Pushing structural information into the yeast interactome by high-throughput protein docking experiments

Structure-based prediction of protein-protein interaction network in rice

Comprehensive protein-protein interaction (PPI) maps are critical for understanding the functional organization of the proteome, but challenging to produce experimentally. Here, we developed a computational method for predicting PPIs based on protein docking. Evaluation of performance on benchmark sets demonstrated the ability of the docking-based method to accurately identify PPIs using predicted protein structures. By employing the docking-based method, we constructed a structurally resolved PPI network consisting of 24,653 interactions between 2,131 proteins, which greatly extends the current knowledge on the rice protein-protein interactome. Moreover, we mapped the trait-associated single nucleotide polymorphisms (SNPs) to the structural interactome, and computationally identified 14 SNPs that had significant consequences on PPI network. The protein structural interactome map provided a resource to facilitate functional investigation of PPI-perturbing alleles associated with agronomically important traits in rice.

Improving classification of correct and incorrect protein-protein docking models by augmenting the training set

Motivation

Protein-protein interactions drive many relevant biological events, such as infection, replication and recognition. To control or engineer such events, we need to access the molecular details of the interaction provided by experimental 3D structures. However, such experiments take time and are expensive; moreover, the current technology cannot keep up with the high discovery rate of new interactions. Computational modeling, like protein-protein docking, can help to fill this gap by generating docking poses. Protein-protein docking generally consists of two parts, sampling and scoring. The sampling is an exhaustive search of the tridimensional space. The caveat of the sampling is that it generates a large number of incorrect poses, producing a highly unbalanced dataset. This limits the utility of the data to train machine learning classifiers.

Results

Using weak supervision, we developed a data augmentation method that we named hAIkal. Using hAIkal, we increased the labeled training data to train several algorithms. We trained and obtained different classifiers; the best classifier has 81% accuracy and 0.51 Matthews' correlation coefficient on the test set, surpassing the state-of-the-art scoring functions.

Availability and implementation

Docking models from Benchmark 5 are available at https://doi.org/10.5281/zenodo.4012018. Processed tabular data are available at https://repository.kaust.edu.sa/handle/10754/666961. Google colab is available at https://colab.research.google.com/drive/1vbVrJcQSf6\_C3jOAmZzgQbTpuJ5zC1RP?usp=sharing.

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

In silico identification of RBD subdomain of spike protein from Pro322-Thr581 for applications in vaccine development against SARS-CoV2

The three-dimensional hybrid structures of coronavirus spike proteins including the C-terminal sequence and receptor binding motif (RBM) was remodeled and energy minimized. Further, protein-protein docking show that Receptor Binding Domain (RBD) of SARS-CoV 2 Lys⁴⁵⁷-Pro⁴⁹⁰ bind on the surface of ACE2 receptor near N-terminal helices to form host-pathogen attachment. In this binding interface, SARS-CoV 2 shows a tight network of hydrogen bonds than other spike proteins from BtRsRaTG13-CoV, SARS-CoV, BtRsBeta-CoV, BtRsCoV-related, Pangolin-CoV (PCoV), human-CoV (hCoV), MERS-CoV (MCoV), Avian-CoV (ACoV) and PEDV1-CoV. Further studies show that subdomains from SARS-CoV 2 RBD Pro³²²-Thr⁵⁸¹, SARS-CoV RBD Pro³⁰⁹-Pro⁵⁷⁵, BtRsRaTG13 RBD Thr⁵⁸¹-Thr³²³, BtRsBeta-CoV RBD Ser³¹¹-Thr⁵⁶⁸, BtRsCoV-related Arg³⁰⁶-Pro⁵⁷⁵ and PCoV RBD Gln³¹⁹-Ser⁵⁸⁹ show binding conformations with ACE2 like their full-length structures of spike proteins. In addition, the subdomains MCoV RBD Gly³⁷²-Val⁶¹⁶, ACoV RBD Gly³⁷²-Val⁶¹⁶ and PEDV1-CoV RBD Ala³¹⁵-Tyr⁶⁷⁵ also binds on the surface of ACE2 similar to their full-length spike proteins. The B-Cell epitope mapping also identified main antigenic determinants predicting that these nine subdomains are highly useful in recombinant vaccine development in inducing cross neutralizing antibodies against SARS-CoV 2 spike protein and inhibits its attachment with ACE2.

Advancing from protein interactomes and gene co-expression networks towards multi-omics-based composite networks: approaches for predicting and extracting biological knowledge

Prediction of biological interaction networks from single-omics data has been extensively implemented to understand various aspects of biological systems. However, more recently, there is a growing interest in integrating multi-omics datasets for the prediction of interactomes that provide a global view of biological systems with higher descriptive capability, as compared to single omics. In this review, we have discussed various computational approaches implemented to infer and analyze two of the most important and well studied interactomes: protein-protein interaction networks and gene co-expression networks. We have explicitly focused on recent methods and pipelines implemented to infer and extract biologically important information from these interactomes, starting from utilizing single-omics data and then progressing towards multi-omics data. Accordingly, recent examples and case studies are also briefly discussed. Overall, this review will provide a proper understanding of the latest developments in protein and gene network modelling and will also help in extracting practical knowledge from them.

Proteome-wide, Structure-Based Prediction of Protein-Protein Interactions/New Molecular Interactions Viewer

Determining the complete Arabidopsis (Arabidopsis thaliana) protein-protein interaction network is essential for understanding the functional organization of the proteome. Numerous small-scale studies and a couple of large-scale ones have elucidated a fraction of the estimated 300,000 binary protein-protein interactions in Arabidopsis. In this study, we provide evidence that a docking algorithm has the ability to identify real interactions using both experimentally determined and predicted protein structures. We ranked 0.91 million interactions generated by all possible pairwise combinations of 1,346 predicted structure models from an Arabidopsis predicted "structure-ome" and found a significant enrichment of real interactions for the top-ranking predicted interactions, as shown by cosubcellular enrichment analysis and yeast two-hybrid validation. Our success rate for computationally predicted, structure-based interactions was 63% of the success rate for published interactions naively tested using the yeast two-hybrid system and 2.7 times better than for randomly picked pairs of proteins. This study provides another perspective in interactome exploration and biological network reconstruction using protein structural information. We have made these interactions freely accessible through an improved Arabidopsis Interactions Viewer and have created community tools for accessing these and ∼2.8 million other protein-protein and protein-DNA interactions for hypothesis generation by researchers worldwide. The Arabidopsis Interactions Viewer is freely available at http://bar.utoronto.ca/interactions2/.

