In silico Prediction of Chemical Ames Mutagenicity

Ligand-based analysis of the antifungal potential of phytosterols and triterpenes isolated from Cryptostegia grandiflora against Candida auris FKBP12

Candida auris, a pathogenic fungus, has posed significant challenges to conventional medical treatments due to its increasing resistance to antifungal agents. Consequently, due to their promising pharmacological properties, there is a compelling interest in exploring novel bioactive compounds, such as phytosterols and triterpenes. This study aimed to conduct virtual screening utilizing computational methods, including ADMET, molecular docking, and molecular dynamics, to assess the activity and feasibility of phytosterols extracted from Cryptostegia grandiflora as potential therapeutic agents. Computational predictions suggest that compounds bearing structural similarities to Fsp3-rich molecules hold promise for inhibiting enzymes and G protein-coupled receptor (GPCR) modulators, with particular emphasis on ursolic acid, which, in its conjugated form, exhibits high oral bioavailability and metabolic stability, rendering it a compelling drug candidate. Molecular docking calculations identified ursolic acid and stigmasterol as promising ligands. While stigmasterol displayed superior affinity during molecular dynamics simulations, it exhibited instability, contrasting with ursolic acid's slightly lower affinity yet sustained stability throughout the dynamic assessments. This suggests that ursolic acid is a robust candidate for inhibiting the FKBP12 isomerase in C. auris. Moreover, further investigations could focus on experimentally validating the molecular docking predictions and evaluating the efficacy of ursolic acid as an FKBP12 isomerase inhibitor in models of C. auris infection.

ToxMPNN: A deep learning model for small molecule toxicity prediction

Machine learning (ML) has shown a great promise in predicting toxicity of small molecules. However, the availability of data for such predictions is often limited. Because of the unsatisfactory performance of models trained on a single toxicity endpoint, we collected toxic small molecules with multiple toxicity endpoints from previous study. The dataset comprises 27 toxic endpoints categorized into seven toxicity classes, namely, carcinogenicity and mutagenicity, acute oral toxicity, respiratory toxicity, irritation and corrosion, cardiotoxicity, CYP450, and endocrine disruption. In addition, a binary classification Common-Toxicity task was added based on the aforementioned dataset. To improve the performance of the models, we added marketed drugs as negative samples. This study presents a toxicity predictive model, ToxMPNN, based on the message passing neural network (MPNN) architecture, aiming to predict the toxicity of small molecules. The results demonstrate that ToxMPNN outperforms other models in capturing toxic features within the molecular structure, resulting in more precise predictions with the ROC_AUC testing score of 0.886 for the Toxicity_drug dataset. Furthermore, it was observed that adding marketed drugs as negative samples not only improves the predictive performance of the binary classification Common-Toxicity task but also enhances the stability of the model prediction. It shows that the graph-based deep learning (DL) algorithms in this study can be used as a trustworthy and effective tool to assess small molecule toxicity in the development of new drugs.

Reducing overconfident errors in molecular property classification using Posterior Network

Deep-learning-based classification models are increasingly used for predicting molecular properties in drug development. However, traditional classification models using the Softmax function often give overconfident mispredictions for out-of-distribution samples, highlighting a critical lack of accurate uncertainty estimation. Such limitations can result in substantial costs and should be avoided during drug development. Inspired by advances in evidential deep learning and Posterior Network, we replaced the Softmax function with a normalizing flow to enhance the uncertainty estimation ability of the model in molecular property classification. The proposed strategy was evaluated across diverse scenarios, including simulated experiments based on a synthetic dataset, ADMET predictions, and ligand-based virtual screening. The results demonstrate that compared with the vanilla model, the proposed strategy effectively alleviates the problem of giving overconfident but incorrect predictions. Our findings support the promising application of evidential deep learning in drug development and offer a valuable framework for further research.

AMPred-CNN: Ames mutagenicity prediction model based on convolutional neural networks

Mutagenicity assessment plays a pivotal role in the safety evaluation of chemicals, pharmaceuticals, and environmental compounds. In recent years, the development of robust computational models for predicting chemical mutagenicity has gained significant attention, driven by the need for efficient and cost-effective toxicity assessments. In this paper, we proposed AMPred-CNN, an innovative Ames mutagenicity prediction model based on Convolutional Neural Networks (CNNs), uniquely employing molecular structures as images to leverage CNNs' powerful feature extraction capabilities. The study employs the widely used benchmark mutagenicity dataset from Hansen et al. for model development and evaluation. Comparative analyses with traditional ML models on different molecular features reveal substantial performance enhancements. AMPred-CNN outshines these models, demonstrating superior accuracy, AUC, F1 score, MCC, sensitivity, and specificity on the test set. Notably, AMPred-CNN is further benchmarked against seven recent ML and DL models, consistently showcasing superior performance with an impressive AUC of 0.954. Our study highlights the effectiveness of CNNs in advancing mutagenicity prediction, paving the way for broader applications in toxicology and drug development.

