Docking Program Research Articles

CoDock-Ligand: combined template-based docking and CNN-based scoring in ligand binding prediction

For ligand binding prediction, it is crucial for molecular docking programs to integrate template-based modeling with a precise scoring function. Here, we proposed the CoDock-Ligand docking method that combines template-based modeling and the GNINA scoring function, a Convolutional Neural Network-based scoring function, for the ligand binding prediction in CASP15. Among the 21 targets, we obtained successful predictions in top 5 submissions for 14 targets and partially successful predictions for 4 targets. In particular, for the most complicated target, H1114, which contains 56 metal cofactors and small molecules, our docking method successfully predicted the binding of most ligands. Analysis of the failed systems showed that the predicted receptor protein presented conformational changes in the backbone and side chains of the binding site residues, which may cause large structural deviations in the ligand binding prediction. In summary, our hybrid docking scheme was efficiently adapted to the ligand binding prediction challenges in CASP15.

Comparative Assessment of Pose Prediction Accuracy in RNA-Ligand Docking.

Structure-based virtual high-throughput screening is used in early-stage drug discovery. Over the years, docking protocols and scoring functions for protein-ligand complexes have evolved to improve the accuracy in the computation of binding strengths and poses. In the past decade, RNA has also emerged as a target class for new small-molecule drugs. However, most ligand docking programs have been validated and tested for proteins and not RNA. Here, we test the docking power (pose prediction accuracy) of three state-of-the-art docking protocols on 173 RNA-small molecule crystal structures. The programs are AutoDock4 (AD4) and AutoDock Vina (Vina), which were designed for protein targets, and rDock, which was designed for both protein and nucleic acid targets. AD4 performed relatively poorly. For RNA targets for which a crystal structure of a bound ligand used to limit the docking search space is available and for which the goal is to identify new molecules for the same pocket, rDock performs slightly better than Vina, with success rates of 48% and 63%, respectively. However, in the more common type of early-stage drug discovery setting, in which no structure of a ligand-target complex is known and for which a larger search space is defined, rDock performed similarly to Vina, with a low success rate of ∼27%. Vina was found to have bias for ligands with certain physicochemical properties, whereas rDock performs similarly for all ligand properties. Thus, for projects where no ligand-protein structure already exists, Vina and rDock are both applicable. However, the relatively poor performance of all methods relative to protein-target docking illustrates a need for further methods refinement.

Assessing Metadynamics and Docking for Absolute Binding Free Energy Calculations Using Severe Acute Respiratory Syndrome Coronavirus 2 Main Protease Inhibitors.

Absolute binding free energy (ABFE) calculations can be an important part of the drug discovery process by identifying molecules that have the potential to be strong binders for a biomolecular target. Recent work has used free energy perturbation (FEP) theory for these calculations, focusing on a set of 16 inhibitors of the severe acute respiratory syndrome coronavirus 2 main protease (Mpro). Herein, the same data set is evaluated by metadynamics (MetaD), four different docking programs, and molecular mechanics with generalized Born and surface area solvation. MetaD yields a Kendall τ distance of 0.28 and Pearson r2 of 0.49, which reflect somewhat less accuracy than that from the ABFE FEP results. Notably, it is demonstrated that an ensemble docking protocol by which each ligand is docked into the 13 crystal structures in this data set provides improved performance, particularly when docking is carried out with Glide XP (Kendall τ distance = 0.20, Pearson r2 = 0.71), Glide SP (Kendall τ distance = 0.19, Pearson r2 = 0.66), or AutoDock 4 (Kendall τ distance = 0.21, Pearson r2 = 0.55). The best results are obtained with "superconsensus" docking by averaging the 52 results for each compound using the 4 docking protocols and all 13 crystal structures (Kendall τ distance = 0.18, Pearson r2 = 0.73).

EasyDock: customizable and scalable docking tool

Docking of large compound collections becomes an important procedure to discover new chemical entities. Screening of large sets of compounds may also occur in de novo design projects guided by molecular docking. To facilitate these processes, there is a need for automated tools capable of efficiently docking a large number of molecules using multiple computational nodes within a reasonable timeframe. These tools should also allow for easy integration of new docking programs and provide a user-friendly program interface to support the development of further approaches utilizing docking as a foundation. Currently available tools have certain limitations, such as lacking a convenient program interface or lacking support for distributed computations. In response to these limitations, we have developed a module called EasyDock. It can be deployed over a network of computational nodes using the Dask library, without requiring a specific cluster scheduler. Furthermore, we have proposed and implemented a simple model that predicts the runtime of docking experiments and applied it to minimize overall docking time. The current version of EasyDock supports popular docking programs, namely Autodock Vina, gnina, and smina. Additionally, we implemented a supplementary feature to enable docking of boron-containing compounds, which are not inherently supported by Vina and smina, and demonstrated its applicability on a set of 55 PDB protein-ligand complexes.

