Search

Abstract Background Proteins often assemble into higher-order complexes to perform their biological functions. Such protein–protein interactions (PPI) are often experimentally measured for pairs of proteins and summarized in a weighted PPI network, to which community detection algorithms can be applied to define the various higher-order protein complexes. Current methods include unsupervised and supervised approaches, often assuming that protein complexes manifest only as dense subgraphs. Utilizing supervised approaches, the focus is not on how to find them in a network, but only on learning which subgraphs correspond to complexes, currently solved using heuristics. However, learning to walk trajectories on a network to identify protein complexes leads naturally to a reinforcement learning (RL) approach, a strategy not extensively explored for community detection. Here, we develop and evaluate a reinforcement learning pipeline for community detection on weighted protein–protein interaction networks to detect new protein complexes. The algorithm is trained to calculate the value of different subgraphs encountered while walking on the network to reconstruct known complexes. A distributed prediction algorithm then scales the RL pipeline to search for novel protein complexes on large PPI networks. Results The reinforcement learning pipeline is applied to a human PPI network consisting of 8k proteins and 60k PPI, which results in 1,157 protein complexes. The method demonstrated competitive accuracy with improved speed compared to previous algorithms. We highlight protein complexes such as C4orf19, C18orf21, and KIAA1522 which are currently minimally characterized. Additionally, the results suggest TMC04 be a putative additional subunit of the KICSTOR complex and confirm the involvement of C15orf41 in a higher-order complex with HIRA, CDAN1, ASF1A, and by 3D structural modeling. Conclusions Reinforcement learning offers several distinct advantages for community detection, including scalability and knowledge of the walk trajectories defining those communities. Applied to currently available human protein interaction networks, this method had comparable accuracy with other algorithms and notable savings in computational time, and in turn, led to clear predictions of protein function and interactions for several uncharacterized human proteins.

Background: Proteins often assemble into higher-order complexes to perform their biological functions. Such protein–protein interactions (PPI) are often experimentally measured for pairs of proteins and summarized in a weighted PPI network, to which community detection algorithms can be applied to define the various higher-order protein complexes. Current methods include unsupervised and supervised approaches, often assuming that protein complexes manifest only as dense subgraphs. Utilizing supervised approaches, the focus is not on how to find them in a network, but only on learning which subgraphs correspond to complexes, currently solved using heuristics. However, learning to walk trajectories on a network to identify protein complexes leads naturally to a reinforcement learning (RL) approach, a strategy not extensively explored for community detection. Here, we develop and evaluate a reinforcement learning pipeline for community detection on weighted protein–protein interaction networks to detect new protein complexes. The algorithm is trained to calculate the value of different subgraphs encountered while walking on the network to reconstruct known complexes. A distributed prediction algorithm then scales the RL pipeline to search for novel protein complexes on large PPI networks. Results: The reinforcement learning pipeline is applied to a human PPI network consisting of 8k proteins and 60k PPI, which results in 1,157 protein complexes. The method demonstrated competitive accuracy with improved speed compared to previous algorithms. We highlight protein complexes such as C4orf19, C18orf21, and KIAA1522 which are currently minimally characterized. Additionally, the results suggest TMC04 be a putative additional subunit of the KICSTOR complex and confirm the involvement of C15orf41 in a higher-order complex with HIRA, CDAN1, ASF1A, and by 3D structural modeling. Conclusions: Reinforcement learning offers several distinct advantages for community detection, including scalability and knowledge of the walk trajectories defining those communities. Applied to currently available human protein interaction networks, this method had comparable accuracy with other algorithms and notable savings in computational time, and in turn, led to clear predictions of protein function and interactions for several uncharacterized human proteins. [ABSTRACT FROM AUTHOR]

