Practical High-Quality Electrostatic Potential Surfaces for Drug Discovery Using a Graph-Convolutional Deep Neural Network

Protein engineering enables a soakable crystal form of human CDK7 primed for high-throughput crystallography and structure-based drug design

Cyclin dependent kinase 7 (CDK7) is an important therapeutic kinase best known for its dual role in cell cycle regulation and gene transcription. Here, we describe the application of protein engineering to generate constructs leading to high resolution crystal structures of human CDK7 in both active and inactive conformations. The active state of the kinase was crystallized by incorporation of an additional surface residue mutation (W132R) onto the double phosphomimetic mutant background (S164D and T170E) that yielded the inactive kinase structure. A novel back-soaking approach was developed to determine crystal structures of several clinical and pre-clinical inhibitors of this kinase, demonstrating the potential utility of the crystal system for structure-based drug design (SBDD). The crystal structures help to rationalize the mode of inhibition and the ligand selectivity profiles versus key anti-targets. The protein engineering approach described here illustrates a generally applicable strategy for structural enablement of challenging molecular targets.

Elucidation of escitalopram oxalate and related antidepressants as putative inhibitors of PTP4A3/PRL-3 protein in hepatocellular carcinoma: A multi-computational investigation

Hepatocellular carcinoma (HCC) persists to be one of the most devastating and deadliest malignancies globally. Recent research into the molecular signaling networks entailed in many malignancies has given some prominent insights that can be leveraged to create molecular therapeutics for combating HCC. Therefore, in the current communication, an in-silico drug repurposing approach has been employed to target the function of PTP4A3/PRL-3 protein in HCC using antidepressants: Fluoxetine hydrochloride, Citalopram, Amitriptyline, Imipramine, and Escitalopram oxalate as the desired ligands. The density function theory (DFT) and chemical absorption, distribution, metabolism, excretion, and toxicity (ADMET) parameters for the chosen ligands were evaluated to comprehend the pharmacokinetics, drug-likeness properties, and bioreactivity of the ligands. The precise interaction mechanism was explored using computational methods such as molecular docking and molecular dynamics (MD) simulation studies to assess the inhibitory effect and the stability of the interactions against the protein of interest. Escitalopram oxalate exhibited a comparatively significant docking score (-7.4 kcal/mol) compared to the control JMS-053 (-6.8 kcal/mol) against the PRL-3 protein. The 2D interaction plots exhibited an array of hydrophobic and hydrogen bond interactions. The findings of the ADMET forecast confirmed that it adheres to Lipinski's rule of five with no violations, and DFT analysis revealed a HOMO-LUMO energy gap of -0.26778 ev, demonstrating better reactivity than the control molecule. The docked complexes were subjected to MD studies (100 ns) showing stable interactions. Considering all the findings, it can be concluded that Escitalopram oxalate and related therapeutics can act as potential pharmacological candidates for targeting the activity of PTP4A3/PRL-3 in HCC.

Fragment-Based Discovery of Allosteric Inhibitors of SH2 Domain-Containing Protein Tyrosine Phosphatase-2 (SHP2)

The ubiquitously expressed protein tyrosine phosphatase SHP2 is required for signaling downstream of receptor tyrosine kinases (RTKs) and plays a role in regulating many cellular processes. Genetic knockdown and pharmacological inhibition of SHP2 suppresses RAS/MAPK signaling and inhibit the proliferation of RTK-driven cancer cell lines. Here, we describe the first reported fragment-to-lead campaign against SHP2, where X-ray crystallography and biophysical techniques were used to identify fragments binding to multiple sites on SHP2. Structure-guided optimization, including several computational methods, led to the discovery of two structurally distinct series of SHP2 inhibitors binding to the previously reported allosteric tunnel binding site (Tunnel Site). One of these series was advanced to a low-nanomolar lead that inhibited tumor growth when dosed orally to mice bearing HCC827 xenografts. Furthermore, a third series of SHP2 inhibitors was discovered binding to a previously unreported site, lying at the interface of the C-terminal SH2 and catalytic domains.

Discovery of new N-Phenylamide Isoxazoline derivatives with high insecticidal activity and reduced honeybee toxicity

Isoxazoline is a novel structure with strong potential for controlling agricultural insect pests, but its high toxicity to honeybees limits its development in agriculture. Herein, a series of N-phenylamide isoxazoline derivatives with low honeybee toxicity were designed and synthesized using the intermediate derivatization method. Bioassay results showed that these compounds exhibited good insecticidal activity. Compounds 3b and 3f showed significant insecticidal effects against Plutella xylostella (P. xylostella) with median lethal concentrations (LC50) of 0.06 and 0.07 mg/L, respectively, comparable to that of fluralaner (LC50 = 0.02 mg/L) and exceeding that of commercial insecticide fluxametamide (LC50 = 0.52 mg/L). It is noteworthy that the acute honeybee toxicities of compounds 3b and 3f (LD50 = 1.43 and 1.63 μg/adult, respectively) were significantly reduced to 1/10 of that of fluralaner (LD50 = 0.14 μg/adult), and were adequate or lower than that of fluxametamide (LD50 = 1.14 μg/adult). Theoretical simulation using molecular docking indicates that compound 3b has similar binding modes with fluralaner and a similar optimal docking pose with fluxametamide when binding to the GABA receptor, which may contribute to its potent insecticidal activity and relatively low toxicity to honey bees. This study provides compounds 3b and 3f as potential new insecticide candidates and provides insights into the development of new isoxazoline insecticides exhibiting both high efficacy and environmental safety.