Exploring molecular fingerprints of different drugs having bile interaction: a stepping stone towards better drug delivery

Bile acids are amphiphilic substances produced naturally in humans. In the context of drug delivery and dosage form design, it is critical to understand whether a drug interacts with bile inside the gastrointestinal (GI) tract or not. This study focuses on the identification of structural fingerprints/features important for bile interaction. Molecular modelling methods such as Bayesian classification and recursive partitioning (RP) studies are executed to find important fingerprints/features for the bile interaction. For the Bayesian classification study, the ROC score of 0.837 and 0.950 are found for the training set and the test set compounds, respectively. The fluorine-containing aliphatic/aromatic group, the branched chain of the alkyl group containing hydroxyl moiety and the phenothiazine ring etc. are identified as good fingerprints having a positive contribution towards bile interactions, whereas, the bad fingerprints such as free carboxylate group, purine, and pyrimidine ring etc. have a negative contribution towards bile interactions. The best tree (tree ID: 1) from the RP study classifies the bile interacting or non-interacting compounds with a ROC score of 0.941 for the training and 0.875 for the test set. Additionally, SARpy and QSAR-Co analyses are also been performed to classify compounds as bile interacting/non-interacting. Moreover, forty-six recently FDA-approved drugs have been screened by the developed SARpy and QSAR-Co models to assess their bile interaction properties. Overall, this attempt may facilitate the researchers to identify bile interacting/non-interacting molecules in a faster way and help in the design of formulations and target-specific drug development.

Development of scoring-assisted generative exploration (SAGE) and its application to dual inhibitor design for acetylcholinesterase and monoamine oxidase B

De novo molecular design is the process of searching chemical space for drug-like molecules with desired properties, and deep learning has been recognized as a promising solution. In this study, I developed an effective computational method called Scoring-Assisted Generative Exploration (SAGE) to enhance chemical diversity and property optimization through virtual synthesis simulation, the generation of bridged bicyclic rings, and multiple scoring models for drug-likeness. In six protein targets, SAGE generated molecules with high scores within reasonable numbers of steps by optimizing target specificity without a constraint and even with multiple constraints such as synthetic accessibility, solubility, and metabolic stability. Furthermore, I suggested a top-ranked molecule with SAGE as dual inhibitors of acetylcholinesterase and monoamine oxidase B through multiple desired property optimization. Therefore, SAGE can generate molecules with desired properties by optimizing multiple properties simultaneously, indicating the importance of de novo design strategies in the future of drug discovery and development. SCIENTIFIC CONTRIBUTION: The scientific contribution of this study lies in the development of the Scoring-Assisted Generative Exploration (SAGE) method, a novel computational approach that significantly enhances de novo molecular design. SAGE uniquely integrates virtual synthesis simulation, the generation of complex bridged bicyclic rings, and multiple scoring models to optimize drug-like properties comprehensively. By efficiently generating molecules that meet a broad spectrum of pharmacological criteria-including target specificity, synthetic accessibility, solubility, and metabolic stability-within a reasonable number of steps, SAGE represents a substantial advancement over traditional methods. Additionally, the application of SAGE to discover dual inhibitors for acetylcholinesterase and monoamine oxidase B not only demonstrates its potential to streamline and enhance the drug development process but also highlights its capacity to create more effective and precisely targeted therapies. This study emphasizes the critical and evolving role of de novo design strategies in reshaping the future of drug discovery and development, providing promising avenues for innovative therapeutic discoveries.