EQUIBIND: A geometric deep learning-based protein-ligand binding prediction method.

Structure-based virtual screening plays a critical role in drug discovery. However, numerous docking programs, such as AutoDock Vina and Glide, are time-consuming due to the necessity of generating numerous molecular conformations and executing steps like scoring, ranking, and refinement for the ligand-receptor complexes. Consequently, achieving rapid and reliable virtual screening remains a noteworthy challenge. Recently, a team of researchers from Massachusetts Institute of Technology, led by Stärk et al., developed an SE(3)-equivariant geometric deep learning based protein-ligand binding prediction approach, EQUIBIND. In comparison to conventional docking methods, EQUIBIND has the capacity to predict the binding modes of small molecules with target proteins rapidly and precisely. It presents an innovative resolution for high-throughput screening of drug-like compounds.

In silico Molecular Docking Analysis of some Terpenoids against 3CLpro of SARS-CoV-2

The recent pandemic of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has raised global health concerns. The main viral protease called 3-chymotrypsin-like cysteine protease (3CLpro) plays an important role in viral replication by polyproteins processing that are translated from viral RNA. Therefore, the present in silico docking study aimed to assess the inhibitory actions of various terpenoids against 3CLpro of SARS-CoV-2. Molecular docking was performed using ArgusLab 4.0.1. a computational docking program and the protein-ligand interaction was visualized by using Pymol 1.7 software. The inhibitory activity of terpenoids like abietic acid, ferruginol, rosmarinic acid, zingiberine, sugiol, kaempferol and betulinic acid was tested against 3CLpro (PDB ID: 6M2N) using molecular docking paradigm while antiviral drugs- remdesivir, darunavir and hydroxychlorquine- were used as standards for comparison. All phyto-constituents showed an effective binding interaction with 6M2N, and the binding affinity was ranged from –8.854 to –13.398 as compared to remdesivir, darunavir and hyroxychlorquine. Amongst tested compounds, abietic acid, ferruginol and betulinic acid exhibited promising enzyme interaction. Results indicate that based upon the binding energy of abietic acid, ferruginol and betulinic acid could be efficient SARS-CoV-2 3CLpro inhibitors. This is supported by the fact that the effects of some terpenoidal phytochemicals especially abietic acid, ferruginol and betulinic acid showed promising enzyme interaction as compared to remdesivir and darunavir. Therefore, further studies are warranted to confirm the effectiveness of abietic acid, ferruginol and betulinic acid for the therapy of COVID-19.

Deep Learning with Geometry-Enhanced Molecular Representation for Augmentation of Large-Scale Docking-Based Virtual Screening.

Structure-based virtual screening has been a crucial tool in drug discovery for decades. However, as the chemical space expands, the existing structure-based virtual screening techniques based on molecular docking and scoring struggle to handle billion-entry ultralarge libraries due to the high computational cost. To address this challenge, people have resorted to machine learning techniques to enhance structure-based virtual screening for efficiently exploring the vast chemical space. In those cases, compounds are usually treated as sequential strings or two-dimensional topology graphs, limiting their ability to incorporate three-dimensional structural information for downstream tasks. We herein propose a novel deep learning protocol, GEM-Screen, which utilizes the geometry-enhanced molecular representation of the compounds docking to a specific target and is trained on docking scores of a small fraction of a library through an active learning strategy to approximate the docking outcome for yet nontraining entries. This protocol is applied to virtual screening campaigns against the AmpC and D4 targets, demonstrating that GEM-Screen enriches more than 90% of the hit scaffolds for AmpC in the top 4% of model predictions and more than 80% of the hit scaffolds for D4 in the same top-ranking size of library. GEM-Screen can be used in conjunction with traditional docking programs for docking of only the top-ranked compounds to avoid the exhaustive docking of the whole library, thus allowing for discovering top-scoring compounds from billion-entry libraries in a rapid yet accurate fashion.

Computational design of novel nanobodies targeting the receptor binding domain of variants of concern of SARS-CoV-2.