Computational Opportunities and Challenges in Finding Cyclic Peptide Modulators of Protein-Protein Interactions

Peptide cyclization can improve stability, conformational constraint, and compactness. However, apart from beta-turn structures, which are well incorporated into cyclic peptides (CPs), many primary peptide structures and functions are markedly altered by cyclization. Accordingly, to mimic linear peptide interfaces with cyclic peptides, it can be beneficial to screen combinatorial cyclic peptide libraries. Computational methods have been developed to screen CPs, but face a number of challenges. Here, we review methods to develop in silico computational libraries, and the potential for screening naturally occurring libraries of CPs. The simplest and most rapid computational pharmacophore methods that estimate peptide three-dimensional structures to be screened versus targets are relatively easy to implement, and while the constraint on structure imposed by cyclization makes them more effective than the same approaches with linear peptides, there are a large number of limiting assumptions. In contrast, full molecular dynamics simulations of cyclic peptide structures not only are costly to implement, but also require careful attention to interpretation, so that not only is the computation time rate limiting, but the interpretation time is also rate limiting due to the analysis of the typically complex underlying conformational space of CPs. A challenge for the field of computational cyclic peptide screening is to bridge this gap effectively. Natural compound libraries of short cyclic peptides, and short cyclized regions of proteins, encoded in the genomes of many organisms present a potential treasure trove of novel functionality which may be screened via combined computational and experimental screening approaches.

Modeling and simulating networks of interdependent protein interactions

Protein interactions are fundamental building blocks of biochemical reaction systems underlying cellular functions. The complexity and functionality of these systems emerge not only from the protein interactions themselves but also from the dependencies between these interactions, as generated by allosteric effects or mutual exclusion due to steric hindrance. Therefore, formal models for integrating and utilizing information about interaction dependencies are of high interest. Here, we describe an approach for endowing protein networks with interaction dependencies using propositional logic, thereby obtaining constrained protein interaction networks ("constrained networks"). The construction of these networks is based on public interaction databases as well as text-mined information about interaction dependencies. We present an efficient data structure and algorithm to simulate protein complex formation in constrained networks. The efficiency of the model allows fast simulation and facilitates the analysis of many proteins in large networks. In addition, this approach enables the simulation of perturbation effects, such as knockout of single or multiple proteins and changes of protein concentrations. We illustrate how our model can be used to analyze a constrained human adhesome protein network, which is responsible for the formation of diverse and dynamic cell-matrix adhesion sites. By comparing protein complex formation under known interaction dependencies versus without dependencies, we investigate how these dependencies shape the resulting repertoire of protein complexes. Furthermore, our model enables investigating how the interplay of network topology with interaction dependencies influences the propagation of perturbation effects across a large biochemical system. Our simulation software CPINSim (for Constrained Protein Interaction Network Simulator) is available under the MIT license at http://github.com/BiancaStoecker/cpinsim and as a Bioconda package (https://bioconda.github.io).

Prediction of Protein-Protein Interactions

The authors provide an overview of physical protein-protein interaction prediction, covering the main strategies for predicting interactions, approaches for assessing predictions, and online resources for accessing predictions. This unit focuses on the main advancements in each of these areas over the last decade. The methods and resources that are presented here are not an exhaustive set, but characterize the current state of the field-highlighting key challenges and achievements. © 2017 by John Wiley & Sons, Inc.

Docking-based modeling of protein-protein interfaces for extensive structural and functional characterization of missense mutations

Next-generation sequencing (NGS) technologies are providing genomic information for an increasing number of healthy individuals and patient populations. In the context of the large amount of generated genomic data that is being generated, understanding the effect of disease-related mutations at molecular level can contribute to close the gap between genotype and phenotype and thus improve prevention, diagnosis or treatment of a pathological condition. In order to fully characterize the effect of a pathological mutation and have useful information for prediction purposes, it is important first to identify whether the mutation is located at a protein-binding interface, and second to understand the effect on the binding affinity of the affected interaction/s. Computational methods, such as protein docking are currently used to complement experimental efforts and could help to build the human structural interactome. Here we have extended the original pyDockNIP method to predict the location of disease-associated nsSNPs at protein-protein interfaces, when there is no available structure for the protein-protein complex. We have applied this approach to the pathological interaction networks of six diseases with low structural data on PPIs. This approach can almost double the number of nsSNPs that can be characterized and identify edgetic effects in many nsSNPs that were previously unknown. This can help to annotate and interpret genomic data from large-scale population studies, and to achieve a better understanding of disease at molecular level.

A benchmark testing ground for integrating homology modeling and protein docking

Protein docking procedures carry out the task of predicting the structure of a protein-protein complex starting from the known structures of the individual protein components. More often than not, however, the structure of one or both components is not known, but can be derived by homology modeling on the basis of known structures of related proteins deposited in the Protein Data Bank (PDB). Thus, the problem is to develop methods that optimally integrate homology modeling and docking with the goal of predicting the structure of a complex directly from the amino acid sequences of its component proteins. One possibility is to use the best available homology modeling and docking methods. However, the models built for the individual subunits often differ to a significant degree from the bound conformation in the complex, often much more so than the differences observed between free and bound structures of the same protein, and therefore additional conformational adjustments, both at the backbone and side chain levels need to be modeled to achieve an accurate docking prediction. In particular, even homology models of overall good accuracy frequently include localized errors that unfavorably impact docking results. The predicted reliability of the different regions in the model can also serve as a useful input for the docking calculations. Here we present a benchmark dataset that should help to explore and solve combined modeling and docking problems. This dataset comprises a subset of the experimentally solved 'target' complexes from the widely used Docking Benchmark from the Weng Lab (excluding antibody-antigen complexes). This subset is extended to include the structures from the PDB related to those of the individual components of each complex, and hence represent potential templates for investigating and benchmarking integrated homology modeling and docking approaches. Template sets can be dynamically customized by specifying ranges in sequence similarity and in PDB release dates, or using other filtering options, such as excluding sets of specific structures from the template list. Multiple sequence alignments, as well as structural alignments of the templates to their corresponding subunits in the target are also provided. The resource is accessible online or can be downloaded at http://cluspro.org/benchmark, and is updated on a weekly basis in synchrony with new PDB releases. Proteins 2016; 85:10-16. © 2016 Wiley Periodicals, Inc.