Sustainable methods for the carboxymethylation and methylation of ursolic acid with dimethyl carbonate under mild and acidic conditions

Ursolic acid is a triterpene plant extract that exhibits significant potential as an anti-cancer, anti-tumour, and anti-inflammatory agent. Its direct use in the pharmaceutical industry is hampered by poor uptake of ursolic acid in the human body coupled with rapid metabolism causing a decrease in bioactivity. Modification of ursolic acid can overcome such issues, however, use of toxic reagents, unsustainable synthetic routes and poor reaction metrics have limited its potential. Herein, we demonstrate the first reported carboxymethylation and/or methylation of ursolic acid with dimethyl carbonate (DMC) as a green solvent and sustainable reagent under acidic conditions. The reaction of DMC with ursolic acid, in the presence of PTSA, ZnCl2, or H2SO4-SiO2 yielded the carboxymethylation product 3β-[[methoxy]carbonyl]oxyurs-12-en-28-oic acid, the methylation product 3β-methoxyurs-12-en-28-oic acid and the dehydration product urs-2,12-dien-28-oic acid. PTSA demonstrated high conversion and selectivity towards the previously unreported carboxymethylation of ursolic acid, while the application of formic acid in the system led to formylation of ursolic acid (3β-formylurs-12-en-28-oic acid) in quantitative yields via esterification, with DMC acting solely as a solvent. Meanwhile, the methylation product of ursolic acid, 3β-methoxyurs-12-en-28-oic acid, was successfully synthesised with FeCl3, demonstrating exceptional conversion and selectivity, >99% and 99%, respectively. Confirmed with the use of qualitative and quantitative green metrics, this result represents a significant improvement in conversion, selectivity, safety, and sustainability over previously reported methods of ursolic acid modification. It was demonstrated that these methods could be applied to other triterpenoids, including corosolic acid. The study also explored the potential pharmaceutical applications of ursolic acid, corosolic acid, and their derivatives, particularly in anti-inflammatory, anti-cancer, and anti-tumour treatments, using molecular ADMET and docking methods. The methods developed in this work have led to the synthesis of novel molecules, thus creating opportunities for the future investigation of biological activity and the modification of a wide range of triterpenoids applying acidic DMC systems to deliver novel active pharmaceutical intermediates.

PTSA-induced synthesis, in silico and nano study of novel ethylquinolin-thiazolo-triazole in cervical cancer

Aim: p-Toluenesulfonic acid-(PTSA) and grinding-induced novel synthesis of ethylquinolin-thiazolo-triazole derivatives was performed using green chemistry. Materials & methods: Development of a nanoconjugate drug-delivery system of ethylquinolin-thiazolo-triazole was carried out with D-α-tocopheryl polyethylene glycol succinate (TPGS) and the formulation was further characterized by transmission electron microscopy, atomic force microscopy, dynamic light scattering and in vitro drug release assay. The effect of 3a nanoparticles was assessed against a cervical cancer cell line (HeLa) through the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay, and the effect on apoptosis was determined. Results & discussion: The 3a nanoparticles triggered the apoptotic mode of cell death after increasing the intracellular reactive oxygen level by enhancing cellular uptake of micelles. Furthermore, in silico studies revealed higher absorption, distribution, metabolism, elimination and toxicity properties and bioavailability of the enzyme tyrosine protein kinase. Conclusion: The 3a nanoparticles enhanced the therapeutic potential and have higher potential for targeted drug delivery against cervical cancer.

New Theobromine Apoptotic Analogue with Anticancer Potential Targeting the EGFR Protein: Computational and In Vitro Studies

AIM

The aim of this study was to design and examine a novel epidermal growth factor receptor (EGFR) inhibitor with apoptotic properties by utilizing the essential structural characteristics of existing EGFR inhibitors as a foundation.

METHOD

The study began with the natural alkaloid theobromine and developed a new semisynthetic derivative (T-1-PMPA). Computational ADMET assessments were conducted first to evaluate its anticipated safety and general drug-likeness. Deep density functional theory (DFT) computations were initially performed to validate the three-dimensional (3D) structure and reactivity of T-1-PMPA. Molecular docking against the EGFR proteins was conducted to investigate T-1-PMPA's binding affinity and inhibitory potential. Additional molecular dynamics (MD) simulations over 200 ns along with MM-GPSA, PLIP, and principal component analysis of trajectories (PCAT) experiments were employed to verify the binding and inhibitory properties of T-1-PMPA. Afterward, T-1-PMPA was semisynthesized to validate the proposed design and in silico findings through several in vitro examinations.

