Protein-ligand Interaction Fingerprints Research Articles

Identification of novel CDK2 inhibitors by a multistage virtual screening method based on SVM, pharmacophore and docking model

Cyclin-dependent kinase 2 (CDK2) is the family of Ser/Thr protein kinases that has emerged as a highly selective with low toxic cancer therapy target. A multistage virtual screening method combined by SVM, protein-ligand interaction fingerprints (PLIF) pharmacophore and docking was utilised for screening the CDK2 inhibitors. The evaluation of the validation set indicated that this method can be used to screen large chemical databases because it has a high hit-rate and enrichment factor (80.1% and 332.83 respectively). Six compounds were screened out from NCI, Enamine and Pubchem database. After molecular dynamics and binding free energy calculation, two compounds had great potential as novel CDK2 inhibitors and they also showed selective inhibition against CDK2 in the kinase activity assay.

Identification of potential inhibitors for HCV NS5b of genotype 4a by combining dynamic simulation, protein–ligand interaction fingerprint, 3D pharmacophore, docking and 3D QSAR

HCV NS5B polymerase has been one of the most attractive targets for developing new drugs for HCV infection and many drugs were successfully developed, but all of them were designed for targeting Hepatitis C Virus genotype 1 (HCV GT1). Hepatitis C virus genotype 4a (HCV GT4a) dominant in Egypt has paid less attention. Here, we describe our protocol of virtual screening in identification of novel potential potent inhibitors for HCV NS5B polymerase of GT4a using homology modeling, protein–ligand interaction fingerprint (PLIF), docking, pharmacophore, and 3D CoMFA quantitative structure activity relationship (QSAR). Firstly, a high-quality 3D model of HCV NS5B polymerase of GT4a was constructed using crystal structure of HCV NS5B polymerase of GT1 (PDB ID: 3hkw) as a template. Then, both the model and the template were simulated to compare conformational stability. PLIF was generated using five crystal structures of HCV NS5B (PDB ID: 4mia, 4mib, 4mk9, 4mka, and 4mkb), which revealed the most important residues and their interactions with the co-crystalized ligands. After that, a 3D pharmacophore model was developed from the generated PLIF data and then used as a screening filter for 17000328 drug-like zinc database compounds. 900 compounds passed the pharmacophore filter and entered the docking-based virtual screening stage. Finally, a 3D CoMFA QSAR was developed using 42 compounds as a training and 19 compounds as a test set. The 3D CoMFA QSAR was used to design and screen some potential inhibitors, these compounds were further evaluated by the docking stage. The highest ranked five hits from docking result (compounds (p1–p4) and compound q1) were selected for further analysis.Communicated by Ramaswamy H. Sarma

Discovery of potent apoptosis signal-regulating kinase 1 inhibitors via integrated computational strategy and biological evaluation

Apoptosis signal-regulating Kinase 1 (ASK1) has been confirmed as a potential therapeutic target for the treatment of non-alcoholic steatohepatitis (NASH) disorder and the discovery of ASK1 inhibitors has attracted increasing attention. In this work, a series of in silico methods including pharmacophore screening, docking binding site analysis, protein-ligand interaction fingerprint (PLIF) similarity investigation and molecular docking were applied to find the potential hits from commercial compound databases. Five compounds with potential inhibitory activity were purchased and submitted to biological activity validation. Thus, one hit compound was discovered with micromolar IC50 value (10.59 μM) against ASK1. Results demonstrated that the integration of computation methods and biological test was quite reliable for the discovery of potent ASK1 inhibitors and the strategy could be extended to other similar targets of interest.

Computer-assisted engineering of the catalytic activity of a carboxylic acid reductase

Carboxylic acid reductases (CARs) play crucial roles in the biosynthesis of optically pure aldehydes with no side products. It has inspired synthetic organic chemists and biotechnologists to exploit them as catalysts in practical applications. However, levels of activity and substrate specificity are not routinely sufficient. Recent developments in protein engineering have produced numerous biocatalysts with new catalytic properties, whereas such efforts in CARs are limited. In this study, we show that the exploitation of information derived from catalytic mechanism analysis and molecular dynamics simulations assisted the semi-rational engineering of a CAR from Segniliparus rugosus (SrCAR) with the aim of increasing activity. Guided by protein-ligand interaction fingerprinting analysis, 17 residues at the substrate binding pockets were first identified. We then performed single site saturation mutagenesis and successfully obtained variants that gave high activities using benzoic acid as the model substrate. As a result, the best mutant K524W enabled 99% conversion and 17.28 s−1 mM−1kcat/Km, with 7- and 2-fold improvement compared to the wild-type, respectively. The engineered catalyst K524W as well as a second variant K524Q proved to be effective in the reduction of other benzoic acid derivatives. Insight into the source of enhanced activity was gained by molecular dynamics simulations.