The COVID-19 pandemic has created an urgent need for effective therapeutic and diagnostic strategies to manage the disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). However, the emergence of numerous variants of concern (VOCs) has made it challenging to develop targeted therapies that are broadly specific in neutralizing the virus. In this study, we aimed to develop neutralizing nanobodies (Nbs) using computational techniques that can effectively neutralize the receptor-binding domain (RBD) of SARS-CoV-2 VOCs. We evaluated the performance of different protein-protein docking programs and identified HDOCK as the most suitable program for Nb/RBD docking with high accuracy. Using this approach, we designed 14 novel Nbs with high binding affinity to the VOC RBDs. The Nbs were engineered with mutated amino acids that interacted with key amino acids of the RBDs, resulting in higher binding affinity than human angiotensin-converting enzyme 2 (ACE2) and other viral RBDs or haemagglutinins (HAs). The successful development of these Nbs demonstrates the potential of molecular modeling as a low-cost and time-efficient method for engineering effective Nbs against SARS-CoV-2. The engineered Nbs have the potential to be employed in RBD-neutralizing assays, facilitating the identification of novel treatment, prevention, and diagnostic strategies against SARS-CoV-2.

Role of Docking in Anticancer Drug Discovery

Abstract: The computational method is widely used in the field of drug design as well as discovery. It aids the drug discovery and design process by making the procedure faster while also ensuring fewer human errors. Cancer is a condition with the development of abnormal cells expressing features like uncontrolled growth and cell division. This leads to abnormal tissue enlargement and interrupts the normal functioning of the tissue. Computational methods, mainly the molecular docking method, have been utilised extensively in the field of anticancer drug discovery. Docking is a virtual screening method that can be performed on a large database of compounds. Molecular docking helps in identifying the predominant binding modes of a ligand with a protein whose three-dimensional structure is known. The docking process can predict the method of inhibition of the target molecule by the ligand molecule. Utilities of molecular docking include structure-activity relationship studies, lead identification by virtual screening, optimization of the identified lead, combinatorial library design and more. This review discusses the process of docking, its role in anticancer drug discovery, and a comparison of different docking software. Docking programs are used to make the docking process much more quick, efficient, and with fewer human errors, as it mostly depends on computational algorithms. A description of some representative studies in anticancer drug discovery related to selected docking software, Autodock, SwissDock, ICM, GOLD and Glide, are also mentioned. This paper concludes by emphasizing the importance of docking programs in the field of drug discovery and how it influences the modern drug discovery processes.

A multitier virtual screening of antagonists targeting PD-1/PD-L1 interface for the management of triple-negativebreastcancer.

Immunotherapies are promising therapeutic options for the management of triple-negative breast cancer because of its high mutation rate and genomic instability. Of note, the blockade of the immune checkpoint protein PD-1 and its ligand PD-L1 has been proven to be an efficient and potent strategy to combat triple-negative breast cancer.To date, various anti-PD-1/anti-PD-L1 antibodies have been approved. However, the intrinsic constraints of these therapeutic antibodies significantly limit their application, making small molecules a potentially significant option for PD-1/PD-L1 inhibition. In light of this, the current study aims to use a high-throughput virtual screening technique to identify potential repurposed candidates as PD-L1 inhibitors. Thus, the present study explored binding efficiency of 2509 FDA-approved compounds retrieved from the drug bank database against PD-L1 protein. The binding affinity of the compounds was determined using the glide XP docking programme. Furthermore, prime-MM/GBSA, DFT calculations, and RF score were used to precisely re-score the binding free energy of the docked complexes. In addition, the ADME and toxicity profiles for the lead compounds were also examined to address PK/PD characteristics. Altogether, the screening process identified three molecules, namely DB01238, DB06016 and DB01167 as potential therapeutics for the PD-L1 protein. To conclude, a molecular dynamic simulation of 100ns was run to characterise the stability and inhibitory action of the three lead compounds. The results from the simulation study confirm the robust structural and thermodynamic stability of DB01238 than other investigated molecules.Thus, our findings hypothesize that DB01238 could serve as potential PD-L1 inhibitor in the near future for triple-negative breast cancer patients.

A Structural Model for the Core Nup358-BicD2 Interface.