RESULTS

DFT studies indicated T-1-PMPA's reactivity using electrostatic potential, global reactive indices, and total density of states. Molecular docking, MD simulations, MM-GPSA, PLIP, and ED suggested the binding and inhibitory properties of T-1-PMPA against the EGFR protein. The in silico ADMET predicted T-1-PMPA's safety and general drug-likeness. In vitro experiments demonstrated that T-1-PMPA effectively inhibited EGFRWT and EGFR790m, with IC50 values of 86 and 561 nM, respectively, compared to Erlotinib (31 and 456 nM). T-1-PMPA also showed significant suppression of the proliferation of HepG2 and MCF7 malignant cell lines, with IC50 values of 3.51 and 4.13 μM, respectively. The selectivity indices against the two cancer cell lines indicated the overall safety of T-1-PMPA. Flow cytometry confirmed the apoptotic effects of T-1-PMPA by increasing the total percentage of apoptosis to 42% compared to 31, and 3% in Erlotinib-treated and control cells, respectively. The qRT-PCR analysis further supported the apoptotic effects by revealing significant increases in the levels of Casp3 and Casp9. Additionally, T-1-PMPA controlled the levels of TNFα and IL2 by 74 and 50%, comparing Erlotinib's values (84 and 74%), respectively.

CONCLUSION

In conclusion, our study's findings suggest the potential of T-1-PMPA as a promising apoptotic anticancer lead compound targeting the EGFR.

Using test-time augmentation to investigate explainable AI: inconsistencies between method, model and human intuition

Stakeholders of machine learning models desire explainable artificial intelligence (XAI) to produce human-understandable and consistent interpretations. In computational toxicity, augmentation of text-based molecular representations has been used successfully for transfer learning on downstream tasks. Augmentations of molecular representations can also be used at inference to compare differences between multiple representations of the same ground-truth. In this study, we investigate the robustness of eight XAI methods using test-time augmentation for a molecular-representation model in the field of computational toxicity prediction. We report significant differences between explanations for different representations of the same ground-truth, and show that randomized models have similar variance. We hypothesize that text-based molecular representations in this and past research reflect tokenization more than learned parameters. Furthermore, we see a greater variance between in-domain predictions than out-of-domain predictions, indicating XAI measures something other than learned parameters. Finally, we investigate the relative importance given to expert-derived structural alerts and find similar importance given irregardless of applicability domain, randomization and varying training procedures. We therefore caution future research to validate their methods using a similar comparison to human intuition without further investigation. SCIENTIFIC CONTRIBUTION: In this research we critically investigate XAI through test-time augmentation, contrasting previous assumptions about using expert validation and showing inconsistencies within models for identical representations. SMILES augmentation has been used to increase model accuracy, but was here adapted from the field of image test-time augmentation to be used as an independent indication of the consistency within SMILES-based molecular representation models.

Identification of bioactive compounds of Zanthoxylum armatum as potential inhibitor of pyruvate kinase M2 (PKM2): Computational and virtual screening approaches

PKM2 (Pyruvate kinase M2) is the isoform of pyruvate kinase which is known to catalyse the last step of glycolysis that is responsible for energy production. This specific isoform is known to be highly expressed in certain cancerous conditions. Considering the role of this protein in various cancer conditions, we used PKM2 as a target protein to identify the potential compounds against this target. In this study, we have examined 96 compounds of Zanthoxylum armatum using an array of computational and in silico tools. The compounds were assessed for toxicity then their anticancer potential was predicted. The virtual screening was done with molecular docking followed by a detailed examination using molecular dynamics simulation. The majority of the compounds showed a higher probability of being antineoplastic. Based on toxicity, predicted anticancer potential, binding affinity, and binding site, three compounds (nevadensin, asarinin, and kaempferol) were selected as hit compounds. The binding energy of these compounds with PKM2 ranged from -7.7 to -8.3 kcal/mol and all hit compounds interact at the active site of the protein. The selected hit compounds formed a stable complex with PKM2 when simulated under physiological conditions. The dynamic analysis showed that these compounds remained attached to the active site till the completion of molecular simulation. MM-PBSA analysis showed that nevadensin exhibited a higher affinity towards PKM2 compared to asarinin and kaempferol. These compounds need to be assessed properties in vivo and in vitro to validate their efficacy.