Bioisosteric modification of Linezolid identified the potential M. tuberculosis protein synthesis inhibitors to overcome the myelosuppression and serotonergic toxicity associated with Linezolid in the treatment of the multi-drug resistance tuberculosis (MDR-TB)

Linezolid is the first and only oxazolidinone antibacterial drug was approved in the last 35 years. It exhibits bacteriostatic efficacy against M. tuberculosis and is a crucial constituent of the BPaL regimen (Bedaquiline, Pretomanid, and Linezolid), which was authorized by the FDA in 2019 for the treatment of XDR-TB or MDR-TB. Despite its unique mechanism of action, Linezolid carries a considerable risk of toxicity, including myelosuppression and serotonin syndrome (SS), which is caused by inhibition of mitochondrial protein synthesis (MPS) and monoamine oxidase (MAO), respectively. Based on the structure toxicity relationship (STR) of Linezolid, in this work, we used a bioisosteric replacement approach to optimize the structure of Linezolid at the C-ring and/or C-5 position for myelosuppression and serotogenic toxicity. Extensive hierarchical multistep docking, drug likeness prediction, molecular binding interactions analyses, and toxicity assessment identified three promising compounds (3071, 7549 and 9660) as less toxic potential modulators of Mtb EthR protein. Compounds 3071, 7549 and 9660 were having the significant docking score of -12.696 Kcal/mol, -12.681 Kcal/mol and -15.293 Kcal/mol towards the Mtb EthR protein with less MAO-A and B affinity [compound 3071: MAO A (-4.799 Kcal/mol) and MAO B (-6.552 Kcal/mol); compound 7549: MAO A (> -2.00 Kcal/mol) and MAO B (> -2.00 Kcal/mol) and compound 9660: MAO A (> -5.678 Kcal/mol) and MAO B (> -6.537Kcal/mol) and none of them shown the Leukopenia as a side effect due to the Myelosuppression. The MD simulation results and binding free energy estimations correspond well with docking analyses, indicating that the proposed compounds bind and inhibit the EthR protein more effectively than Linezolid. The quantum mechanical and electrical characteristics were evaluated using density functional theory (DFT), which also demonstrated that the proposed compounds are more reactive than Linezolid.Communicated by Ramaswamy H. Sarma.

Antibactericidal Ir(III) and Ru(II) Complexes with Phosphine-Alkaloid Conjugate and Their Interactions with Biomolecules: A Case of N-Methylphenethylamine

The phosphine ligand (Ph2 PCH2 N(CH3 )(CH2 )2 Ph, PNMPEA) obtained by the reaction of the (hydroxymethyl)diphenylphosphine with naturally occurring alkaloid N-methylphenethylamine, was used to synthesize the half-sandwich iridium(III) (Ir(η5 -Cp*)Cl2 Ph2 PCH2 N(CH3 )(CH2 )2 Ph, IrPNMPEA) and ruthenium(II) (Ru(η6 -p-cymene)Cl2 Ph2 PCH2 N(CH3 )(CH2 )2 Ph, RuPNMPEA) complexes. They were characterized using a vast array of methods, including 1D and 2D NMR, ESI(+)MS spectrometry, elemental analysis, cyclic voltammetry (CV), electron spectroscopy in the UV-Vis range (absorption, fluorescence) and density functional theory (DFT). The initial antimicrobial activity in vitro toward Gram-positive and Gram-negative bacterial strains was examined, indicating that both complexes are selective towards Gram-positive bacteria, e. g., Staphylococcus aureus, where the IrPNMPEA has been more bactericidal compared to RuPNMPEA. Additionally, the interactions of these compounds with various biomolecules, such as DNA (ctDNA, plasmid DNA, 9-ethylguanine (9-EtG), and 9-methyladenine (9-MeA)), nicotinamide adenine dinucleotide (NADH), glutathione (GSH), and ascorbic acid (Asc) were described. The results showed that both Ir(III) and Ru(II) complexes accelerate the oxidation process of NADH, GSH and Asc that appeared to occur by an electron transfer mechanism. Interestingly, only IrPNMPEA leads to the formation of various biomolecule adducts, which can explain its higher activity. Furthermore, RuPNMPEA and IrPNMPEA have been interacting with the DNA through weak noncovalent interactions.

Open-Source Machine Learning in Computational Chemistry

The field of computational chemistry has seen a significant increase in the integration of machine learning concepts and algorithms. In this Perspective, we surveyed 179 open-source software projects, with corresponding peer-reviewed papers published within the last 5 years, to better understand the topics within the field being investigated by machine learning approaches. For each project, we provide a short description, the link to the code, the accompanying license type, and whether the training data and resulting models are made publicly available. Based on those deposited in GitHub repositories, the most popular employed Python libraries are identified. We hope that this survey will serve as a resource to learn about machine learning or specific architectures thereof by identifying accessible codes with accompanying papers on a topic basis. To this end, we also include computational chemistry open-source software for generating training data and fundamental Python libraries for machine learning. Based on our observations and considering the three pillars of collaborative machine learning work, open data, open source (code), and open models, we provide some suggestions to the community.

Design, synthesis, and computational studies of novel imidazo[1,2-a]pyrimidine derivatives as potential dual inhibitors of hACE2 and spike protein for blocking SARS-CoV-2 cell entry

In the present work, a new series of imidazo[1,2-a]pyrimidine Schiff base derivatives have been obtained using an easy and conventional synthetic route. The synthesized compounds were spectroscopically characterized using ¹H, ¹³C NMR, LC-MS(ESI), and FT-IR techniques. Green metric calculations indicate adherence to several green chemistry principles. The energy of Frontier Molecular Orbitals (FMO), Molecular Electrostatic Potential (MEP), Quantum Theory of Atoms in Molecules (QTAIM), and Reduced Density Gradient (RDG) were determined by the Density Functional Theory (DFT) method at B3LYP/6-31 G (d, p) as the basis set. Moreover, molecular docking studies targeting the human ACE2 and the spike, key entrance proteins of the severe acute respiratory syndrome coronavirus-2 were carried out along with hACE2 natural ligand Angiotensin II, the MLN-4760 inhibitor as well as the Cannabidiolic Acid CBDA which has been demonstrated to bind to the spike protein and block cell entry. The molecular modeling results showed auspicious results in terms of binding affinity as the top-scoring compound exhibited a remarkable affinity (-9.1 and -7.3 kcal/mol) to the ACE2 and spike protein respectively compared to CBDA (-5.7 kcal/mol), the MLN-4760 inhibitor (-7.3 kcal/mol), and angiotensin II (-9.2 kcal/mol). These findings suggest that the synthesized compounds may potentially act as effective entrance inhibitors, preventing the SARS-CoV-2 infection of human cells. Furthermore, in silico, ADMET, and drug-likeness prediction expressed promising drug-like characteristics.

