Automated extraction of information on protein-protein interactions from the biological literature

MOTIVATION: To understand biological process, we must clarify how proteins interact with each other. However, since information about protein-protein interactions still exists primarily in the scientific literature, it is not accessible in a computer-readable format. Efficient processing of large amounts of interactions therefore needs an intelligent information extraction method. Our aim is to develop an efficient method for extracting information on protein-protein interaction from scientific literature. RESULTS: We present a method for extracting information on protein-protein interactions from the scientific literature. This method, which employs only a protein name dictionary, surface clues on word patterns and simple part-of-speech rules, achieved high recall and precision rates for yeast (recall = 86.8% and precision = 94.3%) and Escherichia coli (recall = 82.5% and precision = 93.5%). The result of extraction suggests that our method should be applicable to any species for which a protein name dictionary is constructed. AVAILABILITY: The program is available on request from the authors.     

Discovering patterns to extract protein-protein interactions from full texts

MOTIVATION: Although there are several databases storing protein-protein interactions, most such data still exist only in the scientific literature. They are scattered in scientific literature written in natural languages, defying data mining efforts. Much time and labor have to be spent on extracting protein pathways from literature. Our aim is to develop a robust and powerful methodology to mine protein-protein interactions from biomedical texts. RESULTS: We present a novel and robust approach for extracting protein-protein interactions from literature. Our method uses a dynamic programming algorithm to compute distinguishing patterns by aligning relevant sentences and key verbs that describe protein interactions. A matching algorithm is designed to extract the interactions between proteins. Equipped only with a dictionary of protein names, our system achieves a recall rate of 80.0% and precision rate of 80.5%. AVAILABILITY: The program is available on request from the authors.     

Mining physical protein-protein interactions from the literature

Background:

Deciphering physical protein-protein interactions is fundamental to elucidating both the functions of proteins and biological processes. The development of high-throughput experimental technologies such as the yeast two-hybrid screening has produced an explosion in data relating to interactions. Since manual curation is intensive in terms of time and cost, there is an urgent need for text-mining tools to facilitate the extraction of such information. The BioCreative (Critical Assessment of Information Extraction systems in Biology) challenge evaluation provided common standards and shared evaluation criteria to enable comparisons among different approaches.

Results:

During the benchmark evaluation of BioCreative 2006, all of our results ranked in the top three places. In the task of filtering articles irrelevant to physical protein interactions, our method contributes a precision of 75.07%, a recall of 81.07%, and an AUC (area under the receiver operating characteristic curve) of 0.847. In the task of identifying protein mentions and normalizing mentions to molecule identifiers, our method is competitive among runs submitted, with a precision of 34.83%, a recall of 24.10%, and an F₁ score of28.5%. In extracting protein interaction pairs, our profile-based method was competitive on the SwissProt-only subset (precision = 36.95%, recall = 32.68%, and F₁ score = 30.40%) and on the entire dataset (30.96%, 29.35%, and26.20%, respectively). From the biologist's point of view, however, these findings are far from satisfactory. The error analysis presented in this report provides insight into how performance could be improved: three-quarters of false negatives were due to protein normalization problems (532/698), and about one-quarter were due to problems with correctly extracting interactions for this system.

Conclusion:

We present a text-mining framework to extract physical protein-protein interactions from the literature. Three key issues are addressed, namely filtering irrelevant articles, identifying protein names and normalizing them to molecule identifiers, and extracting protein-protein interactions. Our system is among the top three performers in the benchmark evaluation of BioCreative 2006. The tool will be helpful for manual interaction curation and can greatly facilitate the process of extracting protein-protein interactions.

     

Augmented transitive relationships with high impact protein distillation in protein interaction prediction

Predicting new protein-protein interactions is important for discovering novel functions of various biological pathways. Predicting these interactions is a crucial and challenging task. Moreover, discovering new protein-protein interactions through biological experiments is still difficult. Therefore, it is increasingly important to discover new protein interactions. Many studies have predicted protein-protein interactions, using biological features such as Gene Ontology (GO) functional annotations and structural domains of two proteins. In this paper, we propose an augmented transitive relationships predictor (ATRP), a new method of predicting potential protein interactions using transitive relationships and annotations of protein interactions. In addition, a distillation of virtual direct protein-protein interactions is proposed to deal with unbalanced distribution of different types of interactions in the existing protein-protein interaction databases. Our results demonstrate that ATRP can effectively predict protein-protein interactions. ATRP achieves an 81% precision, a 74% recall and a 77% F-measure in average rate in the prediction of direct protein-protein interactions. Using the generated benchmark datasets from KUPS to evaluate of all types of the protein-protein interaction, ATRP achieved a 93% precision, a 49% recall and a 64% F-measure in average rate. This article is part of a Special Issue entitled: Computational Methods for Protein Interaction and Structural Prediction.     