Prioritization of the ecotoxicological hazard of PAHs towards aquatic species spanning three trophic levels using 2D-QSTR, read-across and machine learning-driven modelling approaches

Polycyclic aromatic hydrocarbons (PAHs) represent a common group of environmental pollutants that endanger various aquatic organisms via various pathways. To better prioritize the ecotoxicological hazard of PAHs to aquatic environment, we used 2D descriptors-based quantitative structure-toxicity relationship (QSTR) to assess the toxicity of PAHs toward six aquatic model organisms spanning three trophic levels. According to strict OECD guideline, six easily interpretable, transferable and reproducible 2D-QSTR models were constructed with high robustness and reliability. A mechanistic interpretation unveiled the key structural factors primarily responsible for controlling the aquatic ecotoxicity of PAHs. Furthermore, quantitative read-across and different machine learning approaches were employed to validate and optimize the modelling approach. Importantly, the optimum QSTR models were further applied for predicting the ecotoxicity of hundreds of untested/unknown PAHs gathered from Pesticide Properties Database (PPDB). Especially, we provided a priority list in terms of the toxicity of unknown PAHs to six aquatic species, along with the corresponding mechanistic interpretation. In summary, the models can serve as valuable tools for aquatic risk assessment and prioritization of untested or completely new PAHs chemicals, providing essential guidance for formulating regulatory policies.

Isolation, Characterization, Genome Annotation, and Evaluation of Tyrosinase Inhibitory Activity in Secondary Metabolites of Paenibacillus sp. JNUCC32: A Comprehensive Analysis through Molecular Docking and Molecular Dynamics Simulation

A potential strain, Paenibacillus sp. JNUCC32, was isolated and subjected to whole-genome sequencing. Genome functional annotation revealed its active metabolic capabilities. This study aimed to investigate the pivotal secondary metabolites in the biological system. Fermentation and extraction were performed, resulting in the isolation of seven known compounds: tryptophol (1), 3-(4-hydroxyphenyl)propionic acid (2), ferulic acid (3), maculosin (4), brevianamide F (5), indole-3-acetic acid (6), and butyric acid (7). Tryptophol exhibited favorable pharmacokinetic properties and demonstrated certain tyrosinase inhibitory activity (IC50 = 999 μM). For further analysis of its inhibition mechanism through molecular docking and molecular dynamics (MD) simulation, tryptophol formed three hydrogen bonds and a pro-Michaelis complex with tyrosinase (binding energy = -5.3 kcal/mol). The MD simulation indicated favorable stability for the tryptophol-mushroom tyrosinase complex, primarily governed by hydrogen bond interactions. The crucial residues VAL-283 and HIS-263 in the docking were also validated. This study suggests tryptophol as a potential candidate for antibrowning agents and dermatological research.

Synthesis, structural, characterization, antibacterial and antibiotic modifying activity, ADMET study, molecular docking and dynamics of chalcone (E)-1-(4-aminophenyl)-3-(4-nitrophenyl)prop-2-en-1-one in strains of Staphylococcus aureus carrying NorA and MepA efflux pumps

Chalcones have an open chain flavonoid structure that can be obtained from natural sources or by synthesis and are widely distributed in fruits, vegetables, and tea. They have a simple and easy to handle structure due to the α-β-unsaturated bridge responsible for most biological activities. The facility to synthesize chalcones combined with its efficient in combating serious bacterial infections make these compounds important agents in the fight against microorganisms. In this work, the chalcone (E)-1-(4-aminophenyl)-3-(4-nitrophenyl)prop-2-en-1-one (HDZPNB) was characterized by spectroscopy and electronic methods. In addition, microbiological tests were performed to investigate the modulator potential and efflux pump inhibition on S. aureus multi-resistant strains. The modulating effect of HDZPNB chalcone in association with the antibiotic norfloxacin, on the resistance of the S. aureus 1199 strain, resulted in increase the MIC. In addition, when HDZPNB was associated with ethidium bromide (EB), it caused an increase in the MIC value, thus not inhibiting the efflux pump. For the strain of S. aureus 1199B, carrying the NorA pump, the HDZPNB associated with norfloxacin showed no modulatory, and when the chalcone was used in association with EB, it had no inhibitory effect on the efflux pump. For the tested strain of S. aureus K2068, which carries the MepA pump, it can be observed that the chalcone together the antibiotic resulted in an increase the MIC. On the other hand, when chalcone was used in association with EB, it caused a decrease in bromide MIC, equal to the reduction caused by standard inhibitors. Thus, these results indicate that the HDZPNB could also act as an inhibitor of the S. aureus gene overexpressing pump MepA. The molecular docking reveals that chalcone has a good binding energies -7.9 for HDZPNB/MepA complexes, molecular dynamics simulations showed that Chalcone/MetA complexes showed good stability of the structure in an aqueous solution, and ADMET study showed that the chalcone has a good oral bioavailability, high passive permeability, low risk of efflux, low clearance rate and low toxic risk by ingestion. The microbiological tests show that the chalcone can be used as a possible inhibitor of the Mep A efflux pump.Communicated by Ramaswamy H. Sarma.

