Protein-protein Binding Sites Research Articles

Whole exome sequencing identified a homozygous novel variant in DOP1A gene in the Pakistan family with neurodevelopmental disabilities: case report and literature review.

Hereditary neurodevelopmental disorders (NDDs) are prevalent in poorly prognostic pediatric diseases, but the pathogenesis of NDDs is still unclear. Irregular myelination could be one of the possible causes of NDDs. Here, whole exome sequencing was carried out for a consanguineous Pakistani family with NDDs to identify disease-associated variants. The co-segregation of candidate variants in the family was validated using Sanger sequencing. The potential impact of the gene on NDDs has been supported by conservation analysis, protein prediction, and expression analysis. A novel homozygous variant DOP1A(NM_001385863.1):c.2561A>G was identified. It was concluded that the missense variant might affect the protein-protein binding sites of the critical MEC interaction region of DOP1A, and DOP1A-MON2 may cause stability deficits in Golgi-endosome protein traffic. Proteolipid protein (PLP) and myelin-associate glycoprotein (MAG) could be targets of the DOP1A-MON2 Golgi-endosome traffic complex, especially during the fetal stage and the early developmental stages. This further supports the perspective that disorganized myelinogenesis due to congenital DOP1A deficiency might cause neurodevelopmental disorders (NDDs). Our case study revealed the potential pathway of myelinogenesis-relevant NDDs and identified DOP1A as a potential NDDs-relevant gene in humans.

Checkpoint Inhibitor PD-1H/VISTA Affects Myeloma Bone Disease By Osteoclast Cytoskeleton Remodeling

Introduction We have recently described the critical role of checkpoint inhibitor PD-1H/VISTA in osteoclast activation and multiple myeloma bone disease. Highly expressed in osteoclast precursors and mature osteoclasts, PD-1H functions as the receptor for osteoclastogenic factor MMP-13 and mediates MMP-13 induced osteoclast activation and myeloma bone disease ( Nat Commun14, 4271 (2023)). Despite the extensive functional study of PD-1H in immune disease models, the PD-1H interacting cellular complex is yet to be defined. Here we report how PD-1H interacts with cytoskeleton proteins contributing to osteoclast remodeling in multiple myeloma bone disease. Methods and Results To identify interacting proteins of PD-1H, PD-1H-His 6 recombinant protein was overexpressed in mouse mononuclear BMCs, and PD-1H-His 6 binding protein complex was pulled down by Ni-NTA agarose beads followed by mass spectrum analysis (Figure A). LIM and SH3 domain protein 1 (LASP1) was one of the most abundant cytoskeleton proteins identified in the PD-1H complex. We therefore first confirmed its binding to PD-1H by two-way co-IP assays (Figure B and C). Co-localization of LASP1 and PD-1H were further confirmed by immunofluorescence assay when co-expressed in HEK293 cells (Figure D). LASP1 is an F-actin-binding protein that stabilizes F-actin bundles and is involved in podosome regulation in macrophages (PLoS ONE 7(4): e35340). It has four domains: LIM, two nebulin-like repeats (N1 and N2) and SH3 domains. While nebulin-like repeats mediate binding to F-actin, SH3 domain is the major protein-protein binding site (PLoS ONE 7(4): e35340). In co-IP assay using PD-1H and domain deleted LASP1, deletion of LIM (△LIM) or either one or both nebulin-like repeats (△N1, △N2 and △N1/2) did not affect the binding to PD-1H. Instead, deletion of SH3 (△SH3) in LASP1 lead to loss of PD-1H binding, demonstrating that PD-1H binds to LASP1 at the SH3 domain (Figure E). Next, the association of LASP1 and PD-1H in the osteoclast cytoskeleton was demonstrated by confocal immunofluorescence microscopy. LASP1 co-localized with F-actin (Figure F) and PD-1H (Figure G) at the sealing ring in mature osteoclasts. Interestingly, although MMP-13 stimulation increased the size of osteoclasts and the sealing ring, MMP-13 did not change the co-localization pattern of LASP1 with PD-1H or F-actin. Conclusions Taken together, this study reveals the novel mechanism of PD-1H regulation of osteoclasts cytoskeleton reorganization, which is critical for osteoclast bone resorption activity. PD-1H directly associates with LASP1 at its C-terminal SH3 domain. Although LASP1 was reported to regulates podosomes function in macrophages, its role in osteoclast cytoskeletons remain largely obscure. The future study will be focused on delineating the role of LASP1 in osteoclast function, and its role in MMP-13/PD-1H signaling regulated myeloma bone disease.