Abstract Background Structural modeling of protein-protein interactions produces a large number of putative configurations of the protein complexes. Identification of the near-native models among them is a serious challenge. Publicly available results of biomedical research may provide constraints on the binding mode, which can be essential for the docking. Our text-mining (TM) tool, which extracts binding site residues from the PubMed abstracts, was successfully applied to protein docking (Badal et al., PLoS Comput Biol, 2015; 11: e1004630). Still, many extracted residues were not relevant to the docking. Results We present an extension of the TM tool, which utilizes natural language processing (NLP) for analyzing the context of the residue occurrence. The procedure was tested using generic and specialized dictionaries. The results showed that the keyword dictionaries designed for identification of protein interactions are not adequate for the TM prediction of the binding mode. However, our dictionary designed to distinguish keywords relevant to the protein binding sites led to considerable improvement in the TM performance. We investigated the utility of several methods of context analysis, based on dissection of the sentence parse trees. The machine learning-based NLP filtered the pool of the mined residues significantly more efficiently than the rule-based NLP. Constraints generated by NLP were tested in docking of unbound proteins from the DOCKGROUND X-ray benchmark set 4. The output of the global low-resolution docking scan was post-processed, separately, by constraints from the basic TM, constraints re-ranked by NLP, and the reference constraints. The quality of a match was assessed by the interface root-mean-square deviation. The results showed significant improvement of the docking output when using the constraints generated by the advanced TM with NLP. Conclusions The basic TM procedure for extracting protein-protein binding site residues from the PubMed abstracts was significantly advanced by the deep parsing (NLP techniques for contextual analysis) in purging of the initial pool of the extracted residues. Benchmarking showed a substantial increase of the docking success rate based on the constraints generated by the advanced TM with NLP.

Background: Protein-protein interactions (PPIs) can offer compelling evidence for protein function, especially when viewed in the context of proteome-wide interactomes. Bacteria have been popular subjects of interactome studies: more than six different bacterial species have been the subjects of comprehensive interactome studies while several more have had substantial segments of their proteomes screened for interactions. The protein interactomes of several bacterial species have been completed, including several from prominent human pathogens. The availability of interactome data has brought challenges, as these large data sets are difficult to compare across species, limiting their usefulness for broad studies of microbial genetics and evolution. Results: In this study, we use more than 52,000 unique protein-protein interactions (PPIs) across 349 different bacterial species and strains to determine their conservation across data sets and taxonomic groups. When proteins are collapsed into orthologous groups (OGs) the resulting meta-interactome still includes more than 43,000 interactions, about 14,000 of which involve proteins of unknown function. While conserved interactions provide support for protein function in their respective species data, we found only 429 PPIs (∼1% of the available data) conserved in two or more species, rendering any cross-species interactome comparison immediately useful. The meta-interactome serves as a model for predicting interactions, protein functions, and even full interactome sizes for species with limited to no experimentally observed PPI, including Bacillus subtilis and Salmonella enterica which are predicted to have up to 18,000 and 31,000 PPIs, respectively. Conclusions: In the course of this work, we have assembled cross-species interactome comparisons that will allow interactomics researchers to anticipate the structures of yet-unexplored microbial interactomes and to focus on well-conserved yet uncharacterized interactors for further study. Such conserved interactions should provide evidence for important but yet-uncharacterized aspects of bacterial physiology and may provide targets for anti-microbial therapies. [ABSTRACT FROM AUTHOR]

Background: Proteins play an important role in biological processes in living organisms. Many protein functions are based on interaction with other proteins. The structural information is important for adequate description of these interactions. Sets of protein structures determined in both bound and unbound states are essential for benchmarking of the docking procedures. However, the number of such proteins in PDB is relatively small. A radical expansion of such sets is possible if the unbound structures are computationally simulated. Results: The DOCKGROUND public resource provides data to improve our understanding of protein-protein interactions and to assist in the development of better tools for structural modeling of protein complexes, such as docking algorithms and scoring functions. A large set of simulated unbound protein structures was generated from the bound structures. The modeling protocol was based on 1 ns Langevin dynamics simulation. The simulated structures were validated on the ensemble of experimentally determined unbound and bound structures. The set is intended for large scale benchmarking of docking algorithms and scoring functions. Conclusions: A radical expansion of the unbound protein docking benchmark set was achieved by simulating the unbound structures. The simulated unbound structures were selected according to criteria from systematic comparison of experimentally determined bound and unbound structures. The set is publicly available at http://dockground.compbio.ku.edu. [ABSTRACT FROM AUTHOR]