Crystal structure of [1,2,4]triazolo[4,3-b]pyridazine derivatives as BRD4 bromodomain inhibitors and structure-activity relationship study

BRD4 contains two tandem bromodomains (BD1 and BD2) that recognize acetylated lysine for epigenetic reading, and these bromodomains are promising therapeutic targets for treating various diseases, including cancers. BRD4 is a well-studied target, and many chemical scaffolds for inhibitors have been developed. Research on the development of BRD4 inhibitors against various diseases is actively being conducted. Herein, we propose a series of [1,2,4]triazolo[4,3-b]pyridazine derivatives as bromodomain inhibitors with micromolar IC₅₀ values. We characterized the binding modes by determining the crystal structures of BD1 in complex with four selected inhibitors. Compounds containing [1,2,4] triazolo[4,3-b]pyridazine derivatives offer promising starting molecules for designing potent BRD4 BD inhibitors.

Experimental and computational studies of tautomerism pyridine carbonyl thiosemicarbazide derivatives

Tautomerism is one of the most important phenomena to consider when designing biologically active molecules. In this work, we use NMR spectroscopy, IR, and X-ray analysis as well as quantum-chemical calculations in the gas phase and in a solvent to study tautomerism of 1- (2-, 3- and 4-pyridinecarbonyl)-4-substituted thiosemicarbazide derivatives. The tautomer containing both carbonyl and thione groups turned out to be the most stable. The results of the calculations are consistent with the experimental data obtained from NMR and IR spectroscopy and with the crystalline forms from the X-ray studies. The obtained results broaden the knowledge in the field of structural studies of the thiosemicarbazide scaffold, which will translate into an understanding of the interactions of compounds with a potential molecular target.

Scalable Inhibitors of the Nsp3-Nsp4 Coupling in SARS-CoV-2

The human Betacoronavirus SARS-CoV-2 is a novel pathogen claiming millions of lives and causing a global pandemic that has disrupted international healthcare systems, economies, and communities. The virus is fast mutating and presenting more infectious but less lethal versions. Currently, some small-molecule therapeutics have received FDA emergency use authorization for the treatment of COVID-19, including Lagevrio (molnupiravir) and Paxlovid (nirmaltrevir/ritonavir), which target the RNA-dependent RNA polymerase and the 3CLpro main protease, respectively. Proteins downstream in the viral replication process, specifically the nonstructural proteins (Nsps1-16), are potential drug targets due to their crucial functions. Of these Nsps, Nsp4 is a particularly promising drug target due to its involvement in the SARS-CoV viral replication and double-membrane vesicle formation (mediated via interaction with Nsp3). Given the degree of sequence conservation of these two Nsps across the Betacoronavirus clade, their protein-protein interactions and functions are likely to be conserved as well in SARS-CoV-2. Through AlphaFold2 and its recent advancements, protein structures were generated of Nsp3 and 4 lumenal loops of interest. Then, using a combination of molecular docking suites and an existing library of lead-like compounds, we virtually screened 7 million ligands to identify five putative ligand inhibitors of Nsp4, which could present an alternative pharmaceutical approach against SARS-CoV-2. These ligands exhibit promising lead-like properties (ideal molecular weight and log P profiles), maintain fixed-Nsp4-ligand complexes in molecular dynamics (MD) simulations, and tightly associate with Nsp4 via hydrophobic interactions. Additionally, alternative peptide inhibitors based on Nsp3 were designed and shown in MD simulations to provide a highly stable binding to the Nsp4 protein. Finally, these therapeutics were attached to dendrimer structures to promote their multivalent binding with Nsp4, especially its large flexible luminal loop (Nsp4LLL). The therapeutics tested in this study represent many different approaches for targeting large flexible protein structures, especially those localized to the ER. This study is the first work targeting the membrane rearrangement system of viruses and will serve as a potential avenue for treating viruses with similar replicative function.

Spectroscopic and quantum chemical investigations to explore the effect of intermolecular interactions in a diuretic drug: Hydrochlorothiazide

Hydrochlorothiazide (HCTZ) being a diuretic drug widely used in anti-hypertensive therapy as it lowers the blood pressure by reducing the reabsorption of electrolytes in kidney resulting an increment of urine output and lowering the blood pressure. The purpose of the present work is to study the structural, vibrational and chemical properties of HCTZ based on its monomeric, dimeric and trimeric models by utilizing computational methods and experimental techniques. Density functional theory (DFT) with functional B3LYP and 6-311++G (d, p) basis set was used for a detailed computational study. Monomeric, dimeric and trimeric models of HCTZ have been studied for a better understanding of inter- and intramolecular hydrogen bonding. FT-IR (400-3800 cm^-1) and FT-Raman (100-3600 cm^-1) spectroscopy have been utilized for the characterization of HCTZ. The shifting in wavenumber of NH₂ and OSO group were observed in dimer and trimer due to the formation of intermolecular hydrogen bonding. Quantum theory of atoms in molecules (QTAIM) along with natural bond orbital (NBO) analysis were performed to examine the nature and strength of hydrogen bonding which showed that all the interactions were medium and partially covalent in nature; transition from LP(3)O15 → σ*(H46 → N32) and LP(3)O39 → σ*(H74 → N51) were responsible for the formation of O15•••H46 and O39•••H74 H-bonds, respectively. HOMO-LUMO energies predicted the chemical reactivity and stability of the molecule and the energy gap for dimer (4.6240 eV) and trimer (4.0493 eV) was found to be lesser than the monomer (5.0888 eV) which showed that the dimer and trimer have predicted more chemical reactivity in comparison to monomer. The most electronegative electrostatic potential was observed around the OSO group and the most electropositive potential around the amide group from MEPS analysis. Global as well as local reactivity descriptors have predicted the reactivity and hence, stability of the title molecule.

FP-GNN: a versatile deep learning architecture for enhanced molecular property prediction

Accurate prediction of molecular properties, such as physicochemical and bioactive properties, as well as ADME/T (absorption, distribution, metabolism, excretion and toxicity) properties, remains a fundamental challenge for molecular design, especially for drug design and discovery. In this study, we advanced a novel deep learning architecture, termed FP-GNN (fingerprints and graph neural networks), which combined and simultaneously learned information from molecular graphs and fingerprints for molecular property prediction. To evaluate the FP-GNN model, we conducted experiments on 13 public datasets, an unbiased LIT-PCBA dataset and 14 phenotypic screening datasets for breast cell lines. Extensive evaluation results showed that compared to advanced deep learning and conventional machine learning algorithms, the FP-GNN algorithm achieved state-of-the-art performance on these datasets. In addition, we analyzed the influence of different molecular fingerprints, and the effects of molecular graphs and molecular fingerprints on the performance of the FP-GNN model. Analysis of the anti-noise ability and interpretation ability also indicated that FP-GNN was competitive in real-world situations. Collectively, FP-GNN algorithm can assist chemists, biologists and pharmacists in predicting and discovering better molecules with desired functions or properties.