Dynein motors facilitate the majority of minus-end-directed transport events on microtubules. The dynein adaptor Bicaudal D2 (BicD2) recruits the dynein machinery to several cellular cargo for transport, including Nup358, which facilitates a nuclear positioning pathway that is essential for the differentiation of distinct brain progenitor cells. Previously, we showed that Nup358 forms a "cargo recognition α-helix" upon binding to BicD2; however, the specifics of the BicD2-Nup358 interface are still not well understood. Here, we used AlphaFold2, complemented by two additional docking programs (HADDOCK and ClusPro) as well as mutagenesis, to show that the Nup358 cargo-recognition α-helix binds to BicD2 between residues 747 and 774 in an anti-parallel manner, forming a helical bundle. We identified two intermolecular salt bridges that are important to stabilize the interface. In addition, we uncovered a secondary interface mediated by an intrinsically disordered region of Nup358 that is directly N-terminal to the cargo-recognition α-helix and binds to BicD2 between residues 774 and 800. This is the same BicD2 domain that binds to the competing cargo adapter Rab6, which is important for the transport of Golgi-derived and secretory vesicles. Our results establish a structural basis for cargo recognition and selection by the dynein adapter BicD2, which facilitates transport pathways that are important for brain development.

Testing the Ability of Protein–Protein Docking Programs to Model Known Complexes: A Project for the Biochemistry Classroom

Testing the Ability of Protein–Protein Docking Programs to Model Known Complexes: A Project for the Biochemistry Classroom

Benchmarking different docking protocols for predicting the binding poses of ligands complexed with cyclooxygenase enzymes and screening chemical libraries.

Non-steroidal anti-inflammatory drugs (NSAIDs) constitute an important class of pharmaceuticals acting on cyclooxygenase COX-1 and COX-2 enzymes. Due to their numerous severe side effects, it is necessary to search for new selective, safe, and effective anti-inflammatory drugs. In silico design of novel therapeutics plays an important role in nowadays drug discovery pipelines. In most cases, the design strategies require the use of molecular docking calculations. The docking procedure may require case-specific condition for a successful result. Additionally, many different docking programs are available, which highlights the importance of identifying the most proper docking method and condition for a given problem. In the current work, the performances of five popular molecular docking programs, namely, GOLD, AutoDock, FlexX, Molegro Virtual Docker (MVD) and Glide to predict the binding mode of co- crystallized inhibitors in the structures of known complexes available for cyclooxygenases were evaluated. Furthermore, the best performers, Glide, AutoDock, GOLD and FlexX, were further evaluated in docking-based virtual screening of libraries consisted of active ligands and decoy molecules for cyclooxygenase enzymes and the obtained docking scores were assessed by receiver operating characteristics (ROC) analysis. The results of docking experiments indicated that Glide program outperformed other docking programs by correctly predicting the binding poses (RMSD less than 2 Å) of all studied co-crystallized ligands of COX-1 and COX-2 enzymes (i.e., the performance was 100%). However, the performances of the other studied docking methods for correctly predicting the binding poses of the ligands were between 59% to 82%. Virtual screening results treated by ROC analysis revealed that all tested methods are useful tools for classification and enrichment of molecules targeting COX enzymes. The obtained AUCs range between 0.61-0.92 with enrichment factors of 8 - 40 folds. The obtained results support the importance of choosing appropriate docking method for predicting ligand-receptor binding modes, and provide specific information about docking calculations on COXs ligands.

Accurate ligand-protein docking in CASP15 using the ClusPro LigTBM server.

In the ligand prediction category of CASP15, the challenge was to predict the positions and conformations of small molecules binding to proteins that were provided as amino acid sequences or as models generated by the AlphaFold2 program. For most targets, we used our template-based ligand docking program ClusPro ligTBM, also implemented as a public server available at https://ligtbm.cluspro.org/. Since many targets had multiple chains and a number of ligands, several templates, and some manual interventions were required. In a few cases, no templates were found, and we had to use direct docking using the Glide program. Nevertheless, ligTBM was shown to be a very useful tool, and by any ranking criteria, our group was ranked among the top five best-performing teams. In fact, all the best groups used template-based docking methods. Thus, it appears that the AlphaFold2-generated models, despite the high accuracy of the predicted backbone, have local differences from the x-ray structure that make the use of direct docking methods more challenging. The results of CASP15 confirm that this limitation can be frequently overcome by homology-based docking.

