Prediction of calcium-binding sites by combining loop-modeling with machine learning

SCAMPER: Accurate Type-Specific Prediction of Calcium-Binding Residues Using Sequence-Derived Features

Understanding molecular mechanisms involved in calcium-protein interactions and modeling corresponding docking rely on the accurate identification of calcium-binding residues (CaBRs). The defects of experimentally annotating protein functions enhances the development of computational approaches that correctly identify calcium-binding interactions. Studies have reported that current methods severely cross-predict residues that interact with other types of molecules (e.g., nucleic acids, proteins, and small ligands) as CaBRs. In this study, a novel predictor named SCAMPER (Selective CAlciuM-binding PrEdictoR) is proposed for the accurate and specific prediction of CaBRs. SCAMPER is designed using newly compiled dataset with complete UniProt sequences and annotations, which include calcium-binding, nucleic acid-binding, protein-binding, and small ligand-binding residues. We use a novel designed two-layer scheme to perform predictions as well as penalize cross-predictions. Empirical tests on an independent test dataset reveals that the proposed method significantly outperforms state-of-the-art predictors. SCAMPER is proved to be capable of distinguishing CaBRs from different types of metal-ion binding residues. We further perform CaBRs predictions on the whole human proteome, and use the results to hypothesize calcium-binding proteins (CaBPs). The latest experimental verified CaBPs and GO analysis prove the accuracy of our predictions. We implement the proposed method and share the data at http://www.inforstation.com/webservers/SCAMPER/.

mebipred: identifying metal binding potential in protein sequence

Motivation

metal-binding proteins have a central role in maintaining life processes. Nearly one-third of known protein structures contain metal ions that are used for a variety of needs, such as catalysis, DNA/RNA binding, protein structure stability, etc. Identifying metal-binding proteins is thus crucial for understanding the mechanisms of cellular activity. However, experimental annotation of protein metal-binding potential is severely lacking, while computational techniques are often imprecise and of limited applicability.

Results

we developed a novel machine learning-based method, mebipred, for identifying metal-binding proteins from sequence-derived features. This method is over 80% accurate in recognizing proteins that bind metal ion-containing ligands; the specific identity of 11 ubiquitously present metal ions can also be annotated. mebipred is reference-free, i.e. no sequence alignments are involved, and is thus faster than alignment-based methods; it is also more accurate than other sequence-based prediction methods. Additionally, mebipred can identify protein metal-binding capabilities from short sequence stretches, e.g. translated sequencing reads, and, thus, may be useful for the annotation of metal requirements of metagenomic samples. We performed an analysis of available microbiome data and found that ocean, hot spring sediments and soil microbiomes use a more diverse set of metals than human host-related ones. For human microbiomes, physiological conditions explain the observed metal preferences. Similarly, subtle changes in ocean sample ion concentration affect the abundance of relevant metal-binding proteins. These results highlight mebipred’s utility in analyzing microbiome metal requirements.

Availability and implementation

mebipred is available as a web server at services.bromberglab.org/mebipred and as a standalone package at https://pypi.org/project/mymetal/.

Supplementary information

Supplementary data are available at Bioinformatics online.

Machine Learning Approaches for Metalloproteins

Metalloproteins are a family of proteins characterized by metal ion binding, whereby the presence of these ions confers key catalytic and ligand-binding properties. Due to their ubiquity among biological systems, researchers have made immense efforts to predict the structural and functional roles of metalloproteins. Ultimately, having a comprehensive understanding of metalloproteins will lead to tangible applications, such as designing potent inhibitors in drug discovery. Recently, there has been an acceleration in the number of studies applying machine learning to predict metalloprotein properties, primarily driven by the advent of more sophisticated machine learning algorithms. This review covers how machine learning tools have consolidated and expanded our comprehension of various aspects of metalloproteins (structure, function, stability, ligand-binding interactions, and inhibitors). Future avenues of exploration are also discussed.