A CNN-LSTM Ensemble Model for Predicting Protein-Protein Interaction Binding Sites.

Proteins commonly perform biological functions through protein-protein interactions (PPIs). The knowledge of PPI sites is imperative for the understanding of protein functions, disease mechanisms, and drug design. Traditional biological experimental methods for studying PPI sites still incur considerable drawbacks, including long experimental time and high labor costs. Therefore, many computational methods have been proposed for predicting PPI sites. However, achieving high prediction performance and overcoming severe data imbalance remain challenging issues. In this paper, we propose a new sequence-based deep learning model called CLPPIS (standing for CNN-LSTM ensemble based PPI Sites prediction). CLPPIS consists of CNN and LSTM components, which can capture spatial features and sequential features simultaneously. Further, it utilizes a novel feature group as input, which has 7 physicochemical, biophysical, and statistical properties. Besides, it adopts a batch-weighted loss function to reduce the interference of imbalance data. Our work suggests that the integration of protein spatial features and sequential features provides important information for PPI sites prediction. Evaluation on three public benchmark datasets shows that our CLPPIS model significantly outperforms existing state-of-the-art methods.

DeepBindPPI: Protein-Protein Binding Site Prediction Using Attention Based Graph Convolutional Network.

Due to the importance of protein-protein interactions in defence mechanism of living body, attempts were made to investigate its attributes, including, but not limited to, binding affinity, and binding region. Contemporary strategies for binding site prediction largely resort to deep learning techniques but turned out to be low precision models. As laboratory experiments for drug discovery tasks utilize this information, increased false positives devalue the computational methods. This emphasize the need to develop enhanced strategies. DeepBindPPI employs deep learning technique to predict the binding regions of proteins, particularly antigen-antibody interaction sites. The results obtained are applied in a docking environment to confirm their correctness. An integration of graph convolutional network with attention mechanism predicts interacting amino acids with improved precision. The model learns the determining factors in interaction from a general pool of proteins and is then fine-tuned using antigen-antibody data. Comparison of the proposed method with existing techniques shows that the developed model has comparable performance. The use of a separate spatial network clearly improved the precision of the proposed method from 0.4 to 0.5. An attempt to utilize the interface information for docking using the HDOCK server gives promising results, with high-quality structures appearing in the top10 ranks.

Learning the protein language of proteome-wide protein-protein binding sites via explainable ensemble deep learning

Protein-protein interactions (PPIs) govern cellular pathways and processes, by significantly influencing the functional expression of proteins. Therefore, accurate identification of protein-protein interaction binding sites has become a key step in the functional analysis of proteins. However, since most computational methods are designed based on biological features, there are no available protein language models to directly encode amino acid sequences into distributed vector representations to model their characteristics for protein-protein binding events. Moreover, the number of experimentally detected protein interaction sites is much smaller than that of protein-protein interactions or protein sites in protein complexes, resulting in unbalanced data sets that leave room for improvement in their performance. To address these problems, we develop an ensemble deep learning model (EDLM)-based protein-protein interaction (PPI) site identification method (EDLMPPI). Evaluation results show that EDLMPPI outperforms state-of-the-art techniques including several PPI site prediction models on three widely-used benchmark datasets including Dset_448, Dset_72, and Dset_164, which demonstrated that EDLMPPI is superior to those PPI site prediction models by nearly 10% in terms of average precision. In addition, the biological and interpretable analyses provide new insights into protein binding site identification and characterization mechanisms from different perspectives. The EDLMPPI webserver is available at http://www.edlmppi.top:5002/.

Prediction of protein-protein interaction sites in intrinsically disordered proteins.

