Ligand-binding Residues Research Articles

Molecular characterization and computational analysis of a highly specific L-glutaminase from a marine bacterium Bacillus australimaris NIOT30

An alkaline active L-glutaminase (BALG) producing bacterium was screened and identified from seamount sediment samples of the Arabian Sea. The isolate was confirmed to be Bacillus australimaris NIOT30 based on morphological characteristics and 16 S rRNA gene sequencing. The glutaminase gene, balg was PCR amplified, cloned and expressed in E. coli BL21 (DE3) host. The molecular weight of purified BALG was estimated to be 36 kDa and the enzyme showed a specific activity of 507 ± 27 Umg−1 against L-glutamine under optimal assay conditions of pH 7.0 and temperature at 37 °C for 15 min. The enzyme showed maximum activity at pH 7 and retained 95% activity at pH 10. BALG retained a relative activity of about 82% and 45% at 45 °C and 60 °C respectively. The kinetic parameters of BALG, Km and Kcat/Km were determined to be of 210 ± 11 mM and 4.4 × 102 M s−1 respectively. Homology modeling and substrate ligand interaction studies revealed the stability of the enzyme-substrate complex. The present study highlights the characterization of a highly active L-glutaminase from B. australimaris NIOT30. Further, mutational analyses of ligand binding residues would show insights into the affinity of L-Glutaminase.

Pharmacophore-Assisted Covalent Docking Identifies a Potential Covalent Inhibitor for Drug-Resistant Genotype 3 Variants of Hepatitis C Viral NS3/4A Serine Protease.

The emergence of drug-resistance-inducing mutations in Hepatitis C virus (HCV) coupled with genotypic heterogeneity has made targeting NS3/4A serine protease difficult. In this work, we investigated the mutagenic variations in the binding pocket of Genotype 3 (G3) HCV NS3/4A and evaluated ligands for efficacious inhibition. We report mutations at 14 positions within the ligand-binding residues of HCV NS3/4A, including H57R and S139P within the catalytic triad. We then modelled each mutational variant for pharmacophore-based virtual screening (PBVS) followed by covalent docking towards identifying a potential covalent inhibitor, i.e., cpd-217. The binding stability of cpd-217 was then supported by molecular dynamic simulation followed by MM/GBSA binding free energy calculation. The free energy decomposition analysis indicated that the resistant mutants alter the HCV NS3/4A-ligand interaction, resulting in unbalanced energy distribution within the binding site, leading to drug resistance. Cpd-217 was identified as interacting with all NS3/4A G3 variants with significant covalent docking scores. In conclusion, cpd-217 emerges as a potential inhibitor of HCV NS3/4A G3 variants that warrants further in vitro and in vivo studies. This study provides a theoretical foundation for drug design and development targeting HCV G3 NS3/4A.

Investigation of Antibody Tolerance in Methanol for Analytical Purposes: Methanol Effect Patterns and Molecular Mechanisms.

The extraction of targets from biological samples for immunoassays using organic solvents, such as methanol, is often necessary. However, high concentrations of organic solvents in extracts invariably lead to instability of the employed antibody, resulting in poor performance of the immunoassay. Evaluating the tolerance ability and exploring the molecular mechanisms of antibody tolerance in organic solvents are essential for the development of robust immunoassays. In this work, 25 monoclonal antibodies and methanol are utilized as models to address these questions. A novel protocol is initially established to precisely and rapidly determine antibody tolerance in methanol, identifying two distinct methanol effect patterns. Through a detailed investigation of the structural basis, a novel hypothesis regarding methanol effect patterns is proposed, termed "folding-aggregation," which is subsequently validated through molecular dynamics simulations. Furthermore, the investigation of sequence basis reveals significant differences in residue types within the complementarity-determining regions and ligand-binding residues, distinguishing the two antibody methanol effect patterns. Moreover, the methanol effect patterns of the antibodies are defined by germline antibodies. This work represents the first exploration of antibody methanol effect patterns and associated molecular mechanisms, with potential implications for the discovery and engineering of tolerant antibodies for the development of robust immunoassays.