Identifying metal binding amino acids based on backbone geometries as a tool for metalloprotein engineering

Metal cofactors within proteins perform a versatile set of essential cellular functions. In order to take advantage of the diverse functionality of metalloproteins, researchers have been working to design or modify metal binding sites in proteins to rationally tune the function or activity of the metal cofactor. This study has performed an analysis on the backbone atom geometries of metal-binding amino acids among 10 different metal binding sites within the entire protein data bank. A set of 13 geometric parameters (features) was identified that is capable of predicting the presence of a metal cofactor in the protein structure with overall accuracies of up to 97% given only the relative positions of their backbone atoms. The decision tree machine-learning algorithm used can quickly analyze an entire protein structure for the presence of sets of primary metal coordination spheres upon mutagenesis, independent of their original amino acid identities. The methodology was designed for application in the field of metalloprotein engineering. A cluster analysis using the data set was also performed and demonstrated that the features chosen are useful for identifying clusters of structurally similar metal-binding sites.

Solution NMR Structure of Titin N2A Region Ig Domain I83 and Its Interaction with Metal Ions

Titin, the largest single chain protein known so far, has long been known to play a critical role in passive muscle function but recent studies have highlighted titin's role in active muscle function. One of the key elements in this role is the Ca²⁺-dependent interaction between titin's N2A region and the thin filament. An important element in this interaction is I83, the terminal immunoglobulin domain in the N2A region. There is limited structural information about this domain, but experimental evidence suggests that it plays a critical role in the N2A-actin binding interaction. We now report the solution NMR structure of I83 and characterize its dynamics and metal binding properties in detail. Its structure shows interesting relationships to other I-band Ig domains. Metal binding and dynamics data point towards the way the domain is evolutionarily optimized to interact with neighbouring domains. We also identify a calcium binding site on the N-terminal side of I83, which is expected to impact the interdomain interaction with the I82 domain. Together these results provide a first step towards a better understanding of the physiological effects associated with deletion of most of the I83 domain, as occurs in the mdm mouse model, as well as for future investigations of the N2A region.

A deep learning framework to predict binding preference of RNA constituents on protein surface

Protein-RNA interaction plays important roles in post-transcriptional regulation. However, the task of predicting these interactions given a protein structure is difficult. Here we show that, by leveraging a deep learning model NucleicNet, attributes such as binding preference of RNA backbone constituents and different bases can be predicted from local physicochemical characteristics of protein structure surface. On a diverse set of challenging RNA-binding proteins, including Fem-3-binding-factor 2, Argonaute 2 and Ribonuclease III, NucleicNet can accurately recover interaction modes discovered by structural biology experiments. Furthermore, we show that, without seeing any in vitro or in vivo assay data, NucleicNet can still achieve consistency with experiments, including RNAcompete, Immunoprecipitation Assay, and siRNA Knockdown Benchmark. NucleicNet can thus serve to provide quantitative fitness of RNA sequences for given binding pockets or to predict potential binding pockets and binding RNAs for previously unknown RNA binding proteins.

Computational approaches for de novo design and redesign of metal-binding sites on proteins

Metal ions play pivotal roles in protein structure, function and stability. The functional and structural diversity of proteins in nature expanded with the incorporation of metal ions or clusters in proteins. Approximately one-third of these proteins in the databases contain metal ions. Many biological and chemical processes in nature involve metal ion-binding proteins, aka metalloproteins. Many cellular reactions that underpin life require metalloproteins. Most of the remarkable, complex chemical transformations are catalysed by metalloenzymes. Realization of the importance of metal-binding sites in a variety of cellular events led to the advancement of various computational methods for their prediction and characterization. Furthermore, as structural and functional knowledgebase about metalloproteins is expanding with advances in computational and experimental fields, the focus of the research is now shifting towards de novo design and redesign of metalloproteins to extend nature’s own diversity beyond its limits. In this review, we will focus on the computational toolbox for prediction of metal ion-binding sites, de novo metalloprotein design and redesign. We will also give examples of tailor-made artificial metalloproteins designed with the computational toolbox.

Role of phosphatase activity of soluble epoxide hydrolase in regulating simvastatin-activated endothelial nitric oxide synthase