Whole-proteome interaction mining

MOTIVATION: A major post-genomic scientific and technological pursuit is to describe the functions performed by the proteins encoded by the genome. One strategy is to first identify the protein-protein interactions in a proteome, then determine pathways and overall structure relating these interactions, and finally to statistically infer functional roles of individual proteins. Although huge amounts of genomic data are at hand, current experimental protein interaction assays must overcome technical problems to scale-up for high-throughput analysis. In the meantime, bioinformatics approaches may help bridge the information gap required for inference of protein function. In this paper, a previously described data mining approach to prediction of protein-protein interactions (Bock and Gough, 2001, Bioinformatics, 17, 455-460) is extended to interaction mining on a proteome-wide scale. An algorithm (the phylogenetic bootstrap) is introduced, which suggests traversal of a phenogram, interleaving rounds of computation and experiment, to develop a knowledge base of protein interactions in genetically-similar organisms. RESULTS: The interaction mining approach was demonstrated by building a learning system based on 1,039 experimentally validated protein-protein interactions in the human gastric bacterium Helicobacter pylori. An estimate of the generalization performance of the classifier was derived from 10-fold cross-validation, which indicated expected upper bounds on precision of 80% and sensitivity of 69% when applied to related organisms. One such organism is the enteric pathogen Campylobacter jejuni, in which comprehensive machine learning prediction of all possible pairwise protein-protein interactions was performed. The resulting network of interactions shares an average protein connectivity characteristic in common with previous investigations reported in the literature, offering strong evidence supporting the biological feasibility of the hypothesized map. For inferences about complete proteomes in which the number of pairwise non-interactions is expected to be much larger than the number of actual interactions, we anticipate that the sensitivity will remain the same but precision may decrease. We present specific biological examples of two subnetworks of protein-protein interactions in C. jejuni resulting from the application of this approach, including elements of a two-component signal transduction systems for thermoregulation, and a ferritin uptake network.     

Protein complex prediction based on maximum matching with domain-domain interaction

With the development of high-throughput methods for identifying protein-protein interactions, large scale interaction networks are available. Computational methods to analyze the networks to detect functional modules as protein complexes are becoming more important. However, most of the existing methods only make use of the protein-protein interaction networks without considering the structural limitations of proteins to bind together. In this paper, we design a new protein complex prediction method by extending the idea of using domain-domain interaction information. Here we formulate the problem into a maximum matching problem (which can be solved in polynomial time) instead of the binary integer linear programming approach (which can be NP-hard in the worst case). We also add a step to predict domain-domain interactions which first searches the database Pfam using the hidden Markov model and then predicts the domain-domain interactions based on the database DOMINE and InterDom which contain confirmed DDIs. By adding the domain-domain interaction prediction step, we have more edges in the DDI graph and the recall value is increased significantly (at least doubled) comparing with the method of Ozawa et al. (2010) [1] while the average precision value is slightly better. We also combine our method with three other existing methods, such as COACH, MCL and MCODE. Experiments show that the precision of the combined method is improved. This article is part of a Special Issue entitled: Computational Methods for Protein Interaction and Structural Prediction.     