Exploring Impact of Protein Sequence Local Information to Predict Enzyme-Ligand Binding Residues Using Machine Learning

Enzymes are important for various biochemical reactions in living cells. In drug discovery and drug design, identifying small molecule binding residues in enzymes is a crucial step. Although, identifying ligand-binding residues using computational techniques are improving, accurate prediction remains difficult. Therefore, to address this problem, we used the sequence local information, i.e., sequence neighbors around the target residue, and transformed it into two ways. First, into a Chaos Game Representation (CGR) to form a feature vector and train with an Extreme Gradient Boosting classifier (XGB) and second into a non-numeric feature space and apply the conditional probabilistic based approach. Our results suggest that local protein sequence information along with global information can help to develop precise predictors for small molecule binding residues. We hope our observations could facilitate the researchers to investigate more into enzyme-ligand interaction and drug discovery.

Structural basis of ligand specificity and channel activation in an insect gustatory receptor

Gustatory receptors (GRs) are critical for insect chemosensation and are potential targets for controlling pests and disease vectors, making their structural investigation a vital step toward such applications. We present structures of Bombyx mori Gr9 (BmGr9), a fructose-gated cation channel, in agonist-free and fructose-bound states. BmGr9 forms a tetramer similar to distantly related insect odorant receptors (ORs). Upon fructose binding, BmGr9's channel gate opens through helix S7b movements. In contrast to ORs, BmGr9's ligand-binding pocket, shaped by a kinked helix S4 and a shorter extracellular S3-S4 loop, is larger and solvent accessible in both agonist-free and fructose-bound states. Also, unlike ORs, fructose binding by BmGr9 involves helix S5 and a pocket lined with aromatic and polar residues. Structure-based sequence alignments reveal distinct patterns of ligand-binding pocket residue conservation in GR subfamilies associated with different ligand classes. These data provide insight into the molecular basis of GR ligand specificity and function.

Exploring imidazo[4,5-g]quinoline-4,9-dione derivatives as Acinetobacter baumannii efflux pump inhibitor: an in silico approach

Antimicrobial resistance (AMR) is fast becoming a medical crisis affecting the entire global population. World Health Organization (WHO) statistics show that globally 0.7 million people are dying yearly due to the emergence of AMR. By 2050, the expected number of lives lost will be 10 million per year. Acinetobacter baumannii is a dreadful nosocomial pathogen that has developed multidrug resistance (MDR) to several currently prescribed antibiotics worldwide. Overexpression of drug efflux transporters (DETs) is one of the mechanisms of multidrug resistance (MDR) in Acinetobacter baumannii. Therefore, blocking the DET can raise the efficacy of the existing antibiotics by increasing their residence time inside the bacteria. In silico screening of five synthetic compounds against three drug efflux pump from A. baumannii has identified KSA5, a novel imidazo[4,5-g]quinoline-4,9-dione derivative, to block the efflux of antibiotics. Molecular docking and simulation results showed that KSA5 could bind to adeB, adeG, and adeJ by consistently interacting with ligand-binding site residues. KSA5 has a higher binding free energy and a lower HOMO-LUMO energy gap than PAβN, suggesting a better ability to interact and inhibit DETs. Further analysis showed that KSA5 is a drug-like molecule with optimal physicochemical and ADME properties. Hence, KSA5 could be combined with antibiotics to overcome antimicrobial resistance. Communicated by Ramaswamy H. Sarma

Functional characterization of four antenna-biased chemosensory proteins in Dioryctria abietella reveals a broadly tuned olfactory DabiCSP1 and its key residues in ligand-binding