Challenges in structural approaches to cell modeling

Computational modeling is essential for structural characterization of biomolecular mechanisms across the broad spectrum of scales. Adequate understanding of biomolecular mechanisms inherently involves our ability to model them. Structural modeling of individual biomolecules and their interactions has been rapidly progressing. However, in terms of the broader picture, the focus is shifting toward larger systems, up to the level of a cell. Such modeling involves a more dynamic and realistic representation of the interactomes in vivo, in a crowded cellular environment, as well as membranes and membrane proteins, and other cellular components. Structural modeling of a cell complements computational approaches to cellular mechanisms based on differential equations, graph models, and other techniques to model biological networks, imaging data, etc. Structural modeling along with other computational and experimental approaches will provide a fundamental understanding of life at the molecular level and lead to important applications to biology and medicine. A cross section of diverse approaches presented in this review illustrates the developing shift from the structural modeling of individual molecules to that of cell biology. Studies in several related areas are covered: biological networks; automated construction of three-dimensional cell models using experimental data; modeling of protein complexes; prediction of non-specific and transient protein interactions; thermodynamic and kinetic effects of crowding; cellular membrane modeling; and modeling of chromosomes. The review presents an expert opinion on the current state-of-the-art in these various aspects of structural modeling in cellular biology, and the prospects of future developments in this emerging field.

Rigid-Docking Approaches to Explore Protein-Protein Interaction Space

Protein-protein interactions play core roles in living cells, especially in the regulatory systems. As information on proteins has rapidly accumulated on publicly available databases, much effort has been made to obtain a better picture of protein-protein interaction networks using protein tertiary structure data. Predicting relevant interacting partners from their tertiary structure is a challenging task and computer science methods have the potential to assist with this. Protein-protein rigid docking has been utilized by several projects, docking-based approaches having the advantages that they can suggest binding poses of predicted binding partners which would help in understanding the interaction mechanisms and that comparing docking results of both non-binders and binders can lead to understanding the specificity of protein-protein interactions from structural viewpoints. In this review we focus on explaining current computational prediction methods to predict pairwise direct protein-protein interactions that form protein complexes.

Protein-protein docking: from interaction to interactome

The protein-protein docking problem is one of the focal points of activity in computational biophysics and structural biology. The three-dimensional structure of a protein-protein complex, generally, is more difficult to determine experimentally than the structure of an individual protein. Adequate computational techniques to model protein interactions are important because of the growing number of known protein structures, particularly in the context of structural genomics. Docking offers tools for fundamental studies of protein interactions and provides a structural basis for drug design. Protein-protein docking is the prediction of the structure of the complex, given the structures of the individual proteins. In the heart of the docking methodology is the notion of steric and physicochemical complementarity at the protein-protein interface. Originally, mostly high-resolution, experimentally determined (primarily by x-ray crystallography) protein structures were considered for docking. However, more recently, the focus has been shifting toward lower-resolution modeled structures. Docking approaches have to deal with the conformational changes between unbound and bound structures, as well as the inaccuracies of the interacting modeled structures, often in a high-throughput mode needed for modeling of large networks of protein interactions. The growing number of docking developers is engaged in the community-wide assessments of predictive methodologies. The development of more powerful and adequate docking approaches is facilitated by rapidly expanding information and data resources, growing computational capabilities, and a deeper understanding of the fundamental principles of protein interactions.

Surfing the Protein-Protein Interaction Surface Using Docking Methods: Application to the Design of PPI Inhibitors

Blocking protein-protein interactions (PPI) using small molecules or peptides modulates biochemical pathways and has therapeutic significance. PPI inhibition for designing drug-like molecules is a new area that has been explored extensively during the last decade. Considering the number of available PPI inhibitor databases and the limited number of 3D structures available for proteins, docking and scoring methods play a major role in designing PPI inhibitors as well as stabilizers. Docking methods are used in the design of PPI inhibitors at several stages of finding a lead compound, including modeling the protein complex, screening for hot spots on the protein-protein interaction interface and screening small molecules or peptides that bind to the PPI interface. There are three major challenges to the use of docking on the relatively flat surfaces of PPI. In this review we will provide some examples of the use of docking in PPI inhibitor design as well as its limitations. The combination of experimental and docking methods with improved scoring function has thus far resulted in few success stories of PPI inhibitors for therapeutic purposes. Docking algorithms used for PPI are in the early stages, however, and as more data are available docking will become a highly promising area in the design of PPI inhibitors or stabilizers.

Protein-protein interaction network prediction by using rigid-body docking tools: application to bacterial chemotaxis

Core elements of cell regulation are made up of protein-protein interaction (PPI) networks. However, many parts of the cell regulatory systems include unknown PPIs. To approach this problem, we have developed a computational method of high-throughput PPI network prediction based on all-to-all rigid-body docking of protein tertiary structures. The prediction system accepts a set of data comprising protein tertiary structures as input and generates a list of possible interacting pairs from all the combinations as output. A crucial advantage of this docking based method is in providing predictions of protein pairs that increases our understanding of biological pathways by analyzing the structures of candidate complex structures, which gives insight into novel interaction mechanisms. Although such exhaustive docking calculation requires massive computational resources, recent advancements in the computational sciences have made such large-scale calculations feasible. different rigid-body docking tools with different scoring models. We found that the predicted interactions were different between the results from the two tools. When the positive predictions from both of the docking tools were combined, all the core signaling interactions were correctly predicted with the exception of interactions activated by protein phosphorylation. Large-scale PPI prediction using tertiary structures is an effective approach that has a wide range of potential applications. This method is especially useful for identifying novel PPIs of new pathways that control cellular behavior.