N-Cycloamino substituent effects on the packing architecture of ortho-sulfanilamide molecular crystals and their in silico carbonic anhydrase II and IX inhibitory activities

In the search for new `sulfa drugs' with therapeutic properties, o-nitrosulfonamides and N-cycloamino-o-sulfanilamides were synthesized and characterized using techniques including ¹H NMR, ¹³C NMR and FT-IR spectroscopy, and single-crystal X-ray diffraction (SC-XRD). The calculated density functional theory (DFT)-optimized geometry of the molecules showed similar conformations to those obtained by SC-XRD. Molecular docking of N-piperidinyl-o-sulfanilamide and N-indolinyl-o-sulfanilamide supports the notion that o-sulfanilamides are able to bind to human carbonic anhydrase II and IX inhibitors (hCA II and IX; PDB entries 4iwz and 5fl4). Hirshfeld surface analyses and DFT studies of three o-nitrosulfonamides {1-[(2-nitrophenyl)sulfonyl]pyrrolidine, C₁₀H₁₂N₂O₄S, 1, 1-[(2-nitrophenyl)sulfonyl]piperidine, C₁₁H₁₄N₂O₄S, 2, and 1-[(2-nitrophenyl)sulfonyl]-2,3-dihydro-1H-indole, C₁₄H₁₂N₂O₄S, 3} and three N-cycloamino-o-sulfanilamides [2-(pyrrolidine-1-sulfonyl)aniline, C₁₀H₁₄N₂O₂S, 4, 2-(piperidine-1-sulfonyl)aniline, C₁₁H₁₆N₂O₂S, 5, and 2-(2,3-dihydro-1H-indole-1-sulfonyl)aniline, C₁₄H₁₄N₂O₂S, 6] suggested that forces such as hydrogen bonding and π-π interactions hold molecules together and further showed that charge transfer could promote bioactivity and the ability to form biological interactions at the piperidinyl and phenyl moieties.

Cu(II)-Catalysed Hydrocarboxylation of Imines Utilizing CO2 to Synthesize α-Unsaturated Aminocarboxylic Acids

Here, we report the Cu(II)-photocatalysed hydrocarboxylation of imines (C=N) from a series of synthesized Schiff Base derivatives, namely (E)-1-(4-((4-methylbenzylidene)amino)phenyl)ethanone, (E)-1-(3-((5-bromo-2-hydroxybenzylidene)amino)phenyl)ethanone, (E)-4-((5-bromo-2-hydroxybenzylidene)amino)-1,5-dimethyl-2-phenyl-1H-pyrazol-3(2H)-one, and (E)-1,5-dimethyl-4-((4-methylbenzylidene)amino)-2-phenyl-1H-pyrazol-3(2H)-one, with carbon dioxide (CO₂) to generate disubstituted amino acids. Under mild conditions (atmospheric pressure of CO₂, room temperature, and 30 W Blue LED light), good to excellent yields confirming the formation of substituted amino acid unsaturated acid derivatives were obtained. Single crystal X-ray diffraction (SC-XRD) and UV-Vis diffuse reflectance spectroscopy (UV-Vis-DRS) confirmed the square pyramidal geometry of the Cu(II) photocatalyst. Docking and DFT calculations of the substituted amino acid unsaturated acid derivatives showed their potential as antimicrobial molecules.

Nucleophilicity Prediction Using Graph Neural Networks

The quantitative description between chemical reaction rates and nucleophilicity parameters plays a crucial role in organic chemistry. In this regard, the formula proposed by Mayr et al. and the constructed reactivity database are important representatives. However, the determination of Mayr's nucleophilicity parameter N often requires time-consuming experiments with reference electrophiles in the solvent. Several machine learning (ML)-based models have been proposed to realize the data-driven prediction of N in recent years. However, in addition to DFT-calculated electronic descriptors, most of them also use a set of artificially predefined structural descriptors as input, which may result in a biased representation of the nucleophile's structural information depending on descriptors' definition preference. Compared with traditional ML algorithms, graph neural networks (GNNs) can naturally take the molecule's structural information into account by applying the message passing technique. We herein proposed a SchNet-based GNN model that only takes the molecular conformation and solvent type as input. The model achieves a comparable performance to the previous benchmark study on 10-fold cross-validation of 894 data points (R² = 0.91, RMSE = 2.25). To enhance the model's ability to capture the molecule's electronic information, some DFT-calculated parameters are then incorporated into the model via graph global features, and substantial improvement is achieved in the prediction precision (R² = 0.95, RMSE = 1.63). These results demonstrate that both structural and electronic information are important for the prediction of N, and GNN can integrate these two kinds of information more effectively.

Modified Electrostatic Complementary Score Function and Its Application Boundary Exploration in Drug Design

In recent years, machine learning (ML) models have been found to quickly predict various molecular properties with accuracy comparable to high-level quantum chemistry methods. One such example is the calculation of electrostatic potential (ESP). Different ESP prediction ML models were proposed to generate surface molecular charge distribution. Electrostatic complementarity (EC) can apply ESP data to quantify the complementarity between a ligand and its binding pocket, leading to the potential to increase the efficiency of drug design. However, there is not much research discussing EC score functions and their applicability domain. We propose a new EC score function modified from the one originally developed by Bauer and Mackey, and confirm its effectiveness against the available Pearson's R correlation coefficient. Additionally, the applicability domain of the EC score and two indices used to define the EC score application scope will be discussed.