Machine Learning Analysis of τRAMD Trajectories to Decipher Molecular Determinants of Drug-Target Residence Times.

Drug-target residence times can impact drug efficacy and safety, and are therefore increasingly being considered during lead optimization. For this purpose, computational methods to predict residence times, τ, for drug-like compounds and to derive structure-kinetic relationships are desirable. A challenge for approaches based on molecular dynamics (MD) simulation is the fact that drug residence times are typically orders of magnitude longer than computationally feasible simulation times. Therefore, enhanced sampling methods are required. We recently reported one such approach: the τRAMD procedure for estimating relative residence times by performing a large number of random acceleration MD (RAMD) simulations in which ligand dissociation occurs in times of about a nanosecond due to the application of an additional randomly oriented force to the ligand. The length of the RAMD simulations is used to deduce τ. The RAMD simulations also provide information on ligand egress pathways and dissociation mechanisms. Here, we describe a machine learning approach to systematically analyze protein-ligand binding contacts in the RAMD trajectories in order to derive regression models for estimating τ and to decipher the molecular features leading to longer τ values. We demonstrate that the regression models built on the protein-ligand interaction fingerprints of the dissociation trajectories result in robust estimates of τ for a set of 94 drug-like inhibitors of heat shock protein 90 (HSP90), even for the compounds for which the length of the RAMD trajectories does not provide a good estimation of τ. Thus, we find that machine learning helps to overcome inaccuracies in the modeling of protein-ligand complexes due to incomplete sampling or force field deficiencies. Moreover, the approach facilitates the identification of features important for residence time. In particular, we observed that interactions of the ligand with the sidechain of F138, which is located on the border between the ATP binding pocket and a hydrophobic transient sub-pocket, play a key role in slowing compound dissociation. We expect that the combination of the τRAMD simulation procedure with machine learning analysis will be generally applicable as an aid to target-based lead optimization.

Virtual Screening Strategy Combined Bayesian Classification Model, Molecular Docking for Acetyl-CoA Carboxylases Inhibitors.

Acetyl-CoA Carboxylases (ACC) have been an important target for the therapy of metabolic syndrome, such as obesity, hepatic steatosis, insulin resistance, dyslipidemia, non-alcoholic fatty liver disease (NAFLD), non-alcoholic steatohepatitis (NASH), type 2 diabetes (T2DM), and some other diseases. In this study, virtual screening strategy combined with Bayesian categorization modeling, molecular docking and binding site analysis with protein ligand interaction fingerprint (PLIF) was adopted to validate some potent ACC inhibitors. First, the best Bayesian model with an excellent value of Area Under Curve (AUC) value (training set AUC: 0.972, test set AUC: 0.955) was used to screen compounds of validation library. Then the compounds screened by best Bayesian model were further screened by molecule docking again. Finally, the hit compounds evaluated with four percentages (1%, 2%, 5%, 10%) were verified to reveal enrichment rates for the compounds. The combination of the ligandbased Bayesian model and structure-based virtual screening resulted in the identification of top four compounds which exhibited excellent IC 50 values against ACC in top 1% of the validation library. In summary, the whole strategy is of high efficiency, and would be helpful for the discovery of ACC inhibitors and some other target inhibitors.

Development of a Graphical User Interface Application to Identify Marginal and Potent Ligands for Estrogen Receptor Alpha

Employing ensemble Protein-Ligand Interaction Fingerprints (ensPLIF) as descriptors in post retrospective Structure-Based Virtual Screening (SBVS) campaigns Quantitative Structure-Activity Relationship (QSAR) analysis has been proven to significantly increase the predictive ability in the identification of potent ligands for estrogen receptor alpha (ERα). In the research presented in this article, similar approaches have been performed to construct and retrospectively validate an SBVS protocol to identify marginal ligands for ERα. Based on both validated SBVS protocols, a graphical-user-interface (GUI) application to identify if a compound is a non-, moderate or potent ligand for ERα was developed. The GUI application was subsequently used to virtually screen genistin, genistein, daidzin, and daidzein, followed by in vitro test employing a cytotoxic assay using 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT) method.

The Development of Target-Specific Machine Learning Models as Scoring Functions for Docking-Based Target Prediction.