Computational approaches to developing short cyclic peptide modulators of protein-protein interactions

Cyclic peptides are a promising class of bioactive molecules potentially capable of modulating "difficult" targets, such as protein-protein interactions. Cyclic peptides have long been used as therapeutics derived from natural product derivatives, but remain an underexplored class of compounds from the perspective of rational drug design, possibly due to the known weaknesses of peptide drugs in general. While cyclic peptides are non"druglike" by the accepted empirical rules, their unique structure may lend itself to both membrane permeability and proteolytic resistance-the main barriers to oral delivery. The constrained shape of cyclic peptides also lends itself better to virtual screening approaches, and new tools and successes in this area have been recently noted. An increasing number of strategies are available, both to generate and screen cyclic peptide libraries, and best practises and current successes are described within. This chapter will describe various computational strategies for virtual screening cyclic peptides, along with known implementations and applications. We will explore the generation and screening of diverse combinatorial virtual libraries, incorporating a range of cyclization strategies and structural modifications. More advanced approaches covered include evolutionary algorithms designed to aid in screening large structural libraries, machine learning approaches, and harnessing bioinformatics resources to bias cyclic peptide virtual libraries towards known bioactive structures.

Structural bioinformatics of the interactome

The past decade has seen a dramatic expansion in the number and range of techniques available to obtain genome-wide information and to analyze this information so as to infer both the functions of individual molecules and how they interact to modulate the behavior of biological systems. Here, we review these techniques, focusing on the construction of physical protein-protein interaction networks, and highlighting approaches that incorporate protein structure, which is becoming an increasingly important component of systems-level computational techniques. We also discuss how network analyses are being applied to enhance our basic understanding of biological systems and their disregulation, as well as how these networks are being used in drug development.

Assessing the applicability of template-based protein docking in the twilight zone

The structural modeling of protein interactions in the absence of close homologous templates is a challenging task. Recently, template-based docking methods have emerged to exploit local structural similarities to help ab-initio protocols provide reliable 3D models for protein interactions. In this work, we critically assess the performance of template-based docking in the twilight zone. Our results show that, while it is possible to find templates for nearly all known interactions, the quality of the obtained models is rather limited. We can increase the precision of the models at expenses of coverage, but it drastically reduces the potential applicability of the method, as illustrated by the whole-interactome modeling of nine organisms. Template-based docking is likely to play an important role in the structural characterization of the interaction space, but we still need to improve the repertoire of structural templates onto which we can reliably model protein complexes.

Discovery of binding proteins for a protein target using protein-protein docking-based virtual screening

Target structure-based virtual screening, which employs protein-small molecule docking to identify potential ligands, has been widely used in small-molecule drug discovery. In the present study, we used a protein-protein docking program to identify proteins that bind to a specific target protein. In the testing phase, an all-to-all protein-protein docking run on a large dataset was performed. The three-dimensional rigid docking program SDOCK was used to examine protein-protein docking on all protein pairs in the dataset. Both the binding affinity and features of the binding energy landscape were considered in the scoring function in order to distinguish positive binding pairs from negative binding pairs. Thus, the lowest docking score, the average Z-score, and convergency of the low-score solutions were incorporated in the analysis. The hybrid scoring function was optimized in the all-to-all docking test. The docking method and the hybrid scoring function were then used to screen for proteins that bind to tumor necrosis factor-α (TNFα), which is a well-known therapeutic target for rheumatoid arthritis and other autoimmune diseases. A protein library containing 677 proteins was used for the screen. Proteins with scores among the top 20% were further examined. Sixteen proteins from the top-ranking 67 proteins were selected for experimental study. Two of these proteins showed significant binding to TNFα in an in vitro binding study. The results of the present study demonstrate the power and potential application of protein-protein docking for the discovery of novel binding proteins for specific protein targets.

DockRank: ranking docked conformations using partner-specific sequence homology-based protein interface prediction

Selecting near-native conformations from the immense number of conformations generated by docking programs remains a major challenge in molecular docking. We introduce DockRank, a novel approach to scoring docked conformations based on the degree to which the interface residues of the docked conformation match a set of predicted interface residues. DockRank uses interface residues predicted by partner-specific sequence homology-based protein-protein interface predictor (PS-HomPPI), which predicts the interface residues of a query protein with a specific interaction partner. We compared the performance of DockRank with several state-of-the-art docking scoring functions using Success Rate (the percentage of cases that have at least one near-native conformation among the top m conformations) and Hit Rate (the percentage of near-native conformations that are included among the top m conformations). In cases where it is possible to obtain partner-specific (PS) interface predictions from PS-HomPPI, DockRank consistently outperforms both (i) ZRank and IRAD, two state-of-the-art energy-based scoring functions (improving Success Rate by up to 4-fold); and (ii) Variants of DockRank that use predicted interface residues obtained from several protein interface predictors that do not take into account the binding partner in making interface predictions (improving success rate by up to 39-fold). The latter result underscores the importance of using partner-specific interface residues in scoring docked conformations. We show that DockRank, when used to re-rank the conformations returned by ClusPro, improves upon the original ClusPro rankings in terms of both Success Rate and Hit Rate. DockRank is available as a server at http://einstein.cs.iastate.edu/DockRank/.