Investigation of Major Flavonoids from Artemisia argyi as a Potential COVID-19 Drug: Molecular Docking and DFT Calculations

Flavonoids from natural products are well-identified as potential antiviral agents in the treatment of SARS-CoV-2 (COVID-19) infection and related diseases. However, some major species of flavonoids from Chinese traditional folk medicine, such as of Artemisia argyi (A. argyi), have not been evaluated yet. Here, we choose five major flavonoids obtained from A. argyi, namely, Jaceosidin (1), Eupatilin (2), Apigenin (3), Eupafolin (4), and 5,6-Dihydroxy-7,3′,4′-trimethoxyflavone (5), compared to the well-studied Baicalein (6), as potential inhibitors analogs for COVID-19 by computational modeling strategies. The frontier molecular orbitals (FMOs), chemical reactivity descriptors, and electrostatic surface potential (ESP) were performed by density functional theory (DFT) calculations. Additionally, these flavonoids were docked on the main protease (PDB: 6LU7) of SARS-CoV-2 to evaluate the binding affinities. Computational analysis predicted that all of these compounds show a high affinity and might serve as potential inhibitors to SARS-CoV-2, among which compound (5) exhibits the least binding energy (−155.226 kcal/mol). The high binding affinity could be enhanced by increasing the electron repulsion due to the valence shell electron pair repulsion model (VSEPR). Consequently, the major flavonoids in Artemisia argyi have a significant ability to reduce the deterioration of COVID-19 in the terms of DFT calculations and molecular docking.

The emerging potential of interactive virtual reality in drug discovery

INTRODUCTION

The potential of virtual reality (VR) to contribute to drug design and development has been recognized for many years. A recent advance is to use VR not only to visualize and interact with molecules, but also to interact with molecular dynamics simulations 'on the fly' (interactive molecular dynamics in VR, IMD-VR), which is useful for flexible docking and examining binding processes and conformational changes.

AREAS COVERED

The authors use the term 'interactive VR' to refer to software where interactivity is an inherent part of the user VR experience e.g. in making structural modifications or interacting with a physically rigorous molecular dynamics (MD) simulation, as opposed to using VR controllers to rotate and translate the molecule for enhanced visualization. Here, they describe these methods and their application to problems relevant to drug discovery, highlighting the possibilities that they offer in this arena.

EXPERT OPINION

The ease of viewing and manipulating molecular structures and dynamics, using accessible VR hardware, and the ability to modify structures on the fly (e.g. adding or deleting atoms) - and for groups of researchers to work together in the same virtual environment - makes modern interactive VR a valuable tool to add to the armory of drug design and development methods.

Virtual screening of PEBP1 inhibitors by combining 2D/3D-QSAR analysis, hologram QSAR, homology modeling, molecular docking analysis, and molecular dynamic simulations

Human phosphatidylethanolamine binding protein 1 (hPEBP1) is a novel target affecting many cellular signaling pathways involved in the formation of metastases. It can be used in the treatment of many cases of cancer. For these reasons, pharmaceutical companies use computational approaches, including multi-QSAR (2D, 3D, and hologram QSAR) analysis, homology modeling, molecular docking analysis, and molecular dynamic simulations, to speed up the drug discovery process. In this paper, QSAR modeling was conducted using two quantum chemistry optimization methods (AM1 and DFT levels). As per PLS results, we found that the DFT/B3LYP method presents high predictability according to 2D-QSAR, CoMFA, CoMSIA, and hologram QSAR studies, with Q² of 0.81, 0.67, 0.79, and 0.67, and external power with R²_pred of 0.78, 0.58, 0.66, and 0.56, respectively. This result has been validated by CoMFA/CoMSIA graphics, which suggests that electrostatic fields combined with hydrogen bond donor/acceptor fields are beneficial to the antiproliferative activity. While the hologram QSAR models show the contributions of each fragment in improving the activity. The results from QSAR analyses revealed that ursolic acids with heterocyclic rings could improve the activities. Ramachandran plot validated the modeled PEBP1 protein. Molecular docking and MD simulations revealed that the hydrophobic and hydrogen bond interactions are dominant in the PEBP1's pocket. These results were used to predict in silico structures of three new compounds with potential anticancer activity. Similar molecular docking stability studies and molecular dynamics simulations were conducted.

Electrostatic Complementarity in Structure-Based Drug Design

Optimization of electrostatic complementarity is an important strategy in structure-based drug discovery for improving the affinity of molecules against a specific protein target. In this Miniperspective we identify examples where deliberate optimization of protein-ligand electrostatic complementarity or intramolecular electrostatic interactions gave improvements in target affinity (up to 250-fold), physicochemical properties, in vitro properties, and off-target selectivity. We also look retrospectively at a series of factor Xa inhibitors that show an almost 8000-fold range in potency that can be correlated with the calculated electrostatic potential (ESP) surfaces. Recent developments using a graph-convolutional deep neural network to rapidly generate high quality ESP surfaces have the potential to make this useful tool more accessible for a wider audience within the field of medicinal chemistry.

Natural Compound ZINC12899676 Reduces Porcine Epidemic Diarrhea Virus Replication by Inhibiting the Viral NTPase Activity

Porcine epidemic diarrhea virus (PEDV) is an alphacoronavirus (α-CoV) that causes high mortality in suckling piglets, leading to severe economic losses worldwide. No effective vaccine or commercial antiviral drug is readily available. Several replicative enzymes are responsible for coronavirus replication. In this study, the potential candidates targeting replicative enzymes (PLP2, 3CLpro, RdRp, NTPase, and NendoU) were screened from 187,119 compounds in ZINC natural products library, and seven compounds had high binding potential to NTPase and showed drug-like property. Among them, ZINC12899676 was identified to significantly inhibit the NTPase activity of PEDV by targeting its active pocket and causing its conformational change, and ZINC12899676 significantly inhibited PEDV replication in IPEC-J2 cells. It first demonstrated that ZINC12899676 inhibits PEDV replication by targeting NTPase, and then, NTPase may serve as a novel target for anti-PEDV.

Knowledge-based BERT: a method to extract molecular features such as computational chemists

Molecular property prediction models based on machine learning algorithms have become important tools to triage unpromising lead molecules in the early stages of drug discovery. Compared with the mainstream descriptor- and graph-based methods for molecular property predictions, SMILES-based methods can directly extract molecular features from SMILES without human expert knowledge, but they require more powerful algorithms for feature extraction and a larger amount of data for training, which makes SMILES-based methods less popular. Here, we show the great potential of pre-training in promoting the predictions of important pharmaceutical properties. By utilizing three pre-training tasks based on atom feature prediction, molecular feature prediction and contrastive learning, a new pre-training method K-BERT, which can extract chemical information from SMILES like chemists, was developed. The calculation results on 15 pharmaceutical datasets show that K-BERT outperforms well-established descriptor-based (XGBoost) and graph-based (Attentive FP and HRGCN+) models. In addition, we found that the contrastive learning pre-training task enables K-BERT to 'understand' SMILES not limited to canonical SMILES. Moreover, the general fingerprints K-BERT-FP generated by K-BERT exhibit comparative predictive power to MACCS on 15 pharmaceutical datasets and can also capture molecular size and chirality information that traditional binary fingerprints cannot capture. Our results illustrate the great potential of K-BERT in the practical applications of molecular property predictions in drug discovery.