Intrinsically disordered proteins (IDPs) participate in many biological processes by interacting with other proteins, including the regulation of transcription, translation, and the cell cycle. With the increasing amount of disorder sequence data available, it is thus crucial to identify the IDP binding sites for functional annotation of these proteins. Over the decades, many computational approaches have been developed to predict protein-protein binding sites of IDP (IDP-PPIS) based on protein sequence information. Moreover, there are new IDP-PPIS predictors developed every year with the rapid development of artificial intelligence. It is thus necessary to provide an up-to-date overview of these methods in this field. In this paper, we collected 30 representative predictors published recently and summarized the databases, features and algorithms. We described the procedure how the features were generated based on public data and used for the prediction of IDP-PPIS, along with the methods to generate the feature representations. All the predictors were divided into three categories: scoring functions, machine learning-based prediction, and consensus approaches. For each category, we described the details of algorithms and their performances. Hopefully, our manuscript will not only provide a full picture of the status quo of IDP binding prediction, but also a guide for selecting different methods. More importantly, it will shed light on the inspirations for future development trends and principles.

In silico investigation of nonsynonymous single nucleotide polymorphisms in BCL2 apoptosis regulator gene to design novel protein-based drugs against cancer.

BCL2 apoptosis regulator gene encodes Bcl-2 pro-survival protein, which plays an important role to evade apoptosis in various cancers. Moreover, single nucleotide polymorphisms (SNPs) in the BCL2 gene can be nonsynonymous (nsSNPs), which might affect the protein stability and probably its function. Therefore, we implement cutting-edge computational techniques based on the Spherical Polar Fourier and Monte-Carlo algorithms to investigate the impact of these SNPs on the B cell lymphoma-2 (Bcl-2) stability and therapeutic potential of protein-based molecules to inhibit this protein. As a result, we identified two nsSNPs (Q118R and R129C) to be deleterious and highly conserved, having a negative effect on protein stability. Additionally, molecular docking and molecular dynamics simulations confirmed the decreased binding affinity of mutated Bcl-2 variants to bind three-helix bundle protein inhibitor as these mutations occurred in the protein-protein binding site. Overall, this computational approach investigating nsSNPs provides a useful basis for designing novel molecules to inhibit Bcl-2 pro-survival pathway in malignant cells.

ScanNet: an interpretable geometric deep learning model for structure-based protein binding site prediction.

Predicting the functional sites of a protein from its structure, such as the binding sites of small molecules, other proteins or antibodies, sheds light on its function in vivo. Currently, two classes of methods prevail: machine learning models built on top of handcrafted features and comparative modeling. They are, respectively, limited by the expressivity of the handcrafted features and the availability of similar proteins. Here, we introduce ScanNet, an end-to-end, interpretable geometric deep learning model that learns features directly from 3D structures. ScanNet builds representations of atoms and amino acids based on the spatio-chemical arrangement of their neighbors. We train ScanNet for detecting protein-protein and protein-antibody binding sites, demonstrate its accuracy-including for unseen protein folds-and interpret the filters learned. Finally, we predict epitopes of the SARS-CoV-2 spike protein, validating known antigenic regions and predicting previously uncharacterized ones. Overall, ScanNet is a versatile, powerful and interpretable model suitable for functional site prediction tasks. A webserver for ScanNet is available from http://bioinfo3d.cs.tau.ac.il/ScanNet/ .

CSM-Potential: mapping protein interactions and biological ligands in 3D space using geometric deep learning.

Recent advances in protein structural modelling have enabled the accurate prediction of the holo 3D structures of almost any protein, however protein function is intrinsically linked to the interactions it makes. While a number of computational approaches have been proposed to explore potential biological interactions, they have been limited to specific interactions, and have not been readily accessible for non-experts or use in bioinformatics pipelines. Here we present CSM-Potential, a geometric deep learning approach to identify regions of a protein surface that are likely to mediate protein-protein and protein–ligand interactions in order to provide a link between 3D structure and biological function. Our method has shown robust performance, outperforming existing methods for both predictive tasks. By assessing the performance of CSM-Potential on independent blind tests, we show that our method was able to achieve ROC AUC values of up to 0.81 for the identification of potential protein-protein binding sites, and up to 0.96 accuracy on biological ligand classification. Our method is freely available as a user-friendly and easy-to-use web server and API at http://biosig.unimelb.edu.au/csm_potential.