The orientation of the oligophagous cone-feeding moth Dioryctria abietella (Lepidoptera: Pyralidae) to host plants primarily relies on olfactory-related proteins, particularly those candidates highly expressed in antennae. Here, through a combination of expression profile, ligand-binding assay, molecular docking and site-directed mutagenesis strategies, we characterized the chemosensory protein (CSP) gene family in D. abietella. Quantitative real-time PCR (qPCR) analyses revealed the detectable expression of all 22 DabiCSPs in the antennae, of which seven genes were significantly enriched in this tissue. In addition, the majority of the genes (19/22 relatives) had the expression in at least one reproductive tissue. In the interactions of four antenna-dominant DabiCSPs and different chemical classes, DabiCSP1 was broadly tuned to 27 plant-derived odors, three man-made insecticides and one herbicide with high affinities (Ki < 6.60 μM). By contrast, three other DabiCSPs (DabiCSP4, CSP6 and CSP17) exhibited a narrow odor binding spectrum, in response to six compounds for each protein. Our mutation analyses combined with molecular docking simulations and binding assays further identified four key residues (Tyr25, Thr26, Ile65 and Val69) in the interactions of DabiCSP1 and ligands, of which binding abilities of this protein to 12, 15, 16 and three compounds were significantly decreased compared to the wildtype protein, respectively. Our study reveals different odor binding spectra of four DabiCSPs enriched in antennae and identifies key residues responsible for the binding of DabiCSP1 and potentially active compounds for the control of this pest.

Mutation of the PEBP-like domain of the mitoribosomal MrpL35/mL38 protein results in production of nascent chains with impaired capacity to assemble into OXPHOS complexes.

Located in the central protuberance region of the mitoribosome and mitospecific mL38 proteins display homology to PEBP (Phosphatidylethanolamine Binding Protein) proteins, a diverse family of proteins reported to bind anionic substrates/ligands and implicated in cellular signaling and differentiation pathways. In this study, we have performed a mutational analysis of the yeast mitoribosomal protein MrpL35/mL38 and demonstrate that mutation of the PEBP-invariant ligand binding residues Asp(D)232 and Arg(R)288 impacted MrpL35/mL38's ability to support OXPHOS-based growth of the cell. Furthermore, our data indicate these residues exist in a functionally important charged microenvironment, which also includes Asp(D)167 of MrpL35/mL38 and Arg(R)127 of the neighboring Mrp7/bL27m protein. We report that mutation of each of these charged residues resulted in a strong reduction in OXPHOS complex levels that was not attributed to a corresponding inhibition of the mitochondrial translation process. Rather, our findings indicate that a disconnect exists in these mutants between the processes of mitochondrial protein translation and the events required to ensure the competency and/or availability of the newly synthesized proteins to assemble into OXPHOS enzymes. Based on our findings, we postulate that the PEBP-homology domain of MrpL35/mL38, together with its partner Mrp7/bL27m, form a key regulatory region of the mitoribosome.

Structural and biochemical characterisation of the N-carbamoyl-β-alanine amidohydrolase from Rhizobium radiobacter MDC 8606.

N-carbamoyl-β-alanine amidohydrolase (CβAA) constitutes one of the most important groups of industrially relevant enzymes used in the production of optically pure amino acids and derivatives. In this study, a CβAA-encoding gene from Rhizobium radiobacter strain MDC 8606 was cloned and overexpressed in Escherichia coli. The purified recombinant enzyme (RrCβAA) showed a specific activity of 14 U·mg-1 using N-carbamoyl-β-alanine as a substrate with an optimum activity at 55 °C and pH 8.0. In this work, we report also the first prokaryotic CβAA structure at a resolution of 2.0 Å. A discontinuous catalytic domain and a dimerisation domain attached through a flexible hinge region at the domain interface have been revealed. We identify key ligand binding residues, including a conserved glutamic acid (Glu131), histidine (H385) and arginine (Arg291). Our results allowed us to explain the preference of the enzyme for linear carbamoyl substrates, as large and branched carbamoyl substrates cannot fit in the active site of the enzyme. This work envisages the use of RrCβAA from R. radiobacter MDC 8606 for the industrial production of L-α-, L-β- and L-γ-amino acids. The structural analysis provides new insights on enzyme-substrate interaction, which shed light on engineering of CβAAs for high catalytic activity and broad substrate specificity.