Experimental and Computational Structural Studies of 2,3,5-Trisubstituted and 1,2,3,5-Tetrasubstituted Indoles as Non-Competitive Antagonists of GluK1/GluK2 Receptors

The blockade of kainate receptors, in particular with non-competitive antagonists, has—due to their anticonvulsant and neuroprotective properties—therapeutic potential in many central nervous system (CNS) diseases. Deciphering the structural properties of kainate receptor ligands is crucial to designing medicinal compounds that better fit the receptor binding pockets. In light of that fact, here, we report experimental and computational structural studies of four indole derivatives that are non-competitive antagonists of GluK1/GluK2 receptors. We used X-ray studies and Hirshfeld surface analysis to determine the structure of the compounds in the solid state and quantum chemical calculations to compute HOMO and LUMO orbitals and the electrostatic potential. Moreover, non-covalent interaction maps were also calculated. It is worth emphasizing that compounds 3 and 4 are achiral molecules crystallising in non-centrosymmetric space groups, which is a relatively rare phenomenon.

Exploration of plant-derived natural polyphenols toward COVID-19 main protease inhibitors: DFT, molecular docking approach, and molecular dynamics simulations

Recent outbreaks of coronavirus have brought serious challenges to public health around the world, and it is essential to find effective treatments. In this study, the 3C-like proteinase (3CLpro) of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) has been considered as an important drug target because of its role in viral replication. We initially optimized 251 compounds at the PM7 level of theory for docking with 3CLpro, and then we selected the top 12 compounds for further optimization with the B3LYP-D3/6-311G** method and obtained the top four compounds by further molecular docking. Quantum chemistry calculations were performed to predict molecular properties, such as the electrostatic potential and some CDFT descriptors. We also performed molecular dynamics simulations and free energy calculations to determine the relative stability of the selected four potential compounds. We have identified key residues controlling the 3CLpro/ligand binding from per-residue based decomposition of the binding free energy. Convincingly, the comprehensive results support the conclusion that the compounds have the potential to become a candidate for anti-coronavirus treatment.

The combination of molecular dynamics simulations and quantitative calculations as a powerful tool for screening molecules.

DeepNCI: DFT Noncovalent Interaction Correction with Transferable Multimodal Three-Dimensional Convolutional Neural Networks

A multimodal deep learning model, DeepNCI, is proposed for improving noncovalent interactions (NCIs) calculated via density functional theory (DFT). DeepNCI is composed of a three-dimensional convolutional neural network (3D CNN) for abstracting critical and comprehensive features from 3D electron density, and a neural network for modeling one-dimensional quantum chemical properties. By merging features from two networks, DeepNCI is able to reduce the root-mean-square error of DFT-calculated NCI from 1.19 kcal/mol to ∼0.2 kcal/mol for a NCI molecular database (>1000 molecules). The representativeness of the joint features can be visualized by t-distributed stochastic neighbor embedding (t-SNE), where they can distinguish categorized NCI systems quite well. Therefore, the fused model performs better than its component networks. In addition, the 3D CNN takes electron density as inputs that are in the same range, despite the size of molecular systems, so it can promote model applicability and transferability. To clarify the applicability of DeepNCI, an application domain (AD) has been defined with merged features using the K-nearest-neighbor method. The calculations for external test sets are shown that AD can properly monitor the reliability for a prediction. The model transferability is tested with a small database of homolysis bond dissociation energy including only dozens of samples. With NCI database pretrained parameters, the same or better performance than the reported results is achieved by transfer learning. This suggests that the DeepNCI model is transferable and it may transfer to other relative tasks, which possibly can resolve some small sampling problems. The source code of DeepNCI can be freely accessed at https://github.com/wenzelee/DeepNCI.

Structure-activity relationship around PI-2620 highlights the importance of the nitrogen atom position in the tricyclic core

Tau aggregates represent a critical pathology in Alzheimer's disease (AD) and other forms of dementia. The extent of Tau neurofibrillary tangles across defined brain regions corresponds well to the observed level of cognitive decline in AD. Compound 1 (PI-2620) was recently identified as a promising Tau positron emission tomography tracer for AD and non-AD tauopathies. To evaluate the impact of the N-atom position with respect to Tau- and off-target binding, tricyclic core analogs of PI-2620 with nitrogen atoms at different positions were prepared. Affinity to aggregated Tau was evaluated using human AD brain homogenates, and their off-target binding was evaluated in a monoamine oxidase A (MAO-A) competition assay. The novel tricyclic core derivatives all displayed inferior Tau binding or MAO-A off-target selectivity, indicating PI-2620 to be the optimal design for high affinity binding to Tau and high MAO-A selectivity.

Synthesis and fungicidal activity of novel benzimidazole derivatives bearing pyrimidine-thioether moiety against Botrytis cinerea

BACKGROUND

Botrytis cinerea is a serious plant fungus and strongly affects the yield and quality of crops. The main control strategy is the employment of fungicides. To research for efficient fungicide with novel structure, a series of novel benzimidazole derivatives bearing pyrimidine and thioether moieties were designed and synthesized.

RESULTS

Some target compounds such as 4h, 4i, 4k, 4l, 4m, 4s, 4t and 4u exhibited notable fungicidal activities, with half maximal effective concentration (EC₅₀ ) values in the range 0.13-0.24 μg mL^-1 , which means that their activities were comparable or higher than that of carbendazim (EC₅₀ = 0.21 μg mL^-1 ). Among them, N-(4-fluorophenyl)-2-((4-(1H-benzimidazol-2-yl)-6-(4-methoxyphenyl) pyrimidin-2-yl)thio)acetamide (4m) displayed the best activity (EC₅₀ = 0.13 μg mL^-1 ). Molecular electrostatic potential analysis of 4m elucidated that the NH moiety of benzimidazole ring was located in the positive potential region and may generate hydrogen bond with target amino acid residue. Molecular docking analysis revealed that there was one hydrogen bond and one 𝜋-𝜋 interaction between 4m and target protein.