The identification of possible targets for a known bioactive compound is of the utmost importance for drug design and development. Molecular docking is one possible approach for in-silico protein target prediction, whereas a molecule is docked into several different protein structures to identify potential targets. This reverse docking approach is hampered by the limitation of current scoring functions to correctly discriminate between targets and nontargets. In this work, a development of target-specific scoring functions is described that showed improved prediction performances for the correct target prediction of both actives and decoys on three validation data sets. In contrast to pure ligand-based approaches, that are in general faster and include a greater target space, docking-based approaches can cover also unknown chemical space that lies outside the known bioactivity data. These target-specific scoring functions are based on known bioactivity data retrieved from ChEMBL and supervised machine learning approaches. Neural Networks and Support Vector Machines (SVMs) models were trained for 20 different protein targets. Our protein-ligand interaction fingerprint PADIF (Protein Atom Score Contributions Derived Interaction Fingerprint) represents the input for training, whereas the PADIFs are calculated based on docking poses of active and inactive compounds. Different data sets of previously unseen molecules were used for the final evaluation and analysis of the prediction performance of the created models. For a single-target selectivity data set, the correct target model returns in most of the cases the highest probabilities scores for their active molecules and with statistically significant differences from the other targets. These probability scores were also predicted and successfully used to rank the targets for molecules of a multitarget data set with activity data described simultaneously for two, three, and four to seven protein targets.

Protein-ligand interaction fingerprints for accurate prediction of dissociation rates of p38 MAPK Type II inhibitors.

Binding/unbinding kinetics are key determinants of drug potencies. However, there are still a lot of challenges in predicting kinetic properties during early-stage drug development. In this work, position-restrained molecular dynamics simulations combined with energy decomposition were applied to extract protein-ligand interaction (PLI) fingerprints along the unbinding pathway of 20 p38 mitogen-activated protein kinase (p38 MAPK) Type II inhibitors. The results showed that the electrostatic and/or van der Waals interaction fingerprints at three key positions can be used for accurate prediction of the dissociation rate constants (koff) of p38 MAPK Type II inhibitors. The strategy proposed in this paper can provide not only an efficient method of predicting the dissociation rates of the p38 MAPK Type II inhibitors, but also the atom-level mechanism of enthalpy-driven unbinding process.

Flavonoids as Putative Epi-Modulators: Insight into Their Binding Mode with BRD4 Bromodomains Using Molecular Docking and Dynamics.

Flavonoids are widely recognized as natural polydrugs, given their anti-inflammatory, antioxidant, sedative, and antineoplastic activities. Recently, different studies showed that flavonoids have the potential to inhibit bromodomain and extraterminal (BET) bromodomains. Previous reports suggested that flavonoids bind between the Z and A loops of the bromodomain (ZA channel) due to their orientation and interactions with P86, V87, L92, L94, and N140. Herein, a comprehensive characterization of the binding modes of fisetin and the biflavonoid, amentoflavone, is discussed. To this end, both compounds were docked with BET bromodomain 4 (BRD4) using four docking programs. The results were post-processed with protein–ligand interaction fingerprints. To gain further insight into the binding mode of the two natural products, the docking results were further analyzed with molecular dynamics simulations. The results showed that amentoflavone makes numerous contacts in the ZA channel, as previously described for flavonoids and kinase inhibitors. It was also found that amentoflavone can potentially make contacts with non-canonical residues for BET inhibition. Most of these contacts were not observed with fisetin. Based on these results, amentoflavone was experimentally tested for BRD4 inhibition, showing activity in the micromolar range. This work may serve as the basis for scaffold optimization and the further characterization of flavonoids as BET inhibitors.

Novel pharmacological maps of protein lysine methyltransferases: key for target deorphanization