Activation of Munc13-1 by Diacylglycerol (DAG)-Lactones.

Munc13-1 is a key protein necessary for vesicle fusion and neurotransmitter release in the brain. Diacylglycerol (DAG)/phorbol ester binds to its C1 domain in the plasma membrane and activates it. The C1 domain of Munc13-1 and protein kinase C (PKC) are homologous in terms of sequence and structure. In order to identify small-molecule modulators of Munc13-1 targeting the C1 domain, we studied the effect of three DAG-lactones, (R,Z)-(2-(hydroxymethyl)-4-(3-isobutyl-5-methylhexylidene)-5-oxotetrahydrofuran-2-yl)methyl pivalate (JH-131e-153), (E)-(2-(hydroxymethyl)-4-(3-isobutyl-5-methylhexylidene)-5-oxotetrahydrofuran-2-yl)methyl pivalate (AJH-836), and (E)-(2-(hydroxymethyl)-4-(4-nitrobenzylidene)-5-oxotetrahydrofuran-2-yl)methyl 4-(dimethylamino)benzoate (130C037), on Munc13-1 activation using the ligand-induced membrane translocation assay. JH-131e-153 showed higher activation than AJH-836, and 130C037 was not able to activate Munc13-1. To understand the role of the ligand-binding site residues in the activation process, three alanine mutants were generated. For AJH-836, the order of activation was wild-type (WT) Munc13-1 > R592A > W588A > I590A. For JH-131e-153, the order of activation was WT > I590 ≈ R592A ≈ W588A. Overall, the Z isomer of DAG-lactones showed higher potency than the E isomer and Trp-588, Ile-590, and Arg-592 were important for its binding. When comparing the activation of Munc13-1 and PKC, the order of activation for JH-131e-153 was PKCα > Munc13-1 > PKCε and for AJH-836, the order of activation was PKCε > PKCα > Munc13-1. Molecular docking supported higher binding of JH-131e-153 than AJH-836 with the Munc13-1 C1 domain. Our results suggest that DAG-lactones have the potential to modulate neuronal processes via Munc13-1 and can be further developed for therapeutic intervention for neurodegenerative diseases.

In Silico Characterisation of the Aedes aegypti Gustatory Receptors.

Aedes aegypti, also known as the dengue mosquito or the yellow fewer mosquito, is the vector of dengue, chikungunya, Zika, Mayaro and yellow fever viruses. The A. aegypti genome contains an array of gustatory receptor (GR) proteins that are related to the recognition of taste. In this study, we performed in silico molecular characterization of all 72 A. aegypti GRs reported in the latest version of A. aegypti genome AaegL5. Phylogenetic analysis classified the receptors into three major clads. Multiple GRs were found to encode multiple transcripts. Physicochemical attributes such as the aliphatic index, hydropathicity index and isoelectric point indicated that A. aegypti gustatory receptors are highly stable and are tailored to perform under a variety of cellular environments. Analysis for subcellular localization indicated that all the GRs are located either in the extracellular matrix or the plasma membrane. Results also indicated that the GRs are distributed mainly on chromosomes 2 and 3, which house 22 and 49 GRs, respectively, whereas chromosome 1 houses only one GR. NCBI-CDD analysis showed the presence of a highly conserved 7tm_7 chemosensory receptor protein superfamily that includes gustatory and odorant receptors from insect species Anopheles gambiae and Drosophila melanogaster. Further, three significantly enriched ungapped motifs in the protein sequence of all 72 A. aegypti gustatory receptors were found. High-quality 3D models for the tertiary structures were predicted with significantly higher confidence, along with ligand-binding residues. Prediction of S-nitrosylation sites indicated the presence of target cysteines in all the GRs with close proximity to the ligand-bindings sites within the 3D structure of the receptors. In addition, two highly conserved motifs inside the GR proteins were discovered that house a tyrosine (Y) and a cysteine (C) residue which may serve as targets for NO-mediated tyrosine nitration and S-nitrosylation, respectively. This study will help devise strategies for functional genomic studies of these important receptor molecules in A. aegypti and other mosquito species through in vitro and in vivo studies.