CONCLUSIONS

This study demonstrated that the benzimidazole derivatives bearing pyrimidine and thioether moieties can be further optimized as a lead compound for the control of B. cinerea. The combination of molecular electrostatic potential and molecular docking analyses may provide a valuable reference for studying the interaction between the ligand and target protein. © 2021 Society of Chemical Industry.

Protein C-GeM: A Coarse-Grained Electron Model for Fast and Accurate Protein Electrostatics Prediction

The electrostatic potential (ESP) is a powerful property for understanding and predicting electrostatic charge distributions that drive interactions between molecules. In this study, we compare various charge partitioning schemes including fitted charges, density-based quantum mechanical (QM) partitioning schemes, charge equilibration methods, and our recently introduced coarse-grained electron model, C-GeM, to describe the ESP for protein systems. When benchmarked against high quality density functional theory calculations of the ESP for tripeptides and the crambin protein, we find that the C-GeM model is of comparable accuracy to ab initio charge partitioning methods, but with orders of magnitude improvement in computational efficiency since it does not require either the electron density or the electrostatic potential as input.

Synthesis and fungicidal activity of novel 2-(2-alkylthio-6-phenylpyrimidin-4-yl)-1H-benzimidazoles

With the aim of exploring new benzimidazole derivative with high fungicidal activity, a series of novel 2-(2-(alkylthio)-6-phenylpyrimidin-4-yl)-1H-benzimidazoles were designed and synthesized, and their in vitro fungicidal activities were evaluated. Compounds 5a, 5f, 5g, 5h, 5i and 5l exhibited excellent fungicidal activities against Botrytis cinerea, and 5c, 5f, 5h, 5i and 5l displayed notable fungicidal activities against Sclerotinia sclerotiorum. Among them, compound 5i (R¹ = fluorine, R² = benzyl) displayed the best activity towards the two tested fungi. Docking study of 5i with β-tubulin protein revealed that the NH moiety of benzimidazole ring generated a hydrogen bond with Gln-11 residue, and the fluorine atom of benzene ring formed a hydrogen bond with Tyr-208 residue, respectively; the benzene ring of Tyr-222 and the pyrimidine ring of 5i yielded a π-π interaction. The molecular electrostatic potential analysis elucidated the nitrogen atom of benzimidazole ring, fluorine atom of benzene ring and sulfur atom of thioether moiety were located in the negative potential regions, whereas some hydrogen atoms of benzene, benzimidazole and pyrimidine rings were located in the positive potential regions. This analysis demonstrated the reason why 5i can form hydrogen bonds with amino acid residues of target protein.

Graph Neural Networks and Their Current Applications in Bioinformatics

Graph neural networks (GNNs), as a branch of deep learning in non-Euclidean space, perform particularly well in various tasks that process graph structure data. With the rapid accumulation of biological network data, GNNs have also become an important tool in bioinformatics. In this research, a systematic survey of GNNs and their advances in bioinformatics is presented from multiple perspectives. We first introduce some commonly used GNN models and their basic principles. Then, three representative tasks are proposed based on the three levels of structural information that can be learned by GNNs: node classification, link prediction, and graph generation. Meanwhile, according to the specific applications for various omics data, we categorize and discuss the related studies in three aspects: disease prediction, drug discovery, and biomedical imaging. Based on the analysis, we provide an outlook on the shortcomings of current studies and point out their developing prospect. Although GNNs have achieved excellent results in many biological tasks at present, they still face challenges in terms of low-quality data processing, methodology, and interpretability and have a long road ahead. We believe that GNNs are potentially an excellent method that solves various biological problems in bioinformatics research.

Molecular Characterisation of Soybean Osmotins and Their Involvement in Drought Stress Response

Osmotins are multifunctional proteins belonging to the thaumatin-like family related to plant stress responses. To better understand the functions of soybean osmotins in drought stress response, the current study presents the characterisation of four previously described proteins and a novel putative soybean osmotin (GmOLPa-like). Gene and protein structure as well as gene expression analyses were conducted on different tissues and developmental stages of two soybean cultivars with varying dehydration sensitivities (BR16 and EMB48 are highly and slightly sensitive, respectively). The analysed osmotin sequences share the conserved amino acid signature and 3D structure of the thaumatin-like family. Some differences were observed in the conserved regions of protein sequences and in the electrostatic surface potential. P21-like present the most similar electrostatic potential to osmotins previously characterised as promoters of drought tolerance in Nicotiana tabacum and Solanum nigrum. Gene expression analysis indicated that soybean osmotins were differentially expressed in different organs (leaves and roots), developmental stages (R1 and V3), and cultivars in response to dehydration. In addition, under dehydration conditions, the highest level of gene expression was detected for GmOLPa-like and P21-like osmotins in the leaves and roots, respectively, of the less drought sensitive cultivar. Altogether, the results suggest an involvement of these genes in drought stress tolerance.

Molecular dynamics, quantum mechanics and docking studies of some Keap1 inhibitors - An insight into the atomistic mechanisms of their antioxidant potential

Inhibitors of Keap1 would disrupt the covalent interaction between Keap1 and Nrf2 to unleash Nrf2 transcriptional machinery that orchestrates its cellular antioxidant, cytoprotective and detoxification processes thereby, protecting the cells against oxidative stress mediated diseases. In this in silico research, we investigated the Keap1 inhibiting potential of fifty (50) antioxidants using pharmacokinetic ADMET profiling, bioactivity assessment, physicochemical studies, molecular docking investigation, molecular dynamics and Quantum mechanical-based Density Functional Theory (DFT) studies using Keap1 as the apoprotein control. Out of these 50 antioxidants, Maslinic acid (MASA), 18-alpha-glycyrrhetinic acid (18-AGA) and resveratrol stand out by passing the RO5 (Lipinski rule of 5) for the physicochemical properties and ADMET studies. These three compounds also show high binding affinity of -10.6 kJ/mol, -10.4 kJ/mol and -7.8 kJ/mol at the kelch pocket of Keap1 respectively. Analysis of the 20ns trajectories using RMSD, RMSF, ROG and h-bond parameters revealed the stability of these compounds after comparing them with Keap1 apoprotein. Furthermore, the electron donating and accepting potentials of these compounds was used to investigate their reactivity using Density Functional Theory (HOMO and LUMO) and it was revealed that resveratrol had the highest stability based on its low energy gap. Our results predict that the three compounds are potential drug candidates with domiciled therapeutic functions against oxidative stress-mediated diseases. However, resveratrol stands out as the compound with the best stability and therefore, could be the best candidate with the best therapeutic efficacy.