Cross-link guided molecular modeling with ROSETTA

Chemical cross-links identified by mass spectrometry generate distance restraints that reveal low-resolution structural information on proteins and protein complexes. The technology to reliably generate such data has become mature and robust enough to shift the focus to the question of how these distance restraints can be best integrated into molecular modeling calculations. Here, we introduce three workflows for incorporating distance restraints generated by chemical cross-linking and mass spectrometry into ROSETTA protocols for comparative and de novo modeling and protein-protein docking. We demonstrate that the cross-link validation and visualization software Xwalk facilitates successful cross-link data integration. Besides the protocols we introduce XLdb, a database of chemical cross-links from 14 different publications with 506 intra-protein and 62 inter-protein cross-links, where each cross-link can be mapped on an experimental structure from the Protein Data Bank. Finally, we demonstrate on a protein-protein docking reference data set the impact of virtual cross-links on protein docking calculations and show that an inter-protein cross-link can reduce on average the RMSD of a docking prediction by 5.0 Å. The methods and results presented here provide guidelines for the effective integration of chemical cross-link data in molecular modeling calculations and should advance the structural analysis of particularly large and transient protein complexes via hybrid structural biology methods.

Predicting binding within disordered protein regions to structurally characterised peptide-binding domains

Disordered regions of proteins often bind to structured domains, mediating interactions within and between proteins. However, it is difficult to identify a priori the short disordered regions involved in binding. We set out to determine if docking such peptide regions to peptide binding domains would assist in these predictions.We assembled a redundancy reduced dataset of SLiM (Short Linear Motif) containing proteins from the ELM database. We selected 84 sequences which had an associated PDB structures showing the SLiM bound to a protein receptor, where the SLiM was found within a 50 residue region of the protein sequence which was predicted to be disordered. First, we investigated the Vina docking scores of overlapping tripeptides from the 50 residue SLiM containing disordered regions of the protein sequence to the corresponding PDB domain. We found only weak discrimination of docking scores between peptides involved in binding and adjacent non-binding peptides in this context (AUC 0.58).Next, we trained a bidirectional recurrent neural network (BRNN) using as input the protein sequence, predicted secondary structure, Vina docking score and predicted disorder score. The results were very promising (AUC 0.72) showing that multiple sources of information can be combined to produce results which are clearly superior to any single source.We conclude that the Vina docking score alone has only modest power to define the location of a peptide within a larger protein region known to contain it. However, combining this information with other knowledge (using machine learning methods) clearly improves the identification of peptide binding regions within a protein sequence. This approach combining docking with machine learning is primarily a predictor of binding to peptide-binding sites, and is not intended as a predictor of specificity of binding to particular receptors.

Interface-resolved network of protein-protein interactions

We define an interface-interaction network (IIN) to capture the specificity and competition between protein-protein interactions (PPI). This new type of network represents interactions between individual interfaces used in functional protein binding and thereby contains the detail necessary to describe the competition and cooperation between any pair of binding partners. Here we establish a general framework for the construction of IINs that merges computational structure-based interface assignment with careful curation of available literature. To complement limited structural data, the inclusion of biochemical data is critical for achieving the accuracy and completeness necessary to analyze the specificity and competition between the protein interactions. Firstly, this procedure provides a means to clarify the information content of existing data on purported protein interactions and to remove indirect and spurious interactions. Secondly, the IIN we have constructed here for proteins involved in clathrin-mediated endocytosis (CME) exhibits distinctive topological properties. In contrast to PPI networks with their global and relatively dense connectivity, the fragmentation of the IIN into distinctive network modules suggests that different functional pressures act on the evolution of its topology. Large modules in the IIN are formed by interfaces sharing specificity for certain domain types, such as SH3 domains distributed across different proteins. The shared and distinct specificity of an interface is necessary for effective negative and positive design of highly selective binding targets. Lastly, the organization of detailed structural data in a network format allows one to identify pathways of specific binding interactions and thereby predict effects of mutations at specific surfaces on a protein and of specific binding inhibitors, as we explore in several examples. Overall, the endocytosis IIN is remarkably complex and rich in features masked in the coarser PPI, and collects relevant detail of protein association in a readily interpretable format.

Much of the work inside the cell is carried out by proteins interacting with other proteins. Each edge in a protein-protein interaction network reflects these functional interactions and each node a separate protein, creating a complex structure that nevertheless follows well-established global and local patterns related to robust protein function. However, this network is not detailed enough to assess whether a particular protein can bind multiple interaction partners simultaneously through distinct interfaces, or whether the partners targeting a specific interface share similar structural or chemical properties. By breaking each protein node into its constituent interface nodes, we generate and assess such a detailed new network. To sample protein binding interactions broadly and accurately beyond those seen in crystal structures, our method combines computational interface assignment with data from biochemical studies. Using this approach we are able to assign interfaces to the majority of known interactions between proteins involved in the clathrin-mediated endocytosis pathway in yeast. Analysis of this interface-interaction network provides novel insights into the functional specificity of protein interactions, and highlights elements of cooperativity and competition among the proteins. By identifying diverse multi-protein complexes, interface-interaction networks also provide a map for targeted drug development.

Low-resolution structural modeling of protein interactome

Structural characterization of protein-protein interactions across the broad spectrum of scales is key to our understanding of life at the molecular level. Low-resolution approach to protein interactions is needed for modeling large interaction networks, given the significant level of uncertainties in large biomolecular systems and the high-throughput nature of the task. Since only a fraction of protein structures in interactome are determined experimentally, protein docking approaches are increasingly focusing on modeled proteins. Current rapid advancement of template-based modeling of protein-protein complexes is following a long standing trend in structure prediction of individual proteins. Protein-protein templates are already available for almost all interactions of structurally characterized proteins, and about one third of such templates are likely correct.

In silico studies of C 3 metabolic pathway proteins of wheat (Triticum aestivum)

Photosynthesis is essential for plant productivity and critical for plant growth. More than 90% of plants have a C₃ metabolic pathway primarily for carbon assimilation. Improving crop yields for food and fuel is a major challenge for plant biology. To enhance the production of wheat there is need to adopt the strategies that can create the change in plants at the molecular level. During the study we have employed computational bioinformatics and interactomics analysis of C₃ metabolic pathway proteins in wheat. The three-dimensional protein modeling provided insight into molecular mechanism and enhanced understanding of physiological processes and biological systems. Therefore in our study, initially we constructed models for nine proteins involving C₃ metabolic pathway, as these are not determined through wet lab experiment (NMR, X-ray Crystallography) and not available in RCSB Protein Data Bank and UniProt KB. On the basis of docking interaction analysis, we proposed the schematic diagram of C₃ metabolic pathway. Accordingly, there also exist vice versa interactions between 3PGK and Rbcl. Future site and directed mutagenesis experiments in C₃ plants could be designed on the basis of our findings to confirm the predicted protein interactions.