LigBind: Identifying Binding Residues for Over 1000 Ligands with Relation-Aware Graph Neural Networks

Identifying the interactions between proteins and ligands is significant for drug discovery and design. Considering the diverse binding patterns of ligands, the ligand-specific methods are trained per ligand to predict binding residues. However, most of the existing ligand-specific methods ignore shared binding preferences among various ligands and generally only cover a limited number of ligands with a sufficient number of known binding proteins. In this study, we propose a relation-aware framework LigBind with graph-level pre-training to enhance the ligand-specific binding residue predictions for 1159 ligands, which can effectively cover the ligands with a few known binding proteins. LigBind first pre-trains a graph neural network-based feature extractor for ligand-residue pairs and relation-aware classifiers for similar ligands. Then, LigBind is fine-tuned with ligand-specific binding data, where a domain adaptive neural network is designed to automatically leverage the diversity and similarity of various ligand-binding patterns for accurate binding residue prediction. We construct ligand-specific benchmark datasets of 1159 ligands and 16 unseen ligands, which are used to evaluate the effectiveness of LigBind. The results demonstrate the LigBind’s efficacy on large-scale ligand-specific benchmark datasets, and it generalizes well to unseen ligands. LigBind also enables accurate identification of the ligand-binding residues in the main protease, papain-like protease and the RNA-dependent RNA polymerase of SARS-CoV-2. The web server and source codes of LigBind are available at http://www.csbio.sjtu.edu.cn/bioinf/LigBind/ and https://github.com/YYingXia/LigBind/ for academic use.

A computational approach for the identification of distant homologs of bacterial riboswitches based on inverse RNA folding.

Riboswitches are conserved structural ribonucleic acid (RNA) sensors that are mainly found to regulate a large number of genes/operons in bacteria. Presently, >50 bacterial riboswitch classes have been discovered, but only the thiamine pyrophosphate riboswitch class is detected in a few eukaryotes like fungi, plants and algae. One of the most important challenges in riboswitch research is to discover existing riboswitch classes in eukaryotes and to understand the evolution of bacterial riboswitches. However, traditional search methods for riboswitch detection have failed to detect eukaryotic riboswitches besides just one class and any distant structural homologs of riboswitches. We developed a novel approach based on inverse RNA folding that attempts to find sequences that match the shape of the target structure with minimal sequence conservation based on key nucleotides that interact directly with the ligand. Then, to support our matched candidates, we expanded the results into a covariance model representing similar sequences preserving the structure. Our method transforms a structure-based search into a sequence-based search that considers the conservation of secondary structure shape and ligand-binding residues. This method enables us to identify a potential structural candidate in fungi that could be the distant homolog of bacterial purine riboswitches. Further, phylogenomic analysis and evolutionary distribution of this structural candidate indicate that the most likely point of origin of this structural candidate in these organisms is associated with the loss of traditional purine riboswitches. The computational approach could be applicable to other domains and problems in RNA research.

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/.

In silico prediction of molecular mechanisms of toxicity mediated by the leptospiral PF07598 gene family-encoded virulence-modifying proteins.