Molecular Docking: A useful approach of Drug Discovery on the Basis of their Structure

Molecular docking is important tool for drug discovery. It provides a valuable tool for drug design and analysis. The most important application of molecular docking is virtual screening. In this review, we present a introduction of methods of molecular docking, and their development and applications in drug discovery. Many docking programs has developed to visualize the three dimensional structure of molecule and docking score can also get with computational methods. This article includes information on molecular docking, molecular modeling, types of docking, molecular docking models, basic requirement of molecular docking, molecular approach, evaluation, applications software available for molecular docking.

Limited Polymorphism in the Dihydrofolate Reductase (dhfr) and dihydropteroate synthase genes (dhps) of Plasmodium knowlesi isolate from Thailand

BackgroundThe 2022 malaria WHO reported around 4000 P. knowlesi infections in the South-East Asia region. In the same period, 72 positive cases were reported by the Department of Disease Control in Thailand, suggesting a persistent infection. Little is known about dihydrofolate reductase (pkdhfr) and dihydropteroate synthase (pkdhps), putative antimalarial resistance markers for P. knowlesi. The relevant amplification and sequencing protocol are presently unavailable. In this study, we developed a protocol for amplifying and evaluating pkdhps mutations. The haplotype pattern of pkdhfr–pkdhps in Thai isolates was analyzed, and the effects of these pkdhps mutations were predicted by using a computer program. MethodsPkdhps were amplified and sequenced from 28 P. knowlesi samples collected in 2008 and 2020 from nine provinces across Thailand. Combining pkdhfr sequencing data from previous work with pkdhps data to analyze polymorphisms of pkdhfr and pkdhps haplotype. Protein modeling and molecular docking were constructed using two inhibitors, sulfadoxine and sulfamethoxazole, and further details were obtained through analyses of protein–ligand interactions by using the Genetic Optimisation for Ligand Docking program. A phylogenetic tree cluster analysis was reconstructed to compare the P. knowlesi Malaysia isolates. ResultsFive nonsynonymous mutations in the pkdhps were detected outside the equivalence of the binding pocket sites to sulfadoxine and sulfamethoxazole, which are at N391S, E421G, I425R, A449S, and N517S. Based on the modeling and molecular docking analyses, the N391S and N517S mutations located close to the enzyme-binding pocket demonstrated a different docking score and protein–ligand interaction in loop 2 of the enzyme. These findings indicated that it was less likely to induce drug resistance. Of the four haplotypes of pkdhfr–pkdhps, the most common one is the R34L pkdhfr mutation and the pkdhps quadruple mutation (GRSS) at E421G, I425R, A449S, and N517S, which were observed in P. knowlesi in southern Thailand (53.57%). Based on the results of neighbor-joining analysis for pkdhfr and pkdhps, the samples isolated from eastern Thailand displayed a close relationship with Cambodia isolates, while southern Thailand isolates showed a long branch separated from the Malaysian isolates. ConclusionsA new PCR protocol amplification and evaluation of dihydropteroate synthase mutations in Knowlesi (pkdhps) has been developed. The most prevalent pkdhfr-pkdhps haplotypes (53.57%) in southern Thailand are R34L pkdhfr mutation and pkdhps quadruple mutation. Further investigation requires additional phenotypic data from clinical isolates, transgenic lines expressing mutant alleles, or recombinant proteins.

Docking and Molecular Dynamics Simulations Clarify Binding Sites for Interactions of Novel Marine Sulfated Glycans with SARS-CoV-2 Spike Glycoprotein.

The entry of SARS-CoV-2 into the host cell is mediated by its S-glycoprotein (SGP). Sulfated glycans bind to the SGP receptor-binding domain (RBD), which forms a ternary complex with its receptor angiotensin converting enzyme 2. Here, we have conducted a thorough and systematic computational study of the binding of four oligosaccharide building blocks from novel marine sulfated glycans (isolated from Pentacta pygmaea and Isostichopus badionotus) to the non-glycosylated and glycosylated RBD. Blind docking studies using three docking programs identified five potential cryptic binding sites. Extensive site-targeted docking and molecular dynamics simulations using two force fields confirmed only two binding sites (Sites 1 and 5) for these novel, highly charged sulfated glycans, which were also confirmed by previously published reports. This work showed the structural features and key interactions driving ligand binding. A previous study predicted Site 2 to be a potential binding site, which was not observed here. The use of several molecular modeling approaches gave a comprehensive assessment. The detailed comparative study utilizing multiple modeling approaches is the first of its kind for novel glycan-SGP interaction characterization. This study provided insights into the key structural features of these novel glycans as they are considered for development as potential therapeutics.

