar disorder: A bioinformatics approach.

Functional module identification in biological networks may provide new insights into the complex interactions among biomolecules for a better understanding of cellular functional organization. Most of existing functional module identification methods are based on the optimization of network modularity and cluster networks into groups of nodes within which there are a higher-than-expectation number of edges. However, module identification simply based on this topological criterion may not discover certain kinds of biologically meaningful modules within which nodes are sparsely connected but have similar interaction patterns with the rest of the network. In order to unearth more biologically meaningful functional modules, we propose a novel efficient convex programming algorithm based on the subgradient method with heuristic path generation to solve the problem in a recently proposed framework of blockmodel module identification. We have implemented our algorithm for large-scale protein-protein interaction (PPI) networks, including Saccharomyces cerevisia and Homo sapien PPI networks collected from the Database of Interaction Proteins (DIP) and Human Protein Reference Database (HPRD). Our experimental results have shown that our algorithm achieves comparable network clustering performance in comparison to the more time-consuming simulated annealing (SA) optimization. Furthermore, preliminary results for identifying fine-grained functional modules in both biological networks and the comparison with the commonly adopted Markov Clustering (MCL) algorithm have demonstrated the potential of our algorithm to discover new types of modules, within which proteins are sparsely connected but with significantly enriched biological functionalities.