Mechanisms of leptospirosis pathogenesis remain unclear despite the identification of a number of potential leptospiral virulence factors. We recently demonstrated potential mechanisms by which the virulence-modifying (VM) proteins-defined as containing a Domain of Unknown function (DUF1561), encoded by the PF07598 gene family-found only in group 1 pathogenic Leptospira-might mediate the clinical pathogenesis of leptospirosis. VM proteins belongs to classical AB toxin paradigm though have a unique AB domain architecture, unlike other AB toxins such as diphtheria toxin, pertussis toxin, shiga toxin, or ricin toxin which are typically encoded by two or more genes and self-assembled into a multi-domain holotoxin. Leptospiral VM proteins are secreted R-type lectin domain-containing exotoxins with discrete N-terminal ricin B-like domains involved in host cell surface binding, and a C-terminal DNase/toxin domain. Here we use the artificial intelligence-based AlphaFold algorithm and other computational tools to predict and elaborate on details of the VM protein structure-function relationship. Comparative AlphaFold and CD-spectroscopy defined the consistent secondary structure (Helix and ß-sheet) content, and the stability of the functional domains were further supported by molecular dynamics simulation. VM proteins comprises distinctive lectic family (QxW)3 motifs, the Mycoplasma CARDS toxin (D3 domain, aromatic patches), C-terminal similarity with mammalian DNase I. In-silico study proposed that Gln412, Gln523, His533, Thr59 are the high binding energy or ligand binding residues plausibly anticipates in the functional activities. Divalent cation (Mg+2-Gln412) and phosphate ion (PO4]-3-Arg615) interaction further supports the functional activities driven by C-terminal domain. Computation-driven structure-function studies of VM proteins will guide experimentation towards mechanistic understandings of leptospirosis pathogenesis, which underlie development of new therapeutic and preventive measures for this devastating disease.

Potential Disruption of Systemic Hormone Transport by Tobacco Alkaloids Using Computational Approaches.

Tobacco/nicotine is one of the most toxic and addictive substances and continues to pose a significant threat to global public health. The harmful effects of smoking/nicotine affect every system in the human body. Nicotine has been associated with effects on endocrine homeostasis in humans such as the imbalance of gonadal steroid hormones, adrenal corticosteroid hormones, and thyroid hormones. The present study was conducted to characterize the structural binding interactions of nicotine and its three important metabolites, cotinine, trans-3'-hydroxycotinine, and 5'-hydroxycotinine, against circulatory hormone carrier proteins, i.e., sex-hormone-binding globulin (SHBG), corticosteroid-binding globulin (CBG), and thyroxine-binding globulin (TBG). Nicotine and its metabolites formed nonbonded contacts and/or hydrogen bonds with amino acid residues of the carrier proteins. For SHBG, Phe-67 and Met-139 were the most important amino acid residues for nicotine ligand binding showing the maximum number of interactions and maximum loss in ASA. For CBG, Trp-371 and Asn-264 were the most important amino acid residues, and for TBG, Ser-23, Leu-269, Lys-270, Asn-273, and Arg-381 were the most important amino acid residues. Most of the amino acid residues of carrier proteins interacting with nicotine ligands showed a commonality with the interacting residues for the native ligands of the proteins. Taken together, the results suggested that nicotine and its three metabolites competed with native ligands for binding to their carrier proteins. Thus, nicotine and its three metabolites may potentially interfere with the binding of testosterone, estradiol, cortisol, progesterone, thyroxine, and triiodothyronine to their carrier proteins and result in the disbalance of their transport and homeostasis in the blood circulation.

1H-1,2,3-Triazole-4H-chromene-D-glucose hybrid compounds: Synthesis and inhibitory activity against Mycobacterium tuberculosis protein tyrosine phosphatase B.

A series of 1H-1,2,3-triazole-4H-chromene-D-glucose hybrid compounds 7a-w were synthesized using click chemistry of 2-amino-7-propargyloxy-4H-chromene-3-carbonitriles 5a-w. CuNPs@montmorillonite was used as a catalyst in the presence of DIPEA as an additive for this chemistry. All synthesized 1H-1,2,3-triazoles were examined for in vitro inhibition against Mycobacterium tuberculosis protein tyrosine phosphatase B (MtbPtpB). Nine 1H-1,2,3-triazoles, including 7c-e, 7h, 7i, and 7r-t, displayed remarkable inhibitory activity against MtbPtpB with IC50 < 10 μM; compound 7t exhibited the most potent inhibition in vitro with an IC50 value of 0.61 μM. Kinetic studies of the three most active compounds, 7c,h,t, showed their competitive inhibition toward the MtbPtpB enzyme. Induced-fit docking and MM-GBSA studies on the enzyme (PDB: 2OZ5) revealed that the most active compound 7t was more effective against MtbPtpB. Residues Arg64, Arg136, Ash165, Arg166, and Arg63 in the binding pocket were identified as potential ligand-binding hot-spot residues for ligand 7t. The binding free energy calculation by the MM-GBSA approach for ligand 7t indicated that Coulomb, lipophilic, and van der Waals energy terms are major contributors to the inhibitor binding. Furthermore, the stability of the ligand-protein complex and the structural insights into the mode of binding were confirmed by 300-ns molecular dynamics simulation of 7t/2OZ5.