77Se-13C based dipolar correlation experiments to map selenium sites in microcrystalline proteins.

Sulfur-containing sites in proteins are of great importance for both protein structure and function, including enzymatic catalysis, signaling pathways, and recognition of ligands and protein partners. Selenium-77 is an NMR active spin-1/2 nucleus that shares many physiochemical properties with sulfur and can be readily introduced into proteins at sulfur sites without significant perturbations to the protein structure. The sulfur-containing amino acid methionine is commonly found at protein-protein or protein-ligand binding sites. Its selenium-containing counterpart, selenomethionine, has a broad chemical shift dispersion useful for NMR-based studies of complex systems. Methods such as (1H)-77Se-13C double cross polarization or {77Se}-13C REDOR could be valuable to map the local environment around selenium sites in proteins but have not been demonstrated to date. In this work, we explore these dipolar transfer mechanisms for structural characterization of the GB1 V39SeM variant of the model protein GB1 and demonstrate that 77Se-13C based correlations can be used to map the local environment around selenium sites in proteins. We have found that the general detection limit is ~ 5Å, but longer range distances up to ~ 7Å can be observed as well. This study establishes a framework for the future characterization of selenium sites at protein-protein or protein-ligand binding interfaces.

Structure of Enterohemorrhagic Escherichia coli O157:H7 Intimin Virulence Factor Bound to Nanobodies

Enterohemorrhagic E. coli O157:H7 (EHEC) is a food and water-born pathogen that presents a significant risk to human health worldwide. EHEC naturally resides in the intestinal tract of cattle as normal flora but can be transmitted to humans when contaminated food is ingested. Once EHEC invades and colonizes the human intestinal tract, it secretes Shiga-like toxins causing serious and potentially fatal gastrointestinal diseases. Unlike most foodborne bacterial diseases, antibiotics are not recommended for treatment of EHEC as antibiotics can induce enterotoxin release, increasing the risk of hemolytic uremic syndrome. There is currently no effective treatment or vaccine against enterohemorrhagic E. coli. EHEC pathogenicity involves intimin, a virulence factor which mediates bacterial colonization of the host GI tract. Due to its critical role in EHEC pathogenesis, intimin has attracted attention as an antibacterial drug target. Inhibiting the interaction of intimin with its cognate bacterial secreted receptor Tir could neutralize bacterial colonization of the gastrointestinal tract. One novel solution for treating EHEC infection comes from a unique class of antibodies known as nanobodies. Nanobodies are the antigen binding domain of heavy chain antibodies produced by the Camelid family. Intimin specific nanobodies were generated by immunizing a llama followed by phage display and selection for high affinity intimin binding nanobodies. The resultant intimin-specific nanobodies neutralized EHEC in vitro. We hypothesize that nanobodies specifically bind the Tir-binding domain of intimin in order to neutralizing EHEC infection. The goal of this research project is to determine the complex crystal structure of the nanobodies bound to intimin. The structure of nanobodies-intimin would show their protein-protein interaction binding site revealing key amino acids involved in binding. These residues could also be involved in the binding of intimin to its natural ligand, Tir. Thus, revealing such residues could form the bases for identifying and developing novel therapeutics and biosensors for treating and preventing EHEC pathogenesis.We show through size exclusion chromatography that nanobodies bind intimin and form stable complexes. We have set up crystal trails of five nanobodies (Int1, Int2, Int3, Int4, and IntN1) complexed to intimin. Two of the five complexed proteins, Int2-intimin and Int3-intimn formed crystals. The crystal condition for the two protein complexes Int2 and Int3 bound to intimin were optimized by varying the concentration of precipitants and pH range. Crystals of the protein complexes were sent to the Canadian Light Source for X-ray diffraction data collection. Diffraction measurements of Int2-intimin and Int3-intimin complexes were collected and processed in space groups P1 and P222, respectively. Currently, we are analyzing the structure by molecular replacement.