On the binding affinity of macromolecular interactions: daring to ask why proteins interact

Interactions between proteins are orchestrated in a precise and time-dependent manner, underlying cellular function. The binding affinity, defined as the strength of these interactions, is translated into physico-chemical terms in the dissociation constant (K_d), the latter being an experimental measure that determines whether an interaction will be formed in solution or not. Predicting binding affinity from structural models has been a matter of active research for more than 40 years because of its fundamental role in drug development. However, all available approaches are incapable of predicting the binding affinity of protein–protein complexes from coordinates alone. Here, we examine both theoretical and experimental limitations that complicate the derivation of structure–affinity relationships. Most work so far has concentrated on binary interactions. Systems of increased complexity are far from being understood. The main physico-chemical measure that relates to binding affinity is the buried surface area, but it does not hold for flexible complexes. For the latter, there must be a significant entropic contribution that will have to be approximated in the future. We foresee that any theoretical modelling of these interactions will have to follow an integrative approach considering the biology, chemistry and physics that underlie protein–protein recognition.

Computational structural analysis of protein interactions and networks

Protein interactions have been at the focus of computational biology in recent years. In particular, interest has come from two different communities--structural and systems biology. Here, we will discuss key systems and structural biology methods that have been used for analysis and prediction of protein-protein interactions and the insight these approaches have provided on the nature and organization of protein-protein interactions inside cells.

Systematic functional prioritization of protein posttranslational modifications

Protein function is often regulated by posttranslational modifications (PTMs), and recent advances in mass spectrometry have resulted in an exponential increase in PTM identification. However, the functional significance of the vast majority of these modifications remains unknown. To address this problem, we compiled nearly 200,000 phosphorylation, acetylation, and ubiquitination sites from 11 eukaryotic species, including 2,500 newly identified ubiquitylation sites for Saccharomyces cerevisiae. We developed methods to prioritize the functional relevance of these PTMs by predicting those that likely participate in cross-regulatory events, regulate domain activity, or mediate protein-protein interactions. PTM conservation within domain families identifies regulatory "hot spots" that overlap with functionally important regions, a concept that we experimentally validated on the HSP70 domain family. Finally, our analysis of the evolution of PTM regulation highlights potential routes for neutral drift in regulatory interactions and suggests that only a fraction of modification sites are likely to have a significant biological role.

PI2PE: A Suite of Web Servers for Predictions Ranging From Protein Structure to Binding Kinetics

PI²PE (http://pipe.sc.fsu.edu) is a suite of four web servers for predicting a variety of folding- and binding-related properties of proteins. These include the solvent accessibility of amino acids upon protein folding, the amino acids forming the interfaces of protein-protein and protein-nucleic acid complexes, and the binding rate constants of these complexes. Three of the servers debuted in 2007, and have garnered ~2,500 unique users and finished over 30,000 jobs. The functionalities of these servers are now enhanced, and a new sever, for predicting the binding rate constants, is added. Together, these web servers form a pipeline from protein sequence to tertiary structure, then to quaternary structure, and finally to binding kinetics.

Cell-Dock: high-performance protein-protein docking

SUMMARY

The application of docking to large-scale experiments or the explicit treatment of protein flexibility are part of the new challenges in structural bioinformatics that will require large computer resources and more efficient algorithms. Highly optimized fast Fourier transform (FFT) approaches are broadly used in docking programs but their optimal code implementation leaves hardware acceleration as the only option to significantly reduce the computational cost of these tools. In this work we present Cell-Dock, an FFT-based docking algorithm adapted to the Cell BE processor. We show that Cell-Dock runs faster than FTDock with maximum speedups of above 200×, while achieving results of similar quality.

AVAILABILITY AND IMPLEMENTATION

The source code is released under GNU General Public License version 2 and can be downloaded from http://mmb.pcb.ub.es/~cpons/Cell-Dock.

CONTACT

djimenez@ac.upc.edu or juanf@bsc.es

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Established and emerging trends in computational drug discovery in the structural genomics era

Bioinformatics and chemoinformatics approaches contribute to hit discovery, hit-to-lead optimization, safety profiling, and target identification and enhance our overall understanding of the health and disease states. A vast repertoire of computational methods has been reported and increasingly combined in order to address more and more challenging targets or complex molecular mechanisms in the context of large-scale integration of structure and bioactivity data produced by private and public drug research. This review explores some key computational methods directly linked to drug discovery and chemical biology with a special emphasis on compound collection preparation, virtual screening, protein docking, and systems pharmacology. A list of generally freely available software packages and online resources is provided, and examples of successful applications are briefly commented upon.

Constructing structural networks of signaling pathways on the proteome scale

Proteins function through their interactions, and the availability of protein interaction networks could help in understanding cellular processes. However, the known structural data are limited and the classical network node-and-edge representation, where proteins are nodes and interactions are edges, shows only which proteins interact; not how they interact. Structural networks provide this information. Protein-protein interface structures can also indicate which binding partners can interact simultaneously and which are competitive, and can help forecasting potentially harmful drug side effects. Here, we use a powerful protein-protein interactions prediction tool which is able to carry out accurate predictions on the proteome scale to construct the structural network of the extracellular signal-regulated kinases (ERK) in the mitogen-activated protein kinase (MAPK) signaling pathway. This knowledge-based method, PRISM, is motif-based, and is combined with flexible refinement and energy scoring. PRISM predicts protein interactions based on structural and evolutionary similarity to known protein interfaces.