Soluble epoxide hydrolase (sEH) has C-terminal epoxide hydrolase and N-terminal lipid phosphatase activity. Its hydrolase activity is associated with endothelial nitric oxide synthase (eNOS) dysfunction. However, little is known about the role of sEH phosphatase in regulating eNOS activity. Simvastatin, a clinical lipid-lowering drug, also has a pleiotropic effect on eNOS activation. However, whether sEH phosphatase is involved in simvastatin-activated eNOS activity remains elusive. We investigated the role of sEH phosphatase activity in simvastatin-mediated activation of eNOS in endothelial cells (ECs). Simvastain increased the phosphatase activity of sEH, which was diminished by pharmacological inhibitors of sEH phosphatase. In addition, pharmacological inhibition of sEH phosphatase or overexpressing the inactive phosphatase domain of sEH enhanced simvastatin-induced NO bioavailability, tube formation and phosphorylation of eNOS, Akt, and AMP-activated protein kinase (AMPK). In contrast, overexpressing the phosphatase domain of sEH limited the simvastatin-increased NO biosynthesis and eNOS phosphorylation at Ser1179. Simvastatin evoked epidermal growth factor receptor–c-Src–increased Tyr phosphorylation of sEH and formation of an sEH–Akt–AMPK–eNOS complex, which was abolished by the c-Src kinase inhibitor PP1 or c-Src dominant-negative mutant K298M. These findings suggest that sEH phosphatase activity negatively regulates simvastatin-activated eNOS by impeding the Akt–AMPK–eNOS signaling cascade.

High Resolution Prediction of Calcium-Binding Sites in 3D Protein Structures Using FEATURE

Metal-binding proteins are ubiquitous in biological systems ranging from enzymes to cell surface receptors. Among the various biologically active metal ions, calcium plays a large role in regulating cellular and physiological changes. With the increasing number of high-quality crystal structures of proteins associated with their metal ion ligands, many groups have built models to identify Ca(2+) sites in proteins, utilizing information such as structure, geometry, or homology to do the inference. We present a FEATURE-based approach in building such a model and show that our model is able to discriminate between nonsites and calcium-binding sites with a very high precision of more than 98%. We demonstrate the high specificity of our model by applying it to test sets constructed from other ions. We also introduce an algorithm to convert high scoring regions into specific site predictions and demonstrate the usage by scanning a test set of 91 calcium-binding protein structures (190 calcium sites). The algorithm has a recall of more than 93% on the test set with predictions found within 3 Å of the actual sites.

Prediction and analysis of canonical EF hand loop and qualitative estimation of Ca²⁺ binding affinity

The diversity of functions carried out by EF hand-containing calcium-binding proteins is due to various interactions made by these proteins as well as the range of affinity levels for Ca²⁺ displayed by them. However, accurate methods are not available for prediction of binding affinities. Here, amino acid patterns of canonical EF hand sequences obtained from available crystal structures were used to develop a classifier that distinguishes Ca²⁺-binding loops and non Ca²⁺-binding regions with 100% accuracy. To investigate further, we performed a proteome-wide prediction for E. histolytica, and classified known EF-hand proteins. We compared our results with published methods on the E. histolytica proteome scan, and demonstrated our method to be more specific and accurate for predicting potential canonical Ca²⁺-binding loops. Furthermore, we annotated canonical EF-hand motifs and classified them based on their Ca²⁺-binding affinities using support vector machines. Using a novel method generated from position-specific scoring metrics and then tested against three different experimentally derived EF-hand-motif datasets, predictions of Ca²⁺-binding affinities were between 87 and 90% accurate. Our results show that the tool described here is capable of predicting Ca²⁺-binding affinity constants of EF-hand proteins. The web server is freely available at http://202.41.10.46/calb/index.html.

Identifying druggable targets by protein microenvironments matching: application to transcription factors

Druggability of a protein is its potential to be modulated by drug-like molecules. It is important in the target selection phase. We hypothesize that: (i) known drug-binding sites contain advantageous physicochemical properties for drug binding, or “druggable microenvironments” and (ii) given a target, the presence of multiple druggable microenvironments similar to those seen previously is associated with a high likelihood of druggability. We developed DrugFEATURE to quantify druggability by assessing the microenvironments in potential small-molecule binding sites. We benchmarked DrugFEATURE using two data sets. One data set measures druggability using NMR-based screening. DrugFEATURE correlates well with this metric. The second data set is based on historical drug discovery outcomes. Using the DrugFEATURE cutoffs derived from the first, we accurately discriminated druggable and difficult targets in the second. We further identified novel druggable transcription factors with implications for cancer therapy. DrugFEATURE provides useful insight for drug discovery, by evaluating druggability and suggesting specific regions for interacting with drug-like molecules.