Drebrin regulates cytoskeleton dynamics in migrating neurons through interaction with CXCR4

Stromal cell-derived factor-1 (SDF-1) and chemokine receptor type 4 (CXCR4) are regulators of neuronal migration (e.g., GnRH neurons, cortical neurons, and hippocampal granule cells). However, how SDF-1/CXCR4 alters cytoskeletal components remains unclear. Developmentally regulated brain protein (drebrin) stabilizes actin polymerization, interacts with microtubule plus ends, and has been proposed to directly interact with CXCR4 in T cells. The current study examined, in mice, whether CXCR4 under SDF-1 stimulation interacts with drebrin to facilitate neuronal migration. Bioinformatic prediction of protein-protein interaction highlighted binding sites between drebrin and crystallized CXCR4. In migrating GnRH neurons, drebrin, CXCR4, and the microtubule plus-end binding protein EB1 were localized close to the cell membrane. Coimmunoprecipitation (co-IP) confirmed a direct interaction between drebrin and CXCR4 using wild-type E14.5 whole head and a GnRH cell line. Analysis of drebrin knockout (DBN1 KO) mice showed delayed migration of GnRH cells into the brain. A decrease in hippocampal granule cells was also detected, and co-IP confirmed a direct interaction between drebrin and CXCR4 in PN4 hippocampi. Migration assays on primary neurons established that inhibiting drebrin (either pharmacologically or using cells from DBN1 KO mice) prevented the effects of SDF-1 on neuronal movement. Bioinformatic prediction then identified binding sites between drebrin and the microtubule plus end protein, EB1, and super-resolution microscopy revealed decreased EB1 and drebrin coexpression after drebrin inhibition. Together, these data show a mechanism by which a chemokine, via a membrane receptor, communicates with the intracellular cytoskeleton in migrating neurons during central nervous system development.

Functional characterization of an unknown soybean intrinsically disordered protein in vitro and in Escherichia coli

Intrinsically disordered proteins (IDPs) possess a wide range of biological function in all organisms, however the specific functions of most IDPs are still unknown. Soybean LOC protein, LOC for short, is a heat-stable protein, which is more abundant in the stress-resistant radicles. Sequence alignment and phylogenetic analysis showed that LOC is a functionally unknown protein and conserved in Fabaceae. LOC, being enriched in most disorder-promoting residues and depleted in most order-promoting residues, was predicted to contain high levels of intrinsic disorder by several commonly used computational tools. However, it was also predicted to contain two disorder-based protein-protein binding sites and two short α-helical segments. The circular dichroism spectroscopic analysis showed that this protein is mostly disordered in water, but can form more α-helical structure in the presence of SDS and TFE. Functional in vitro studies showed that the LOC protein is able to prevent lactate dehydrogenase inactivation by freeze-thaw at a molar ratio of 10:1. Furthermore, in vivo analyses revealed the survival rate of Escherichia coli over-expressing LOC protein under the conditions of osmotic stress was noticeably increased in comparison with the control. These observations suggest that the intrinsically disordered protein LOC might serve as a chaperone and/or cell protector.

Abstract 5267: Unbiased metabolic profiling of atypical teratoid/rhabdoid tumors predicts sensitivity to the glutamine metabolic inhibitor 6-diazo-5-oxo-L-norleucine

Abstract Atypical teratoid rhabdoid tumors (AT/RT) are deadly tumors of infancy in need of new, targeted therapies. Molecular analyses have revealed 3 epigenetically distinct subgroups of AT/RT. High MYC-expressing tumors are a particularly aggressive subgroup, with a 5-year median survival of 18.5%. MYC is difficult to directly target given large, flat, featureless protein-protein binding sites. We performed the first unbiased metabolic profile of AT/RT cell models by high-performance liquid chromatography-mass spectrometry. This study revealed that high MYC-expressing AT/RT have a unique metabolic profile (Partial Least Squares-Discriminant Analysis). Pathway analysis highlights a dependence on glutamine metabolism in high-MYC expressing AT/RT. High MYC-expressing cell lines grow poorly in glutamine-free media compared to low-MYC cell lines (MTS assay, glutamine-free media vs normal media). Due to this dependence on glutamine metabolism, we hypothesized that high-MYC expressing AT/RT would be especially sensitive to glutamine metabolic inhibitors. We show that the glutamine analogue, 6-diazo-5-oxo-L-norleucine (DON) slows high-MYC expressing AT/RT cell growth, induces high rates of apoptosis, and improves survival in orthotopic mouse models of high-MYC expressing AT/RT (p =0.0027 by log-rank test). In contrast, low-MYC expressing models are relatively insensitive to DON. In uniformly labeled glutamine metabolic analyses of AT/RT cells, DON treatment led to depletion of glutathione levels by preventing the production of glutamate. DON treatment synergized with carboplatin to kill AT/RT cells in vitro and extend the life of mice bearing AT/RT orthotopic xenografts (p< 0.001 by log rank test). We aim to translate these findings into a new clinical trial to improve survival in this deadly disease. Citation Format: Sabrina Z. Wang, Brad Poore, Jesse Alt, Antoinette Price, Sariah Allen, Brent Orr, Rana Rais, Barbara Slusher, Charles Eberhart, Eric H. Raabe, Jeffrey H. Rubens. Unbiased metabolic profiling of atypical teratoid/rhabdoid tumors predicts sensitivity to the glutamine metabolic inhibitor 6-diazo-5-oxo-L-norleucine [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 5267.