Epigenetic therapies are being investigated for the treatment of cancer, cognitive disorders, metabolic alterations and autoinmune diseases. Among the different epigenetic target families, protein lysine methyltransferases (PKMTs), are especially interesting because it is believed that their inhibition may be highly specific at the functional level. Despite its relevance, there are currently known inhibitors against only 10 out of the 50 SET-domain containing members of the PKMT family. Accordingly, the identification of chemical probes for the validation of the therapeutic impact of epigenetic modulation is key. Moreover, little is known about the mechanisms that dictate their substrate specificity and ligand selectivity. Consequently, it is desirable to explore novel methods to characterize the pharmacological similarity of PKMTs, going beyond classical phylogenetic relationships. Such characterization would enable the prediction of ligand off-target effects caused by lack of ligand selectivity and the repurposing of known compounds against alternative targets. This is particularly relevant in the case of orphan targets with unreported inhibitors. Here, we first perform a systematic study of binding modes of cofactor and substrate bound ligands with all available SET domain-containing PKMTs. Protein ligand interaction fingerprints were applied to identify conserved hot spots and contact-specific residues across subfamilies at each binding site; a relevant analysis for guiding the design of novel, selective compounds. Then, a recently described methodology (GPCR-CoINPocket) that incorporates ligand contact information into classical alignment-based comparisons was applied to the entire family of 50 SET-containing proteins to devise pharmacological similarities between them. The main advantage of this approach is that it is not restricted to proteins for which crystallographic data with bound ligands is available. The resulting family organization from the separate analysis of both sites (cofactor and substrate) was retrospectively and prospectively validated. Of note, three hits (inhibition > 50% at 10 µM) were identified for the orphan NSD1.

3D-quantitative structure-activity relationship study for the design of novel enterovirus A71 3C protease inhibitors.

A three-dimensional quantitative structure-activity relationships model of enterovirus A71 3C protease inhibitors was constructed in this study. The protein-ligand interaction fingerprint was analyzed to generate a pharmacophore model. A predictive and reliable three-dimensional quantitative structure-activity relationships model was built based on the Flexible Alignment of AutoGPA. Moreover, three novel compounds (I-III) were designed and evaluated for their biochemical activity against 3C protease and anti-enterovirus A71 activity in vitro. III exhibited excellent inhibitory activity (IC50 =0.031±0.005μM, EC50 =0.036±0.007μM). Thus, this study provides a useful quantitative structure-activity relationships model to develop potent inhibitors for enterovirus A71 3C protease.

A novel interaction fingerprint derived from per atom score contributions: exhaustive evaluation of interaction fingerprint performance in docking based virtual screening

Protein ligand interaction fingerprints are a powerful approach for the analysis and assessment of docking poses to improve docking performance in virtual screening. In this study, a novel interaction fingerprint approach (PADIF, protein per atom score contributions derived interaction fingerprint) is presented which was specifically designed for utilising the GOLD scoring functions’ atom contributions together with a specific scoring scheme. This allows the incorporation of known protein–ligand complex structures for a target-specific scoring. Unlike many other methods, this approach uses weighting factors reflecting the relative frequency of a specific interaction in the references and penalizes destabilizing interactions. In addition, and for the first time, an exhaustive validation study was performed that assesses the performance of PADIF and two other interaction fingerprints in virtual screening. Here, PADIF shows superior results, and some rules of thumb for a successful use of interaction fingerprints could be identified.

Prediction of sensitivity to gefitinib/erlotinib for EGFR mutations in NSCLC based on structural interaction fingerprints and multilinear principal component analysis

BackgroundNon-small cell lung cancer (NSCLC) with activating EGFR mutations, especially exon 19 deletions and the L858R point mutation, is particularly responsive to gefitinib and erlotinib. However, the sensitivity varies for less common and rare EGFR mutations. There are various explanations for the low sensitivity of EGFR exon 20 insertions and the exon 20 T790 M point mutation to gefitinib/erlotinib. However, few studies discuss, from a structural perspective, why less common mutations, like G719X and L861Q, have moderate sensitivity to gefitinib/erlotinib.ResultsTo decode the drug sensitivity/selectivity of EGFR mutants, it is important to analyze the interaction between EGFR mutants and EGFR inhibitors. In this paper, the 30 most common EGFR mutants were selected and the technique of protein-ligand interaction fingerprint (IFP) was applied to analyze and compare the binding modes of EGFR mutant-gefitinib/erlotinib complexes. Molecular dynamics simulations were employed to obtain the dynamic trajectory and a matrix of IFPs for each EGFR mutant-inhibitor complex. Multilinear Principal Component Analysis (MPCA) was applied for dimensionality reduction and feature selection. The selected features were further analyzed for use as a drug sensitivity predictor. The results showed that the accuracy of prediction of drug sensitivity was very high for both gefitinib and erlotinib. Targeted Projection Pursuit (TPP) was used to show that the data points can be easily separated based on their sensitivities to gefetinib/erlotinib.ConclusionsWe can conclude that the IFP features of EGFR mutant-TKI complexes and the MPCA-based tensor object feature extraction are useful to predict the drug sensitivity of EGFR mutants. The findings provide new insights for studying and predicting drug resistance/sensitivity of EGFR mutations in NSCLC and can be beneficial to the design of future targeted therapies and innovative drug discovery.