Automatic Peak Selection by a Benjamini-Hochberg-Based Algorithm

A common issue in bioinformatics is that computational methods often generate a large number of predictions sorted according to certain confidence scores. A key problem is then determining how many predictions must be selected to include most of the true predictions while maintaining reasonably high precision. In nuclear magnetic resonance (NMR)-based protein structure determination, for instance, computational peak picking methods are becoming more and more common, although expert-knowledge remains the method of choice to determine how many peaks among thousands of candidate peaks should be taken into consideration to capture the true peaks. Here, we propose a Benjamini-Hochberg (B-H)-based approach that automatically selects the number of peaks. We formulate the peak selection problem as a multiple testing problem. Given a candidate peak list sorted by either volumes or intensities, we first convert the peaks into -values and then apply the B-H-based algorithm to automatically select the number of peaks. The proposed approach is tested on the state-of-the-art peak picking methods, including WaVPeak [1] and PICKY [2]. Compared with the traditional fixed number-based approach, our approach returns significantly more true peaks. For instance, by combining WaVPeak or PICKY with the proposed method, the missing peak rates are on average reduced by 20% and 26%, respectively, in a benchmark set of 32 spectra extracted from eight proteins. The consensus of the B-H-selected peaks from both WaVPeak and PICKY achieves 88% recall and 83% precision, which significantly outperforms each individual method and the consensus method without using the B-H algorithm. The proposed method can be used as a standard procedure for any peak picking method and straightforwardly applied to some other prediction selection problems in bioinformatics. The source code, documentation and example data of the proposed method is available at http://sfb.kaust.edu.sa/pages/software.aspx.     

An integrated probabilistic model for functional prediction of proteins.

We develop an integrated probabilistic model to combine protein physical interactions, genetic interactions, highly correlated gene expression networks, protein complex data, and domain structures of individual proteins to predict protein functions. The model is an extension of our previous model for protein function prediction based on Markovian random field theory. The model is flexible in that other protein pairwise relationship information and features of individual proteins can be easily incorporated. Two features distinguish the integrated approach from other available methods for protein function prediction. One is that the integrated approach uses all available sources of information with different weights for different sources of data. It is a global approach that takes the whole network into consideration. The second feature is that the posterior probability that a protein has the function of interest is assigned. The posterior probability indicates how confident we are about assigning the function to the protein. We apply our integrated approach to predict functions of yeast proteins based upon MIPS protein function classifications and upon the interaction networks based on MIPS physical and genetic interactions, gene expression profiles, tandem affinity purification (TAP) protein complex data, and protein domain information. We study the recall and precision of the integrated approach using different sources of information by the leave-one-out approach. In contrast to using MIPS physical interactions only, the integrated approach combining all of the information increases the recall from 57% to 87% when the precision is set at 57%-an increase of 30%.     

Inferring protein-protein interactions through high-throughput interaction data from diverse organisms

MOTIVATION: Identifying protein-protein interactions is critical for understanding cellular processes. Because protein domains represent binding modules and are responsible for the interactions between proteins, computational approaches have been proposed to predict protein interactions at the domain level. The fact that protein domains are likely evolutionarily conserved allows us to pool information from data across multiple organisms for the inference of domain-domain and protein-protein interaction probabilities. RESULTS: We use a likelihood approach to estimating domain-domain interaction probabilities by integrating large-scale protein interaction data from three organisms, Saccharomyces cerevisiae, Caenorhabditis elegans and Drosophila melanogaster. The estimated domain-domain interaction probabilities are then used to predict protein-protein interactions in S.cerevisiae. Based on a thorough comparison of sensitivity and specificity, Gene Ontology term enrichment and gene expression profiles, we have demonstrated that it may be far more informative to predict protein-protein interactions from diverse organisms than from a single organism. AVAILABILITY: The program for computing the protein-protein interaction probabilities and supplementary material are available at http://bioinformatics.med.yale.edu/interaction.     

Prediction of protein function using protein-protein interaction data.

Assigning functions to novel proteins is one of the most important problems in the postgenomic era. Several approaches have been applied to this problem, including the analysis of gene expression patterns, phylogenetic profiles, protein fusions, and protein-protein interactions. In this paper, we develop a novel approach that employs the theory of Markov random fields to infer a protein's functions using protein-protein interaction data and the functional annotations of protein's interaction partners. For each function of interest and protein, we predict the probability that the protein has such function using Bayesian approaches. Unlike other available approaches for protein annotation in which a protein has or does not have a function of interest, we give a probability for having the function. This probability indicates how confident we are about the prediction. We employ our method to predict protein functions based on "biochemical function," "subcellular location," and "cellular role" for yeast proteins defined in the Yeast Proteome Database (YPD, www.incyte.com), using the protein-protein interaction data from the Munich Information Center for Protein Sequences (MIPS, mips.gsf.de). We show that our approach outperforms other available methods for function prediction based on protein interaction data. The supplementary data is available at www-hto.usc.edu/~msms/ProteinFunction.     