Natural language processing in text mining for structural modeling of protein complexes

BackgroundStructural 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.ResultsWe 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.ConclusionsThe 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.

A Cylindrical Assembly Model and Dynamics of the Ebola Virus VP40

The Ebola virus, a member of the filovirus family, causes a severe hemorrhagic fever with a high fatality rate in humans. Of the seven proteins that are encoded by the Ebola virus, VP40 is the primary matrix protein and exists in different conformational and oligomeric states. VP40 plays crucial roles in viral assembly and budding at the plasma membrane of the infected cells. This matrix protein is capable of forming virus-like particles without the need for other Ebola proteins. The budding and formation of new virus-like particles requires VP40 hexamers to arrange side-by-side into a cylinder, as well as end-to-end to extend the cylinder to a long filament. However, no experimental three-dimensional structure for any filovirus VP40 cylindrical assembly matrix is currently available. Here, we developed cylindrical assembly models of 42 nm and 92 nm diameters with varying numbers of VP40 hexamers using a protein-protein docking approach. These models match well with the 2D averages of cryo-electron tomograms of authentic virions. In addition, our models form a multilayered matrix that is in good agreement with the experimental observations. Using a pair of hexamer from the models, we studied the energetics of the interactions between the hexamers for different side-by-side arrangements, which indicated the most important residues in the protein-protein binding site responsible for the cylindrical matrix assembly. We also investigated the stability and dynamics of the cylindrical hexamer models using implicit all-atom molecular dynamics simulations, which revealed the flexibility of the cylinders. Our models will provide helpful information to better understand the assembly processes of filoviruses and such structural studies could also lead to the design and development of antiviral drugs.

GenProBiS: web server for mapping of sequence variants to protein binding sites.

Discovery of potentially deleterious sequence variants is important and has wide implications for research and generation of new hypotheses in human and veterinary medicine, and drug discovery. The GenProBiS web server maps sequence variants to protein structures from the Protein Data Bank (PDB), and further to protein–protein, protein–nucleic acid, protein–compound, and protein–metal ion binding sites. The concept of a protein–compound binding site is understood in the broadest sense, which includes glycosylation and other post-translational modification sites. Binding sites were defined by local structural comparisons of whole protein structures using the Protein Binding Sites (ProBiS) algorithm and transposition of ligands from the similar binding sites found to the query protein using the ProBiS-ligands approach with new improvements introduced in GenProBiS. Binding site surfaces were generated as three-dimensional grids encompassing the space occupied by predicted ligands. The server allows intuitive visual exploration of comprehensively mapped variants, such as human somatic mis-sense mutations related to cancer and non-synonymous single nucleotide polymorphisms from 21 species, within the predicted binding sites regions for about 80 000 PDB protein structures using fast WebGL graphics. The GenProBiS web server is open and free to all users at http://genprobis.insilab.org.

IPPBS-Opt: A Sequence-Based Ensemble Classifier for Identifying Protein-Protein Binding Sites by Optimizing Imbalanced Training Datasets.