Molecular Docking and PLIF Studies of Novel Tacrine-Naphtoquinone Hybrids Based on Multi-Target-Directed Ligand Approach for Alzheimer’s Disease

Background: Alzheimer’s disease (AD), the most typical type of dementia and memory loss, is a complicated and progressive neurodegenerative disorder. Due to the multi-factorial etiology of AD, the multi-target-directed ligand (MTDL) approach can be a potential method in seeking new drug candidates for this disease. Methods: In this study, over 200 tacrine-naphtoquinone hybrids have been designed and their drug-likeness, molecular docking, and descriptor analysis were conducted to find out a drug candidate with less toxicity and better binding affinity than tacrine. The Docking analysis was conducted using human acetylcholineesterase (1ACJ), human butyrylcholineesterase (4BDS), and β-secretase (BACE1) (1w51) enzymes using Autodock 4.2 and Vina. Results: Promising results were obtained on the types of interactions. Based on molecular docking on 3 targets as well as protein ligand interaction fingerprint (PLIF) studies, the compounds with better results were introduced as good candidates for synthesis. The validity of docking protocols was confirmed using a set of familiar active ligands and decoys on these targets by means of 2 known statistical metrics such as the receiver-operating characteristic (ROC) and Enrichment Factor (EF). Conclusion: Structure activity relationship (SAR) studies, in these class of compounds, show that the hydroxyethylamine, as a linker, is an essential group to improving binding site to AChE and BACE-1 targets.

BINARY QUANTITATIVE STRUCTURE-ACTIVITY RELATIONSHIP ANALYSIS IN RETROSPECTIVE STRUCTURE-BASED VIRTUAL SCREENING CAMPAIGNS TARGETING ESTROGEN RECEPTOR ALPHA

Objective: The objective of this study is to construct predictive unbiased structure-based virtual screening (SBVS) protocols to identify potent ligands for estrogen receptor alpha by combining molecular docking, protein-ligand interaction fingerprinting (PLIF), and binary quantitative structure-activity relationship (QSAR) analysis using recursive partition and regression tree method.Methods: Employing the enhanced version of a directory of useful decoys, SBVS protocols using molecular docking simulations, and PLIF were constructed and retrospectively validated. To avoid bias, SMILES format of the compounds was used. The predictive abilities of the SBVS protocols were then compared based on the enrichment factor (EF) and the F-measure values.Results: The SBVS protocols resulted in this research were SBVS_1 (employing docking scores of the best pose on every compound to rank the results and selecting compounds within 1% false positives as positive), SBVS_2 (employing decision tree resulted from the binary QSAR analysis using docking scores and PLIF bitstrings of the best pose of every compound as descriptors), and SBVS_3 (employing decision tree resulted from the binary QSAR analysis using ensemble PLIF of the selected poses from optimized docking score as the cutoff). The EF values of SBVS_1, SBVS_2, and SBVS_3 are 28.315, 576.084, and 713.472, respectively, while their F-measure values are 0.310, 0.573, and 0.769, respectively.Conclusion: Highly predictive unbiased SBVS protocols to identify potent estrogen receptor alpha ligands were constructed. Further application in prospective screening is therefore highly suggested.

HIV-1 Non-Nucleoside Reverse Transcriptase Inhibitors: SAR and Lead Optimization Using CoMFA and CoMSIA Studies (1995-2016).

Design of inhibitors for HIV-1 reverse transcriptase inhibition (HIV-1 RT) is one of the successful chemotherapies for the treatment of HIV infection. Among the inhibitors available for HIV-1 RT, non-nucleoside reverse transcriptase inhibitors (NNRTIs) have shown to be very promising and clinically approved drugs. However, the efficiency of many of these drugs has been reduced by the drug-resistant variants of HIV-1 RT. The aim of the current review is to provide a summary of lead optimization strategies from the 3D-QSARs studies on NNRTI class from the past 21 years (1995 to 2016). The conformation dependent-alignment based (CoMFA and CoMSIA) methods have been proven very successful ligand based strategy in the drug design. Here, CoMFA and CoMSIA studies reported for structurally distinct NNRTIs including thiazolobenzimidazole, dipyridodiazepinone, 1,1,3-trioxo [1,2,4]-thiadiazine, formimidoester disulfides, thiocarbamate, thiazolidinone derivatives, etc. have been discussed in detail. In addition, we explore the position of the functional groups that drive the protein-ligand interaction. The structure-activity relationship (SAR) revealed from CoMFA and CoMSIA studies of these drug classes is not only in agreement with the structure-based method but also provides an efficient way of lead optimization. In addition to molecular docking experiments, protein-ligand interaction fingerprints were calculated in order to understand the common binding mode of NNRTI compounds. Overall, this review enlightens the protein-ligand interactions with a detailed SAR discussion for chemotypes. Such discussion will help medicinal chemist to gain a better understanding for the design of novel and promising NNRTI candidates.