Feature-based classification of native and non-native protein-protein interactions: Comparing supervised and semi-supervised learning approaches

Structural knowledge about protein-protein interactions can provide insights to the basic processes underlying cell function. Recent progress in experimental and computational structural biology has led to a rapid growth of experimentally resolved structures and computationally determined near-native models of protein-protein interactions. However, determining whether a protein-protein interaction is physiological or it is the artifact of an experimental or computational method remains a challenging problem. In this work, we have addressed two related problems. The first problem is distinguishing between the experimentally obtained physiological and crystal-packing protein-protein interactions. The second problem is concerned with the classification of near-native and inaccurate docking models. We first defined a universal set of interface features and employed a support vector machines (SVM)-based approach to classify the interactions for both problems, with the accuracy, precision, and recall for the first problem classifier reaching 93%. To improve the classification, we next developed a semi-supervised learning approach for the second problem, using transductive SVM (TSVM). We applied both classifiers to a commonly used protein docking benchmark of 124 complexes. We found that while we reached the classification accuracies of 78.9% for the SVM classifier and 80.3% for the TSVM classifier, improving protein-docking methods by model re-ranking remains a challenging problem.     

Mining literature for protein-protein interactions

MOTIVATION: A central problem in bioinformatics is how to capture information from the vast current scientific literature in a form suitable for analysis by computer. We address the special case of information on protein-protein interactions, and show that the frequencies of words in Medline abstracts can be used to determine whether or not a given paper discusses protein-protein interactions. For those papers determined to discuss this topic, the relevant information can be captured for the Database of Interacting PROTEINS: Furthermore, suitable gene annotations can also be captured. RESULTS: Our Bayesian approach scores Medline abstracts for probability of discussing the topic of interest according to the frequencies of discriminating words found in the abstract. More than 80 discriminating words (e.g. complex, interaction, two-hybrid) were determined from a training set of 260 Medline abstracts corresponding to previously validated entries in the Database of Interacting Proteins. Using these words and a log likelihood scoring function, approximately 2000 Medline abstracts were identified as describing interactions between yeast proteins. This approach now forms the basis for the rapid expansion of the Database of Interacting Proteins.     

Protein evolution is faster outside the cell

Some proteins are highly conserved across all species, whereas others diverge significantly even between closely related species. Attempts have been made to correlate the rate of protein evolution to amino acid composition, protein dispensability, and the number of protein-protein interactions, but in all cases, conflicting studies have shown that the theories are hard to confirm experimentally. The only correlation that is undisputed so far is that highly/broadly expressed proteins seem to evolve at a lower rate. Consequently, it has been suggested that correlations between evolution rate and factors like protein dispensability or the number of protein-protein interactions could be just secondary effects due to differences in expression. The purpose of this study was to analyze mammalian proteins/genes with known subcellular location for variations in evolution rates. We show that proteins that are exported (extracellular proteins) evolve faster than proteins that reside inside the cell (intracellular proteins). We find weak, but significant, correlations between evolution rates and expression levels, percentage of tissues in which the proteins are expressed (expression broadness), and the number of protein interaction partners. More important, we show that the observed difference in evolution rate between extra- and intracellular proteins is largely independent of expression levels, expression broadness, and the number of protein-protein interactions. We also find that the difference is not caused by an overrepresentation of immunological proteins or disulfide bridge-containing proteins among the extracellular data set. We conclude that the subcellular location of a mammalian protein has a larger effect on its evolution rate than any of the other factors studied in this paper, including expression levels/patterns. We observe a difference in evolution rates between extracellular and intracellular proteins for a yeast data set as well and again show that it is completely independent of expression levels.     