Knowledge of protein-protein interactions and their binding sites is indispensable for in-depth understanding of the networks in living cells. With the avalanche of protein sequences generated in the postgenomic age, it is critical to develop computational methods for identifying in a timely fashion the protein-protein binding sites (PPBSs) based on the sequence information alone because the information obtained by this way can be used for both biomedical research and drug development. To address such a challenge, we have proposed a new predictor, called iPPBS-Opt, in which we have used: (1) the K-Nearest Neighbors Cleaning (KNNC) and Inserting Hypothetical Training Samples (IHTS) treatments to optimize the training dataset; (2) the ensemble voting approach to select the most relevant features; and (3) the stationary wavelet transform to formulate the statistical samples. Cross-validation tests by targeting the experiment-confirmed results have demonstrated that the new predictor is very promising, implying that the aforementioned practices are indeed very effective. Particularly, the approach of using the wavelets to express protein/peptide sequences might be the key in grasping the problem’s essence, fully consistent with the findings that many important biological functions of proteins can be elucidated with their low-frequency internal motions. To maximize the convenience of most experimental scientists, we have provided a step-by-step guide on how to use the predictor’s web server (http://www.jci-bioinfo.cn/iPPBS-Opt) to get the desired results without the need to go through the complicated mathematical equations involved.

Identification of protein-protein binding sites by incorporating the physicochemical properties and stationary wavelet transforms into pseudo amino acid composition

With the explosive growth of protein sequences entering into protein data banks in the post-genomic era, it is highly demanded to develop automated methods for rapidly and effectively identifying the protein–protein binding sites (PPBSs) based on the sequence information alone. To address this problem, we proposed a predictor called iPPBS-PseAAC, in which each amino acid residue site of the proteins concerned was treated as a 15-tuple peptide segment generated by sliding a window along the protein chains with its center aligned with the target residue. The working peptide segment is further formulated by a general form of pseudo amino acid composition via the following procedures: (1) it is converted into a numerical series via the physicochemical properties of amino acids; (2) the numerical series is subsequently converted into a 20-D feature vector by means of the stationary wavelet transform technique. Formed by many individual “Random Forest” classifiers, the operation engine to run prediction is a two-layer ensemble classifier, with the 1st-layer voting out the best training data-set from many bootstrap systems and the 2nd-layer voting out the most relevant one from seven physicochemical properties. Cross-validation tests indicate that the new predictor is very promising, meaning that many important key features, which are deeply hidden in complicated protein sequences, can be extracted via the wavelets transform approach, quite consistent with the facts that many important biological functions of proteins can be elucidated with their low-frequency internal motions. The web server of iPPBS-PseAAC is accessible at http://www.jci-bioinfo.cn/iPPBS-PseAAC, by which users can easily acquire their desired results without the need to follow the complicated mathematical equations involved.

Structural neighboring property for identifying protein-protein binding sites.

BackgroundThe protein-protein interaction plays a key role in the control of many biological functions, such as drug design and functional analysis. Determination of binding sites is widely applied in molecular biology research. Therefore, many efficient methods have been developed for identifying binding sites. In this paper, we calculate structural neighboring property through Voronoi diagram. Using 6,438 complexes, we study local biases of structural neighboring property on interface.ResultsWe propose a novel statistical method to extract interacting residues, and interacting patches can be clustered as predicted interface residues. In addition, structural neighboring property can be adopted to construct a new energy function, for evaluating docking solutions. It includes new statistical property as well as existing energy items. Comparing to existing methods, our approach improves overall Fnat value by at least 3%. On Benchmark v4.0, our method has average Irmsd value of 3.31Å and overall Fnat value of 63%, which improves upon Irmsd of 3.89 Å and Fnat of 49% for ZRANK, and Irmsd of 3.99Å and Fnat of 46% for ClusPro. On the CAPRI targets, our method has average Irmsd value of 3.46 Å and overall Fnat value of 45%, which improves upon Irmsd of 4.18 Å and Fnat of 40% for ZRANK, and Irmsd of 5.12 Å and Fnat of 32% for ClusPro.ConclusionsExperiments show that our method achieves better results than some state-of-the-art methods for identifying protein-protein binding sites, with the prediction quality improved in terms of CAPRI evaluation criteria.