Regulation of actin by ion-linked equilibria

Actin assembly, filament mechanical properties, and interactions with regulatory proteins depend on the types and concentrations of salts in solution. Salts modulate actin through both nonspecific electrostatic effects and specific binding to discrete sites. Multiple cation-binding site classes spanning a broad range of affinities (nanomolar to millimolar) have been identified on actin monomers and filaments. This review focuses on discrete, low-affinity cation-binding interactions that drive polymerization, regulate filament-bending mechanics, and modulate interactions with regulatory proteins. Cation binding may be perturbed by actin post-translational modifications and linked equilibria. Partial cation occupancy under physiological and commonly used in vitro solution conditions likely contribute to filament mechanical heterogeneity and structural polymorphism. Site-specific cation-binding residues are conserved in Arp2 and Arp3, and may play a role in Arp2/3 complex activation and actin-filament branching activity. Actin-salt interactions demonstrate the relevance of ion-linked equilibria in the operation and regulation of complex biological systems.

Combining fragment homology modeling with molecular dynamics aims at prediction of Ca²⁺ binding sites in CaBPs

The family of calcium-binding proteins (CaBPs) consists of dozens of members and contributes to all aspects of the cell's function, from homeostasis to learning and memory. However, the Ca²⁺-binding mechanism is still unclear for most of CaBPs. To identify the Ca²⁺-binding sites of CaBPs, this study presented a computational approach which combined the fragment homology modeling with molecular dynamics simulation. For validation, we performed a two-step strategy as follows: first, the approach is used to identify the Ca²⁺-binding sites of CaBPs, which have the EF-hand Ca²⁺-binding site and the detailed binding mechanism. To accomplish this, eighteen crystal structures of CaBPs with 49 Ca²⁺-binding sites are selected to be analyzed including calmodulin. The computational method identified 43 from 49 Ca²⁺-binding sites. Second, we performed the approach to large-conductance Ca²⁺-activated K⁺ (BK) channels which don't have clear Ca²⁺-binding mechanism. The simulated results are consistent with the experimental data. The computational approach may shed some light on the identification of Ca²⁺-binding sites in CaBPs.

eFindSite: improved prediction of ligand binding sites in protein models using meta-threading, machine learning and auxiliary ligands

Molecular structures and functions of the majority of proteins across different species are yet to be identified. Much needed functional annotation of these gene products often benefits from the knowledge of protein-ligand interactions. Towards this goal, we developed eFindSite, an improved version of FINDSITE, designed to more efficiently identify ligand binding sites and residues using only weakly homologous templates. It employs a collection of effective algorithms, including highly sensitive meta-threading approaches, improved clustering techniques, advanced machine learning methods and reliable confidence estimation systems. Depending on the quality of target protein structures, eFindSite outperforms geometric pocket detection algorithms by 15-40 % in binding site detection and by 5-35 % in binding residue prediction. Moreover, compared to FINDSITE, it identifies 14 % more binding residues in the most difficult cases. When multiple putative binding pockets are identified, the ranking accuracy is 75-78 %, which can be further improved by 3-4 % by including auxiliary information on binding ligands extracted from biomedical literature. As a first across-genome application, we describe structure modeling and binding site prediction for the entire proteome of Escherichia coli. Carefully calibrated confidence estimates strongly indicate that highly reliable ligand binding predictions are made for the majority of gene products, thus eFindSite holds a significant promise for large-scale genome annotation and drug development projects. eFindSite is freely available to the academic community at http://www.brylinski.org/efindsite .

Functional fragments of disorder in outer membrane β barrel proteins

The traditional view of “sequence–structure–function” has been amended by the discovery of intrinsically disordered proteins. Almost 50% of PDB structures are now known to have one or more regions of disorder, which are involved in diverse functions. These regions typically possess low aromatic content and sequence complexity as well as high net charge and flexibility. In this study, we examined the composition and contribution of intrinsic disorder in outer membrane β barrel protein functions. Our systematic analysis to find the dual personality (DP) fragments, which often function by disorder–order transitions, revealed the presence of 61 DP fragments with 234 residues in β barrel trans membrane protein structures. It was found that though the disorder is more prevalent in the periplasmic regions, most of the residues which undergo disorder–order transitions are found in the extracellular regions. For example, the calcium binding sites in BtuB protein are found to undergo disorder to order transition upon binding calcium. The conformational change in the cell receptor binding site of the OpcA protein, which is important in host cell interactions of N. meningitidis, was also found to be due to the disorder–order transitions occurring in the presence of the ligand. The natively disordered nature of DP fragments makes it more appropriate to call them “functional fragments of disorder.” The present study provides insight into the roles played by intrinsically disordered regions in outer membrane protein functions.