Extraction of protein interaction information from unstructured text using a context-free grammar

MOTIVATION: As research into disease pathology and cellular function continues to generate vast amounts of data pertaining to protein, gene and small molecule (PGSM) interactions, there exists a critical need to capture these results in structured formats allowing for computational analysis. Although many efforts have been made to create databases that store this information in computer readable form, populating these sources largely requires a manual process of interpreting and extracting interaction relationships from the biological research literature. Being able to efficiently and accurately automate the extraction of interactions from unstructured text, would greatly improve the content of these databases and provide a method for managing the continued growth of new literature being published. RESULTS: In this paper, we describe a system for extracting PGSM interactions from unstructured text. By utilizing a lexical analyzer and context free grammar (CFG), we demonstrate that efficient parsers can be constructed for extracting these relationships from natural language with high rates of recall and precision. Our results show that this technique achieved a recall rate of 83.5% and a precision rate of 93.1% for recognizing PGSM names and a recall rate of 63.9% and a precision rate of 70.2% for extracting interactions between these entities. In contrast to other published techniques, the use of a CFG significantly reduces the complexities of natural language processing by focusing on domain specific structure as opposed to analyzing the semantics of a given language. Additionally, our approach provides a level of abstraction for adding new rules for extracting other types of biological relationships beyond PGSM relationships. AVAILABILITY: The program and corpus are available by request from the authors.     

Finding the evidence for protein-protein interactions from PubMed abstracts

MOTIVATION: Protein-protein interactions play critical roles in biological processes, and many biologists try to find or to predict crucial information concerning these interactions. Before verifying interactions in biological laboratory work, validating them from previous research is necessary. Although many efforts have been made to create databases that store verified information in a structured form, much interaction information still remains as unstructured text. As the amount of new publications has increased rapidly, a large amount of research has sought to extract interactions from the text automatically. However, there remain various difficulties associated with the process of applying automatically generated results into manually annotated databases. For interactions that are not found in manually stored databases, researchers attempt to search for abstracts or full papers. RESULTS: As a result of a search for two proteins, PubMed frequently returns hundreds of abstracts. In this paper, a method is introduced that validates protein-protein interactions from PubMed abstracts. A query is generated from two given proteins automatically and abstracts are then collected from PubMed. Following this, target proteins and their synonyms are recognized and their interaction information is extracted from the collection. It was found that 67.37% of the interactions from DIP-PPI corpus were found from the PubMed abstracts and 87.37% of interactions were found from the given full texts. AVAILABILITY: Contact authors.     

Prediction of functional tertiary interactions and intermolecular interfaces from primary sequence data

Given the availability of sequence information for many species, one can examine how the sequence of a gene varies among different organisms. This is accomplished by aligning the sequences and observing patterns of conservation, mutation and counter-mutation at different positions in the gene. Imbedded in these patterns is information on energetic coupling and macromolecular interactions, which can be deciphered by application of statistical algorithms. Here we report a robust approach for predicting interactions within (or between) any type of biopolymer, including proteins, RNAs and RNA-protein complexes. Rather than maximize the number of predictions, this approach is designed to detect a limited number of highly significant interactions, thereby providing accurate results from alignments that contain a modest number of sequences (20-60). The versatility and accuracy of the algorithm is demonstrated by the successful prediction of important intramolecular interactions within RNAs, modified RNAs, and proteins, as well as the prediction of RNA-protein and protein-protein interactions.     

Discovering regulatory and signalling circuits in molecular interaction networks

MOTIVATION: In model organisms such as yeast, large databases of protein-protein and protein-DNA interactions have become an extremely important resource for the study of protein function, evolution, and gene regulatory dynamics. In this paper we demonstrate that by integrating these interactions with widely-available mRNA expression data, it is possible to generate concrete hypotheses for the underlying mechanisms governing the observed changes in gene expression. To perform this integration systematically and at large scale, we introduce an approach for screening a molecular interaction network to identify active subnetworks, i.e., connected regions of the network that show significant changes in expression over particular subsets of conditions. The method we present here combines a rigorous statistical measure for scoring subnetworks with a search algorithm for identifying subnetworks with high score. RESULTS: We evaluated our procedure on a small network of 332 genes and 362 interactions and a large network of 4160 genes containing all 7462 protein-protein and protein-DNA interactions in the yeast public databases. In the case of the small network, we identified five significant subnetworks that covered 41 out of 77 (53%) of all significant changes in expression. Both network analyses returned several top-scoring subnetworks with good correspondence to known regulatory mechanisms in the literature. These results demonstrate how large-scale genomic approaches may be used to uncover signalling and regulatory pathways in a systematic, integrative fashion.