In Silico Study in MPO and Molecular Docking of the Synthetic Drynaran Analogues Against the Chronic Tinnitus: Modulation of the M1 Muscarinic Acetylcholine Receptor

Tinnitus is a syndrome that affects the human auditory system and is characterized by a perception of sounds in the absence of acoustic stimuli, or in total silence. Research indicates that muscarinic acetylcholine receptors (mAChRs), especially the M1 type, have a fundamental role in the alterations of auditory perceptions of tinnitus. Here, a series of computer-aided tools were used, from molecular surface analysis software to services available on the web for estimating pharmacokinetics and pharmacodynamics. The results infer that the low lipophilicity ligands, that is, the 1a-d alkyl furans, present the best pharmacokinetic profile, as compounds with an optimal alignment between permeability and clearance. However, only ligands 1a and 1b have properties that are safe for the central nervous system, the site of cholinergic modulation. These ligands showed similarity with compounds deposited in the European Molecular Biology Laboratory chemical (ChEMBL) database acting on the mAChRs M1 type, the target selected for the molecular docking test. The simulations suggest that the 1 g ligand can form the ligand-receptor complex with the best affinity energy order and that, together with the 1b ligand, they are competitive agonists in relation to the antagonist Tiotropium, in addition to acting in synergism with the drug Bromazepam in the treatment of chronic tinnitus.

New insight into gut microbiota-derived metabolites to enhance liver regeneration via network pharmacology study

We intended to identify favourable metabolite(s) and pharmacological mechanism(s) of gut microbiota (GM) for liver regeneration (LR) through network pharmacology. We utilized the gutMGene database to obtain metabolites of GM, and targets associated with metabolites as well as LR-related targets were identified using public databases. Furthermore, we performed a molecular docking assay on the active metabolite(s) and target(s) to verify the network pharmacological concept. We mined a total of 208 metabolites in the gutMGene database and selected 668 targets from the SEA (1,256 targets) and STP (947 targets) databases. Finally, 13 targets were identified between 61 targets and the gutMGene database (243 targets). Protein-protein interaction network analysis showed that AKT1 is a hub target correlated with 12 additional targets. In this study, we describe the potential microbe from the microbiota (E. coli), chemokine signalling pathway, AKT1 and myricetin that accelerate LR, providing scientific evidence for further clinical trials.

In Silico and In Vitro Analyses to Repurpose Quercetin as a Human Pancreatic α-Amylase Inhibitor

Human pancreatic α-amylase (HPA), situated at the apex of the starch digestion hierarchy, is an attractive therapeutic approach to precisely regulate blood glucose levels, thereby efficiently managing diabetes. Polyphenols offer a natural and multifaceted approach to moderate postprandial sugar spikes, with their slight modulation in carbohydrate digestion and potential secondary benefits, such as antioxidant and anti-inflammatory effects. Taking into consideration the unfavorable side effects of currently available commercial medications, we aimed to study a library of polyphenols attributed to their remarkable antidiabetic properties and screened the most potent HPA inhibitor via a comprehensive in silico study encompassing molecular docking, molecular mechanics with generalized Born and surface area solvation (MM/GBSA) calculation, molecular dynamics (MD) simulation, density functional theory (DFT) study, and pharmacokinetic properties followed by an in vitro assay. Significant hydrogen bonding with the catalytic triad residues of HPA, prominent MM/GBSA binding energy of -27.03 kcal/mol, and the stable nature of the protein-ligand complex with regard to 100 ns MD simulation screened quercetin as the best HPA inhibitor. Additionally, quercetin showed strong reactivity in the substrate-binding pocket of HPA and exhibited favorable pharmacokinetic properties with a considerable inhibitory concentration (IC50) of 57.37 ± 0.9 μg/mL against α-amylase. This study holds prospects for HPA inhibition and suggests quercetin as an approach to therapy for diabetes; however, it is imperative to conduct further research.