Rewiring of PDZ domain-ligand interaction network contributed to eukaryotic evolution

PDZ domain-mediated interactions have greatly expanded during metazoan evolution, becoming important for controlling signal flow via the assembly of multiple signaling components. The evolutionary history of PDZ domain-mediated interactions has never been explored at the molecular level. It is of great interest to understand how PDZ domain-ligand interactions emerged and how they become rewired during evolution. Here, we constructed the first human PDZ domain-ligand interaction network (PDZNet) together with binding motif sequences and interaction strengths of ligands. PDZNet includes 1,213 interactions between 97 human PDZ proteins and 591 ligands that connect most PDZ protein-mediated interactions (98%) in a large single network via shared ligands. We examined the rewiring of PDZ domain-ligand interactions throughout eukaryotic evolution by tracing changes in the C-terminal binding motif sequences of the PDZ ligands. We found that interaction rewiring by sequence mutation frequently occurred throughout evolution, largely contributing to the growth of PDZNet. The rewiring of PDZ domain-ligand interactions provided an effective means of functional innovations in nervous system development. Our findings provide empirical evidence for a network evolution model that highlights the rewiring of interactions as a mechanism for the development of new protein functions. PDZNet will be a valuable resource to further characterize the organization of the PDZ domain-mediated signaling proteome.

Rewiring of interactions is a powerful tool for the evolution of organism complexity. Rewiring among preexisting proteins provides a simple mechanism for the development of new signaling circuits by redirecting information flows without a gain or loss of genes. Particularly, interactions mediated by short linear motifs can be easily changed by mutations during evolution, resulting in a rewiring of interactions. However, how interaction rewiring of linear motif interactions facilitates the emergence of new protein function during evolution is poorly understood. Here, we systematically investigated the rewiring of interactions mediated by PDZ domains, which are one of the most commonly found peptide recognition modules. We found that PDZ domain-ligand interactions are frequently rewired by C-terminal sequence mutations in PDZ ligands during evolution. Especially, rewiring of PDZ domain-ligand interactions was involved in neuronal function development, occurring concurrently with the emergence of vertebrates and suggesting that reorganization of signaling pathways by rewiring PDZ domain-ligand interactions significantly contributed to the evolution of nervous systems in vertebrates. Our findings highlight the rewiring of interactions as an effective means for functional innovation, providing new insight into eukaryotic evolution, which has not been fully explained by only the expansion of protein families.

Kinetic rate constant prediction supports the conformational selection mechanism of protein binding

The prediction of protein-protein kinetic rate constants provides a fundamental test of our understanding of molecular recognition, and will play an important role in the modeling of complex biological systems. In this paper, a feature selection and regression algorithm is applied to mine a large set of molecular descriptors and construct simple models for association and dissociation rate constants using empirical data. Using separate test data for validation, the predicted rate constants can be combined to calculate binding affinity with accuracy matching that of state of the art empirical free energy functions. The models show that the rate of association is linearly related to the proportion of unbound proteins in the bound conformational ensemble relative to the unbound conformational ensemble, indicating that the binding partners must adopt a geometry near to that of the bound prior to binding. Mirroring the conformational selection and population shift mechanism of protein binding, the models provide a strong separate line of evidence for the preponderance of this mechanism in protein-protein binding, complementing structural and theoretical studies.

Assigning confidence scores to protein-protein interactions

Screens for protein-protein interactions using assays like the yeast two-hybrid system have generated volumes of useful data. The protein interactions from these screens have been used to develop a better understanding of the functions of individual proteins, regulatory pathways, molecular machines, and entire biological systems. The value of this data, however, is limited by the inherent frequency of false positives that arise in most protein interaction screens. Appreciable numbers of false positives can crop up in both low-throughput and high-throughput screens, and even in screens that employ stringent criteria for defining a positive. A number of classification systems have been used to help distinguish false positives from biologically relevant true positives. This chapter describes a system for assigning a confidence score to each interaction based on the probability that it is a true positive. Such confidence scores can be used to prioritize interactions for validation. The scores are also useful for network analysis methods that take advantage of probabilistic edge weights. The scoring method does not rely on gold standard datasets of reliable true positives and true negatives, and thus circumvents the challenges associated with obtaining such datasets. Moreover, the scoring method uses data features that are largely assay-independent, making it useful for interactions obtained from a variety of different technologies and screening methods.

Molecular systems biology of Sic1 in yeast cell cycle regulation through multiscale modeling

Cell cycle control is highly regulated to guarantee the precise timing of events essential for cell growth, i.e., DNA replication onset and cell division. Failure of this control plays a role in cancer and molecules called cyclin-dependent kinase (Cdk) inhibitors (Ckis) exploit a critical function in cell cycle timing. Here we present a multiscale modeling where experimental and computational studies have been employed to investigate structure, function and temporal dynamics of the Cki Sic1 that regulates cell cycle progression in Saccharomyces cerevisiae. Structural analyses reveal molecular details of the interaction between Sic1 and Cdk/cyclin complexes, and biochemical investigation reveals Sic1 function in analogy to its human counterpart p27(Kip1), whose deregulation leads to failure in timing of kinase activation and, therefore, to cancer. Following these findings, a bottom-up systems biology approach has been developed to characterize modular networks addressing Sic1 regulatory function. Through complementary experimentation and modeling, we suggest a mechanism that underlies Sic1 function in controlling temporal waves of cyclins to ensure correct timing of the phase-specific Cdk activities.

Flexibility and binding affinity in protein-ligand, protein-protein and multi-component protein interactions: limitations of current computational approaches

The recognition process between a protein and a partner represents a significant theoretical challenge. In silico structure-based drug design carried out with nothing more than the three-dimensional structure of the protein has led to the introduction of many compounds into clinical trials and numerous drug approvals. Central to guiding the discovery process is to recognize active among non-active compounds. While large-scale computer simulations of compounds taken from a library (virtual screening) or designed de novo are highly desirable in the post-genomic area, many technical problems remain to be adequately addressed. This article presents an overview and discusses the limits of current computational methods for predicting the correct binding pose and accurate binding affinity. It also presents the performances of the most popular algorithms for exploring binary and multi-body protein interactions.