Structural insight from intermolecular interactions and energy framework analyses of 2-substituted 6,7,8,9-tetrahydro-11H-pyrido[2,1-b]quinazolin-11-ones

The crystal structures of three mackinazolinone derivatives (2-amino-6,7,8,9-tetrahydro-11H-pyrido[2,1-b]quinazolin-11-one at room temperature, and 2-nitro-6,7,8,9-tetrahydro-11H-pyrido[2,1-b]quinazolin-11-one and N-(11-oxo-6,8,9,11-tetrahydro-7H-pyrido[2,1-b]quinazolin-2-yl)benzamide at 100 K) are explored using X-ray crystallography. To delineate the different intermolecular interactions and the respective interaction energies in the crystal architectures, energy framework analyses were carried out using the CE-B3LYP/6-31G(d,p) method implemented in the CrystalExplorer software. In the structures the different molecules are linked by C-H...O, C-H...N and N-H...O hydrogen bonds. Together with these hydrogen bonds, C-H...π and C-O...π interactions are involved in the formation of a three-dimensional crystal network. A Hirshfeld surface analysis allows the visualization of the two-dimensional fingerprint plots and the quantification of the contributions of H...H, H...C/C...H and H...O/O...H contacts throughout the different crystal structures. To obtain additional information on the intrinsic properties of our targets and to compare the experimental crystal structures with their respective conformations in the gas phase, quantum chemical calculations at the B3LYP-D3BJ/6-311++G(d,p) level of theory, including Grimme's D3 correction term and BJ damping functions, were carried out to account for intramolecular dispersion interactions. The identified energy gaps between the highest occupied and the lowest unoccupied molecular orbitals (HOMO-LUMO gap) of our targets in the gas phase and in two implicit solvents (methanol and dimethyl sulfoxide) allow us to quantify the impact of different substituents on the reactivity of mackinazolinone derivatives.

Machine learning techniques applied to the drug design and discovery of new antivirals: a brief look over the past decade

Introduction: Drug design and discovery of new antivirals will always be extremely important in medicinal chemistry, taking into account known and new viral diseases that are yet to come. Although machine learning (ML) have shown to improve predictions on the biological potential of chemicals and accelerate the discovery of drugs over the past decade, new methods and their combinations have improved their performance and established promising perspectives regarding ML in the search for new antivirals.Areas covered: The authors consider some interesting areas that deal with different ML techniques applied to antivirals. Recent innovative studies on ML and antivirals were selected and analyzed in detail. Also, the authors provide a brief look at the past to the present to detect advances and bottlenecks in the area.Expert opinion: From classical ML techniques, it was possible to boost the searches for antivirals. However, from the emergence of new algorithms and the improvement in old approaches, promising results will be achieved every day, as we have observed in the case of SARS-CoV-2. Recent experience has shown that it is possible to use ML to discover new antiviral candidates from virtual screening and drug repurposing.

Hyperbolic relational graph convolution networks plus: a simple but highly efficient QSAR-modeling method

Accurate predictions of druggability and bioactivities of compounds are desirable to reduce the high cost and time of drug discovery. After more than five decades of continuing developments, quantitative structure-activity relationship (QSAR) methods have been established as indispensable tools that facilitate fast, reliable and affordable assessments of physicochemical and biological properties of compounds in drug-discovery programs. Currently, there are mainly two types of QSAR methods, descriptor-based methods and graph-based methods. The former is developed based on predefined molecular descriptors, whereas the latter is developed based on simple atomic and bond information. In this study, we presented a simple but highly efficient modeling method by combining molecular graphs and molecular descriptors as the input of a modified graph neural network, called hyperbolic relational graph convolution network plus (HRGCN+). The evaluation results show that HRGCN+ achieves state-of-the-art performance on 11 drug-discovery-related datasets. We also explored the impact of the addition of traditional molecular descriptors on the predictions of graph-based methods, and found that the addition of molecular descriptors can indeed boost the predictive power of graph-based methods. The results also highlight the strong anti-noise capability of our method. In addition, our method provides a way to interpret models at both the atom and descriptor levels, which can help medicinal chemists extract hidden information from complex datasets. We also offer an HRGCN+'s online prediction service at https://quantum.tencent.com/hrgcn/.

Screening Ionic Liquids Based on Ionic Volume and Electrostatic Potential Analyses

Ionic liquids (ILs) are known to have tunable solvation properties, based on the pairing of different anions and cations, but the compositional landscape is vast and challenging to navigate efficiently. Some computational screening protocols are available, but they can be either time-consuming or difficult to implement. Herein, we perform a detailed investigation of the fundamental role of electrostatic interactions in these systems. We effectively develop a bridge between the previous volume-based approach with a quantum structure-property relationship approach to create fast, simple screening guidelines. We propose a new parameter that is applicable to both monovalent and multivalent ions, the ionic polarity index (IPI), which is defined as the ratio of the average electrostatic surface potential (V̅) of the ion to the net charge of the ion (q). The IPI correlation has been tested on a diverse data set of 121 ions, and reliable predictions can be obtained within a homologous series of IL compounds.

Physical stability enhancement and antimicrobial properties of a sodium ionic cocrystal with theophylline

In the present study, we have described the synthesis and characterisation of the theophylline hydrate (Theo hydrate), cocrystal (Theo–Phen·2H₂O) and hydrated sodium co-crystal of theophylline (Na–(Theo)₂ClO·2H₂O), where Theo = theophylline and Phen = 1,10-phenathroline.

Machine Learning Methods in Drug Discovery

The advancements of information technology and related processing techniques have created a fertile base for progress in many scientific fields and industries. In the fields of drug discovery and development, machine learning techniques have been used for the development of novel drug candidates. The methods for designing drug targets and novel drug discovery now routinely combine machine learning and deep learning algorithms to enhance the efficiency, efficacy, and quality of developed outputs. The generation and incorporation of big data, through technologies such as high-throughput screening and high through-put computational analysis of databases used for both lead and target discovery, has increased the reliability of the machine learning and deep learning incorporated techniques. The use of these virtual screening and encompassing online information has also been highlighted in developing lead synthesis pathways. In this review, machine learning and deep learning algorithms utilized in drug discovery and associated techniques will be discussed. The applications that produce promising results and methods will be reviewed.