Therapeutic Study of Cinnamic Acid Derivative for Oxidative Stress Ablation: The Computational and Experimental Answers

This study aimed to examine the therapeutic activity of the cinnamic acid derivative KAD-7 (N'-(2,4-dichlorobenzylidene)-3-(4-methoxyphenyl) acrylohydrazide) on Fe2+-induced oxidative hepatic injury via experimental and computational models. In addition, the role of ATPase and ectonucleoside triphosphate diphosphohydrolase (ENTPDase) in the coordination of cellular signals is speculated upon to proffer suitable therapeutics for metabolic stress disorder upon their inhibition. While we know little about therapeutics with flexible dual inhibitors for these protein targets, this study was designed to screen KAD-7's (N'-(2,4-dichlorobenzylidene)-3-(4-methoxyphenyl) acrylohydrazide) inhibitory potential for both protein targets. We induced oxidative hepatic damage via the incubation of hepatic tissue supernatant with 0.1 mM FeSO4 for 30 min at 37 °C. We achieved the treatment by incubating the hepatic tissues with KAD-7 under the same conditions. The catalase (CAT), glutathione (GSH), malondialdehyde (MDA), ATPase, and ENTPDase activity were all measured in the tissues. We predicted how the drug candidate would work against ATPase and ENTPDase targets using molecular methods. When hepatic injury was induced, there was a significant decrease in the levels of the GSH, CAT, and ENTPDase (p < 0.05) activities. In contrast, we found a noticeable rise in the MDA levels and ATPase activity. KAD-7 therapy resulted in lower levels of these activities overall (p < 0.05), as compared to the control levels. We found the compound to have a strong affinity for ATPase (-7.1 kcal/mol) and ENTPDase (-7.4 kcal/mol), and a better chemical reactivity than quercetin. It also met all drug-likeness parameters. Our study shows that KAD-7 can protect the liver from damage caused by FeSO4 by reducing oxidative stress and purinergic actions. Our studies indicate that KAD-7 could be developed as a therapeutic option since it can flexibly inhibit both ATPase and ENTPDase.

Computer-aided analysis for identification of novel analogues of ketoprofen based on molecular docking, ADMET, drug-likeness and DFT studies for the treatment of inflammation

Computer-based drug design is increasingly used in strategies for discovering new molecules for therapeutic purposes. The targeted drug is ketoprofen (KTP), which belongs to the family of non-steroidal anti-inflammatory drugs, which are widely used for the treatment of pain, fever, inflammation and certain types of cancers. In an attempt to rationalize the search for 72 new potential anti-inflammatory compounds on the COX-2 enzyme, we carried out an in silico protocol that successfully combines molecular docking towards COX-2 receptor (5F1A), ADMET pharmacokinetic parameters, drug-likeness rules and molecular electrostatic potential (MEP). It was found that six of the compounds analyzed satisfy with the associated values to physico-chemical properties as key evaluation parameters for the drug-likeness and demonstrate a hydrophobic character which makes their solubility in aqueous media difficult and easy in lipids. All the compounds presented good ADMET profile and they showed an interaction with the amino acids responsible for anti-inflammatory activity of the COX-2 isoenzyme. The calculation of the MEP of the six analogues reveals new preferential sites involving the formation of new bonds. Consequently, this result allowed us to understand the origin of the potential increase in the anti-inflammatory activity of the candidates. Finally, it was obtained that six compounds have a binding mode, binding energy, and stability in the active site of COX-2 like the reference drug ketoprofen, suggesting that these compounds could become a powerful candidate in the inhibition of the COX-2 enzyme.Communicated by Ramaswamy H. Sarma.

Anti-breast cancer potential of a new xanthine derivative: In silico, antiproliferative, selectivity, VEGFR-2 inhibition, apoptosis induction and migration inhibition studies

BACKGROUND

The overexpression of VEGFR-2 receptors in breast cancer provides a valuable approach to anticancer strategies. Targeting VEGFR-2, a new semisynthetic compound (T-1-MCPAB) has been designed.

METHODS

Computational methods (ADMET, toxicity, DFT, Molecular Docking, Molecular Dynamics Simulations, MM-GBSA, PLIP, and PCAT) were conducted. In addition to the semi-synthesis, in vitro studies (anti-VEGFR-2, anti-proliferative, flow cytometry, and wound scratch assay) were employed.