Identification of cation-binding sites on actin that drive polymerization and modulate bending stiffness

The assembly of actin monomers into filaments and networks plays vital roles throughout eukaryotic biology, including intracellular transport, cell motility, cell division, determining cellular shape, and providing cells with mechanical strength. The regulation of actin assembly and modulation of filament mechanical properties are critical for proper actin function. It is well established that physiological salt concentrations promote actin assembly and alter the overall bending mechanics of assembled filaments and networks. However, the molecular origins of these salt-dependent effects, particularly if they involve nonspecific ionic strength effects or specific ion-binding interactions, are unknown. Here, we demonstrate that specific cation binding at two discrete sites situated between adjacent subunits along the long-pitch helix drive actin polymerization and determine the filament bending rigidity. We classify the two sites as "polymerization" and "stiffness" sites based on the effects that mutations at the sites have on salt-dependent filament assembly and bending mechanics, respectively. These results establish the existence and location of the cation-binding sites that confer salt dependence to the assembly and mechanics of actin filaments.

Using multiple microenvironments to find similar ligand-binding sites: application to kinase inhibitor binding

The recognition of cryptic small-molecular binding sites in protein structures is important for understanding off-target side effects and for recognizing potential new indications for existing drugs. Current methods focus on the geometry and detailed chemical interactions within putative binding pockets, but may not recognize distant similarities where dynamics or modified interactions allow one ligand to bind apparently divergent binding pockets. In this paper, we introduce an algorithm that seeks similar microenvironments within two binding sites, and assesses overall binding site similarity by the presence of multiple shared microenvironments. The method has relatively weak geometric requirements (to allow for conformational change or dynamics in both the ligand and the pocket) and uses multiple biophysical and biochemical measures to characterize the microenvironments (to allow for diverse modes of ligand binding). We term the algorithm PocketFEATURE, since it focuses on pockets using the FEATURE system for characterizing microenvironments. We validate PocketFEATURE first by showing that it can better discriminate sites that bind similar ligands from those that do not, and by showing that we can recognize FAD-binding sites on a proteome scale with Area Under the Curve (AUC) of 92%. We then apply PocketFEATURE to evolutionarily distant kinases, for which the method recognizes several proven distant relationships, and predicts unexpected shared ligand binding. Using experimental data from ChEMBL and Ambit, we show that at high significance level, 40 kinase pairs are predicted to share ligands. Some of these pairs offer new opportunities for inhibiting two proteins in a single pathway.

Small molecule drugs may interact with many proteins. Some of these interactions may cause unexpected effects, including side effects or potentially useful therapeutic effects. One way to predict these effects is to analyze the three-dimensional structure of target proteins, and identify new binding sites for small molecule drugs. Several methods have been proposed for predicting new binding sites, relying on geometric and functional complementarity of the sites and the small molecules. In this paper, we report on a new method for identifying novel protein-drug interactions by analyzing the similarity between binding sites in proteins. The method has relatively weak geometric requirements and allows for conformational change or dynamics in both the ligand and protein. Our results show that geometric flexibility is useful for effectively comparing sites. We have applied the method to evolutionarily distant kinases, and find unexpected shared inhibitor binding. Our results may be valuable for drug repurposing in order to find novel uses for existing kinase inhibitors.

ContPro: A web tool for calculating amino acid contact distances in protein from 3D -structures at different distance threshold

To investigate the functional sites on a protein and the prediction of binding sites (residues)in proteins, it is often required to identify the binding site residues at different distance threshold from protein three dimensional (3D)structures. For the study of a particular protein chain and its interaction with the ligand in complex form, researchers have to parse the output of different available tools or databases for finding binding-site residues. Here we have developed a tool for calculating amino acid contact distances in proteins at different distance threshold from the 3D-structure of the protein. For an input of protein 3D-structure, ContPro can quickly find all binding-site residues in the protein by calculating distances and also allows researchers to select the different distance threshold, protein chain and ligand of interest. Additionally, it can also parse the protein model (in case of multi model protein coordinate file)and the sequence of selected protein chain in Fasta format from the input 3D-structure. The developed tool will be useful for the identification and analysis of binding sites of proteins from 3D-structure at different distance thresholds.

AVAILABILITY

IT CAN BE ACCESSED AT: http://procarb.org/contpro/