Accelerating protein docking in ZDOCK using an advanced 3D convolution library

Computational prediction of the 3D structures of molecular interactions is a challenging area, often requiring significant computational resources to produce structural predictions with atomic-level accuracy. This can be particularly burdensome when modeling large sets of interactions, macromolecular assemblies, or interactions between flexible proteins. We previously developed a protein docking program, ZDOCK, which uses a fast Fourier transform to perform a 3D search of the spatial degrees of freedom between two molecules. By utilizing a pairwise statistical potential in the ZDOCK scoring function, there were notable gains in docking accuracy over previous versions, but this improvement in accuracy came at a substantial computational cost. In this study, we incorporated a recently developed 3D convolution library into ZDOCK, and additionally modified ZDOCK to dynamically orient the input proteins for more efficient convolution. These modifications resulted in an average of over 8.5-fold improvement in running time when tested on 176 cases in a newly released protein docking benchmark, as well as substantially less memory usage, with no loss in docking accuracy. We also applied these improvements to a previous version of ZDOCK that uses a simpler non-pairwise atomic potential, yielding an average speed improvement of over 5-fold on the docking benchmark, while maintaining predictive success. This permits the utilization of ZDOCK for more intensive tasks such as docking flexible molecules and modeling of interactomes, and can be run more readily by those with limited computational resources.

On the analysis of protein-protein interactions via knowledge-based potentials for the prediction of protein-protein docking

Development of effective methods to screen binary interactions obtained by rigid-body protein-protein docking is key for structure prediction of complexes and for elucidating physicochemical principles of protein-protein binding. We have derived empirical knowledge-based potential functions for selecting rigid-body docking poses. These potentials include the energetic component that provides the residues with a particular secondary structure and surface accessibility. These scoring functions have been tested on a state-of-art benchmark dataset and on a decoy dataset of permanent interactions. Our results were compared with a residue-pair potential scoring function (RPScore) and an atomic-detailed scoring function (Zrank). We have combined knowledge-based potentials to score protein-protein poses of decoys of complexes classified either as transient or as permanent protein-protein interactions. Being defined from residue-pair statistical potentials and not requiring of an atomic level description, our method surpassed Zrank for scoring rigid-docking decoys where the unbound partners of an interaction have to endure conformational changes upon binding. However, when only moderate conformational changes are required (in rigid docking) or when the right conformational changes are ensured (in flexible docking), Zrank is the most successful scoring function. Finally, our study suggests that the physicochemical properties necessary for the binding are allocated on the proteins previous to its binding and with independence of the partner. This information is encoded at the residue level and could be easily incorporated in the initial grid scoring for Fast Fourier Transform rigid-body docking methods.

Sequence- and interactome-based prediction of viral protein hotspots targeting host proteins: a case study for HIV Nef

Virus proteins alter protein pathways of the host toward the synthesis of viral particles by breaking and making edges via binding to host proteins. In this study, we developed a computational approach to predict viral sequence hotspots for binding to host proteins based on sequences of viral and host proteins and literature-curated virus-host protein interactome data. We use a motif discovery algorithm repeatedly on collections of sequences of viral proteins and immediate binding partners of their host targets and choose only those motifs that are conserved on viral sequences and highly statistically enriched among binding partners of virus protein targeted host proteins. Our results match experimental data on binding sites of Nef to host proteins such as MAPK1, VAV1, LCK, HCK, HLA-A, CD4, FYN, and GNB2L1 with high statistical significance but is a poor predictor of Nef binding sites on highly flexible, hoop-like regions. Predicted hotspots recapture CD8 cell epitopes of HIV Nef highlighting their importance in modulating virus-host interactions. Host proteins potentially targeted or outcompeted by Nef appear crowding the T cell receptor, natural killer cell mediated cytotoxicity, and neurotrophin signaling pathways. Scanning of HIV Nef motifs on multiple alignments of hepatitis C protein NS5A produces results consistent with literature, indicating the potential value of the hotspot discovery in advancing our understanding of virus-host crosstalk.

Towards the prediction of protein interaction partners using physical docking

Prediction of physical protein-protein interactions represents a key challenge in computational systems biology. This study provides a proof-of-principle that high-throughput in silico protein docking results can be used to predict interaction partners.

Deciphering the whole network of protein interactions for a given proteome (‘interactome') is the goal of many experimental and computational efforts in Systems Biology. Separately the prediction of the structure of protein complexes by docking methods is a well-established scientific area. To date, docking programs have not been used to predict interaction partners. We provide a proof of principle for such an approach. Using a set of protein complexes representing known interactors in their unbound form, we show that a standard docking program can distinguish the true interactors from a background of 922 non-redundant potential interactors. We additionally show that true interactions can be distinguished from non-likely interacting proteins within the same structural family. Our approach may be put in the context of the proposed ‘funnel-energy model'; the docking algorithm may not find the native complex, but it distinguishes binding partners because of the higher probability of favourable models compared with a collection of non-binders. The potential exists to develop this proof of principle into new approaches for predicting interaction partners and reconstructing biological networks.

Prediction of protein-protein interactions: unifying evolution and structure at protein interfaces

The vast majority of the chores in the living cell involve protein-protein interactions. Providing details of protein interactions at the residue level and incorporating them into protein interaction networks are crucial toward the elucidation of a dynamic picture of cells. Despite the rapid increase in the number of structurally known protein complexes, we are still far away from a complete network. Given experimental limitations, computational modeling of protein interactions is a prerequisite to proceed on the way to complete structural networks. In this work, we focus on the question 'how do proteins interact?' rather than 'which proteins interact?' and we review structure-based protein-protein interaction prediction approaches. As a sample approach for modeling protein interactions, PRISM is detailed which combines structural similarity and evolutionary conservation in protein interfaces to infer structures of complexes in the protein interaction network. This will ultimately help us to understand the role of protein interfaces in predicting bound conformations.