Exploration on learning molecular docking with deep learning models

A deep learning‐powered VS approach combined with two free docking programs are proposed and evaluated for screening an ultra‐large compound library to obtain diverse potential active compounds rapidly and efficiently. We found that it is a practical and transferable strategy to significantly reduce computational cost.BackgroundMolecular docking‐based virtual screening (VS) aims to choose ligands with potential pharmacological activities from millions or even billions of molecules. This process could significantly cut down the number of compounds that need to be experimentally tested. However, during the docking calculation, many molecules have low affinity for a particular protein target, which waste a lot of computational resources.MethodsWe implemented a fast and practical molecular screening approach called DL‐DockVS (deep learning dock virtual screening) by using deep learning models (regression and classification models) to learn the outcomes of pipelined docking programs step‐by‐step.ResultsIn this study, we showed that this approach could successfully weed out compounds with poor docking scores while keeping compounds with potentially high docking scores against 10 DUD‐E protein targets. A self‐built dataset of about 1.9 million molecules was used to further verify DL‐DockVS, yielding good results in terms of recall rate, active compounds enrichment factor and runtime speed.ConclusionsWe comprehensively evaluate the practicality and effectiveness of DL‐DockVS against 10 protein targets. Due to the improvements of runtime and maintained success rate, it would be a useful and promising approach to screen ultra‐large compound libraries in the age of big data. It is also very convenient for researchers to make a well‐trained model of one specific target for predicting other chemical libraries and high docking‐score molecules without docking computation again.

IN SILICO EVALUATION OF SARS-COV-2 PAPAIN-LIKE PROTEASE INHIBITORY ACTIVITY OF SOME FDA-APPROVED DRUGS

Objective: In this study, it was aimed to perform in silico studies on the papain-like protease structure of SARS-CoV-2 (PDB: 7JIT) of 1300 FDA-approved drugs downloaded from the ZINC database. Material and Method: A molecular docking study was performed with PLpro (PDB ID: 7JIT) using four different molecular docking programs for a total of 1300 FDA-approved drugs obtained from the ZINC database. Conivaptan and amphotericin B were obtained in docking analysis with AutoDock Vina and Sybyl-X, respectively. Docking analysis with Glide SP and Glide XP resulted in fludarabine and panobinostat, respectively. Molecular dynamics simulations were performed for a period of 120 ns to check the stability of these four drugs. Result and Discussion: The reliability of the results obtained using four different molecular docking programs on the SARS-CoV-2 papain-like protease of 1300 drug molecules was checked by reinserting the co-crystal ligand. Protein-ligand interactions between fludarabine, conivaptan, amphotericin-B, panobinostat, and PLpro were given. In the molecular dynamics study, RMSD, RMSF, Rg, and SASA analyses were performed for four systems. It was observed that RMSD remained constant for all 120 ns for all four systems except for amphotericin B, which deviated slightly towards the end of 120 ns. No significant fluctuation was noticed in the RMSF graphics for all four systems.

Molecular Docking Analysis of Novel Thiourea Derivatives of Naproxen with Potential Antitumor Activity

Abstract Naproxen, as a propionic acid derivative, causes serious gastrointestinal side effects due to the presence of free carboxylic group. In that sense, masking of carboxylic group with other pharmacophores may be a promising strategy to decrease gastrointestinal toxicity. Thiourea derivatives have been intensively investigated as potential antitumor drugs, whereby their activity is based on potential inhibition of protein kinases, topoisomerases, carbonic anhydrase and sirtuins. In addition, it was shown that inhibition of certain protein kinases might reverse resistance to chemotherapeutic drugs by enhancing the cell death in the presence of low concentrations of drug. Twenty new thiourea derivatives of naproxen were designed and their binding to four selected protein kinases involved in tumor multidrug resistance (AKT2, mTOR, EGFR and VEGFR1) was estimated using two molecular docking programs (AutoDock Vina and OEDocking). According to OEDocking, the highest potential to inhibit AKT2 and mTor has derivative 1, while derivative 20 demonstrates the highest potential towards EGFR and VEGFR1. According to AutoDock Vina, the highest potential for inhibition of EGFR, AKT2 and VEGFR1 have derivatives 16 and 17. Therefore, derivatives 1, 16, 17 and 20 are potentially the most potent protein kinase inhibitors that could be further synthesized and tested for anticancer activity.