Background: Protein interactions play a key role in life processes. Characterization of conformational properties of protein-protein interactions is important for understanding the mechanisms of protein association. The rapidly increasing amount of experimentally determined structures of proteins and protein-protein complexes provides foundation for research on protein interactions and complex formation. The knowledge of the conformations of the surface side chains is essential for modeling of protein complexes. The purpose of this study was to analyze and compare dihedral angle distribution functions of the side chains at the interface and non-interface areas in bound and unbound proteins.Results: To calculate the dihedral angle distribution functions, the configuration space was divided into grid cells. Statistical analysis showed that the similarity between bound and unbound interface and non-interface surface depends on the amino acid type and the grid resolution. The correlation coefficients between the distribution functions increased with the grid spacing increase for all amino acid types. The Manhattan distance showing the degree of dissimilarity between the distribution functions decreased accordingly. Short residues with one or two dihedral angles had higher correlations and smaller Manhattan distances than the longer residues. Met and Arg had the slowest growth of the correlation coefficient with the grid spacing increase. The correlations between the interface and non-interface distribution functions had a similar dependence on the grid resolution in both bound and unbound states. The interface and non-interface differences between bound and unbound distribution functions, caused by biological protein-protein interactions or crystal contacts, disappeared at the 70° grid spacing for interfaces and 30° for non-interface surface, which agrees with an average span of the side-chain rotamers. Conclusions: The two-fold difference in the critical grid spacing indicates larger conformational changes upon binding at the interface than at the rest of the surface. At the same time, transitions between rotamers induced by interactions across the interface or the crystal packing are rare, with most side chains having local readjustments that do not change the rotameric state. The analysis is important for better understanding of protein interactions and development of flexible docking approaches. [ABSTRACT FROM AUTHOR]

Abstract Background Protein-protein interactions are involved in most cellular processes, and their detailed physico-chemical and structural characterization is needed in order to understand their function at the molecular level. In-silico docking tools can complement experimental techniques, providing three-dimensional structural models of such interactions at atomic resolution. In several recent studies, protein structures have been modeled as networks (or graphs), where the nodes represent residues and the connecting edges their interactions. From such networks, it is possible to calculate different topology-based values for each of the nodes, and to identify protein regions with high centrality scores, which are known to positively correlate with key functional residues, hot spots, and protein-protein interfaces. Results Here we show that this correlation can be efficiently used for the scoring of rigid-body docking poses. When integrated into the pyDock energy-based docking method, the new combined scoring function significantly improved the results of the individual components as shown on a standard docking benchmark. This improvement was particularly remarkable for specific protein complexes, depending on the shape, size, type, or flexibility of the proteins involved. Conclusions The network-based representation of protein structures can be used to identify protein-protein binding regions and to efficiently score docking poses, complementing energy-based approaches.

Background: Structural modeling of protein-protein interactions produces a large number of putative configurations of the protein complexes. Identification of the near-native models among them is a serious challenge. Publicly available results of biomedical research may provide constraints on the binding mode, which can be essential for the docking. Our text-mining (TM) tool, which extracts binding site residues from the PubMed abstracts, was successfully applied to protein docking (Badal et al., PLoS Comput Biol, 2015; 11: e1004630). Still, many extracted residues were not relevant to the docking. Results: We present an extension of the TM tool, which utilizes natural language processing (NLP) for analyzing the context of the residue occurrence. The procedure was tested using generic and specialized dictionaries. The results showed that the keyword dictionaries designed for identification of protein interactions are not adequate for the TM prediction of the binding mode. However, our dictionary designed to distinguish keywords relevant to the protein binding sites led to considerable improvement in the TM performance. We investigated the utility of several methods of context analysis, based on dissection of the sentence parse trees. The machine learning-based NLP filtered the pool of the mined residues significantly more efficiently than the rule-based NLP. Constraints generated by NLP were tested in docking of unbound proteins from the DOCKGROUND X-ray benchmark set 4. The output of the global low-resolution docking scan was post-processed, separately, by constraints from the basic TM, constraints re-ranked by NLP, and the reference constraints. The quality of a match was assessed by the interface root-mean-square deviation. The results showed significant improvement of the docking output when using the constraints generated by the advanced TM with NLP. Conclusions: The basic TM procedure for extracting protein-protein binding site residues from the PubMed abstracts was significantly advanced by the deep parsing (NLP techniques for contextual analysis) in purging of the initial pool of the extracted residues. Benchmarking showed a substantial increase of the docking success rate based on the constraints generated by the advanced TM with NLP. [ABSTRACT FROM AUTHOR]