BindWeb: A web server for ligand binding residue and pocket prediction from protein structures.

Knowledge of protein-ligand interactions is beneficial for biological process analysis and drug design. Given the complexity of the interactions and the inadequacy of experimental data, accurate ligand binding residue and pocket prediction remains challenging. In this study, we introduce an easy-to-use web server BindWeb for ligand-specific and ligand-general binding residue and pocket prediction from protein structures. BindWeb integrates a graph neural network GraphBind with a hybrid convolutional neural network and bidirectional long short-term memory network DELIA to identify binding residues. Furthermore, BindWeb clusters the predicted binding residues to binding pockets with mean shift clustering. The experimental results and case study demonstrate that BindWeb benefits from the complementarity of two base methods. BindWeb is freely available for academic use at http://www.csbio.sjtu.edu.cn/bioinf/BindWeb/.

Protein-DNA Binding Residue Prediction via Bagging Strategy and Sequence-Based Cube-Format Feature.

Protein-DNA interactions play an important role in diverse biological processes. Accurately identifying protein-DNA binding residues is a critical but challenging task for protein function annotations and drug design. Although wet-lab experimental methods are the most accurate way to identify protein-DNA binding residues, they are time consuming and labor intensive. There is an urgent need to develop computational methods to rapidly and accurately predict protein-DNA binding residues. In this study, we propose a novel sequence-based method, named PredDBR, for predicting DNA-binding residues. In PredDBR, for each query protein, its position-specific frequency matrix (PSFM), predicted secondary structure (PSS), and predicted probabilities of ligand-binding residues (PPLBR) are first generated as three feature sources. Secondly, for each feature source, the sliding window technique is employed to extract the matrix-format feature of each residue. Then, we design two strategies, i.e., square root (SR) and average (AVE), to separately transform PSFM-based and two predicted feature source-based, i.e., PSS-based and PPLBR-based, matrix-format features of each residue into three corresponding cube-format features. Finally, after serially combining the three cube-format features, the ensemble classifier is generated via applying bagging strategy to multiple base classifiers built by the framework of 2D convolutional neural network. The computational experimental results demonstrate that the proposed PredDBR achieves an average overall accuracy of 93.7% and a Mathew's correlation coefficient of 0.405 on two independent validation datasets and outperforms several state-of-the-art sequenced-based protein-DNA binding residue predictors. The PredDBR web-server is available at https://jun-csbio.github.io/PredDBR/.

Prediction of metal ion ligand binding residues by adding disorder value and propensity factors based on deep learning algorithm.

Proteins need to interact with different ligands to perform their functions. Among the ligands, the metal ion is a major ligand. At present, the prediction of protein metal ion ligand binding residues is a challenge. In this study, we selected Zn2+, Cu2+, Fe2+, Fe3+, Co2+, Mn2+, Ca2+ and Mg2+ metal ion ligands from the BioLip database as the research objects. Based on the amino acids, the physicochemical properties and predicted structural information, we introduced the disorder value as the feature parameter. In addition, based on the component information, position weight matrix and information entropy, we introduced the propensity factor as prediction parameters. Then, we used the deep neural network algorithm for the prediction. Furtherly, we made an optimization for the hyper-parameters of the deep learning algorithm and obtained improved results than the previous IonSeq method.