Binary Quantitative Structure-Activity Relationship Analysis to Increase the Predictive Ability of Structure-Based Virtual Screening Campaigns Targeting Cyclooxygenase-2

Structure-Based Virtual Screening (SBVS) campaigns employing Protein-Ligand Interaction Fingerprints (PLIF) identification have served as a powerful strategy in fragments and ligands identification, both retro- and prospectively. Most of the SBVS campaigns employed PLIF by comparing them to a reference PLIF to calculate the Tanimoto-coefficient. Since the approach was reference dependent, it could lead to a very different discovery path if a different reference was used. In this article, references independent approach, i.e. decision trees construction using docking score and PLIF as the descriptors to increase the predictive ability of the SBVS campaigns in the identification of ligands for cyclooxygenase-2 is presented. The results showed that the binary Quantitative-Structure Activity Relationship (QSAR) analysis could significantly increase the predictive ability of the SBVS campaign. Moreover, the selected decision tree could also pinpoint the molecular determinants of the ligands binding to cyclooxygenase-2.

Enhance the performance of current scoring functions with the aid of 3D protein-ligand interaction fingerprints

BackgroundIn structure-based drug design, binding affinity prediction remains as a challenging goal for current scoring functions. Development of target-biased scoring functions provides a new possibility for tackling this problem, but this approach is also associated with certain technical difficulties. We previously reported the Knowledge-Guided Scoring (KGS) method as an alternative approach (BMC Bioinformatics, 2010, 11, 193–208). The key idea is to compute the binding affinity of a given protein-ligand complex based on the known binding data of an appropriate reference complex, so the error in binding affinity prediction can be reduced effectively.ResultsIn this study, we have developed an upgraded version, i.e. KGS2, by employing 3D protein-ligand interaction fingerprints in reference selection. KGS2 was evaluated in combination with four scoring functions (X-Score, ChemPLP, ASP, and GoldScore) on five drug targets (HIV-1 protease, carbonic anhydrase 2, beta-secretase 1, beta-trypsin, and checkpoint kinase 1). In the in situ scoring test, considerable improvements were observed in most cases after application of KGS2. Besides, the performance of KGS2 was always better than KGS in all cases. In the more challenging molecular docking test, application of KGS2 also led to improved structure-activity relationship in some cases.ConclusionsKGS2 can be applied as a convenient “add-on” to current scoring functions without the need to re-engineer them, and its application is not limited to certain target proteins as customized scoring functions. As an interpolation method, its accuracy in principle can be improved further with the increasing knowledge of protein-ligand complex structures and binding affinity data. We expect that KGS2 will become a practical tool for enhancing the performance of current scoring functions in binding affinity prediction. The KGS2 software is available upon contacting the authors.

IFPTarget: A Customized Virtual Target Identification Method Based on Protein-Ligand Interaction Fingerprinting Analyses.

Small-molecule target identification is an important and challenging task for chemical biology and drug discovery. Structure-based virtual target identification has been widely used, which infers and prioritizes potential protein targets for the molecule of interest (MOI) principally via a scoring function. However, current "universal" scoring functions may not always accurately identify targets to which the MOI binds from the retrieved target database, in part due to a lack of consideration of the important binding features for an individual target. Here, we present IFPTarget, a customized virtual target identification method, which uses an interaction fingerprinting (IFP) method for target-specific interaction analyses and a comprehensive index (Cvalue) for target ranking. Evaluation results indicate that the IFP method enables substantially improved binding pose prediction, and Cvalue has an excellent performance in target ranking for the test set. When applied to screen against our established target library that contains 11,863 protein structures covering 2842 unique targets, IFPTarget could retrieve known targets within the top-ranked list and identified new potential targets for chemically diverse drugs. IFPTarget prediction led to the identification of the metallo-β-lactamase VIM-2 as a target for quercetin as validated by enzymatic inhibition assays. This study provides a new in silico target identification tool and will aid future efforts to develop new target-customized methods for target identification.