RESULTS

ADME and toxicity profiles of T-1-MCPAB studies indicated its overall drug-likeness showing results much better than Sorafenib. Then, T-1-MCPAB's exact 3D structure, stability, and reactivity were evoked by the DFT calculations. Molecular docking, molecular dynamics simulations, MM-GPSA, PLIP, and PCAT studies denoted the correct binding and inhibiting potential of T-1-MCPAB, towards VEGFR-2 protein. After the semisynthesis, T-1-MCPAB inhibited VEGFR-2 with an IC50 of 0.135 µM, which was comparable to sorafenib's IC50 of 0.0591 µM. T-1-MCPAB also showed a notable performance against MCF7 and T47D breast cancer cell lines with IC50 values of 30.95 µM and 63.64 µM, respectively, and had high selectivity index values of 3.7 and 1.8, respectively. Furthermore, T-1-MCPAB influenced early and late apoptosis and significantly decreased the potential of MCF7 cells to heal and migrate.

CONCLUSION

T-1-MCPAB is a promising VEGFR-2 inhibitor with potential for breast cancer treatment. Further chemical and biological studies are needed to explore its potential as a therapeutic agent.

Review of machine learning and deep learning models for toxicity prediction

The ever-increasing number of chemicals has raised public concerns due to their adverse effects on human health and the environment. To protect public health and the environment, it is critical to assess the toxicity of these chemicals. Traditional in vitro and in vivo toxicity assays are complicated, costly, and time-consuming and may face ethical issues. These constraints raise the need for alternative methods for assessing the toxicity of chemicals. Recently, due to the advancement of machine learning algorithms and the increase in computational power, many toxicity prediction models have been developed using various machine learning and deep learning algorithms such as support vector machine, random forest, k-nearest neighbors, ensemble learning, and deep neural network. This review summarizes the machine learning- and deep learning-based toxicity prediction models developed in recent years. Support vector machine and random forest are the most popular machine learning algorithms, and hepatotoxicity, cardiotoxicity, and carcinogenicity are the frequently modeled toxicity endpoints in predictive toxicology. It is known that datasets impact model performance. The quality of datasets used in the development of toxicity prediction models using machine learning and deep learning is vital to the performance of the developed models. The different toxicity assignments for the same chemicals among different datasets of the same type of toxicity have been observed, indicating benchmarking datasets is needed for developing reliable toxicity prediction models using machine learning and deep learning algorithms. This review provides insights into current machine learning models in predictive toxicology, which are expected to promote the development and application of toxicity prediction models in the future.

Descriptor-Free Deep Learning QSAR Model for the Fraction Unbound in Human Plasma

Chemical-specific parameters are either measured in vitro or estimated using quantitative structure-activity relationship (QSAR) models. The existing body of QSAR work relies on extracting a set of descriptors or fingerprints, subset selection, and training a machine learning model. In this work, we used a state-of-the-art natural language processing model, Bidirectional Encoder Representations from Transformers, which allowed us to circumvent the need for calculation of these chemical descriptors. In this approach, simplified molecular-input line-entry system (SMILES) strings were embedded in a high-dimensional space using a two-stage training approach. The model was first pre-trained on a masked SMILES token task and then fine-tuned on a QSAR prediction task. The pre-training task learned meaningful high-dimensional embeddings based upon the relationships between the chemical tokens in the SMILES strings derived from the "in-stock" portion of the ZINC 15 dataset─a large dataset of commercially available chemicals. The fine-tuning task then perturbed the pre-trained embeddings to facilitate prediction of a specific QSAR endpoint of interest. The power of this model stems from the ability to reuse the pre-trained model for multiple different fine-tuning tasks, reducing the computational burden of developing multiple models for different endpoints. We used our framework to develop a predictive model for fraction unbound in human plasma (fu,p). This approach is flexible, requires minimum domain expertise, and can be generalized for other parameters of interest for rapid and accurate estimation of absorption, distribution, metabolism, excretion, and toxicity.

DeepAmes: A deep learning-powered Ames test predictive model with potential for regulatory application

The Ames assay is required by the regulatory agencies worldwide to assess the mutagenic potential risk of consumer products. As well as this in vitro assay, in silico approaches have been widely used to predict Ames test results as outlined in the International Council for Harmonization (ICH) guidelines. Building on this in silico approach, here we describe DeepAmes, a high performance and robust model developed with a novel deep learning (DL) approach for potential utility in regulatory science. DeepAmes was developed with a large and consistent Ames dataset (>10,000 compounds) and was compared with other five standard Machine Learning (ML) methods. Using a test set of 1,543 compounds, DeepAmes was the best performer in predicting the outcome of Ames assay. In addition, DeepAmes yielded the best and most stable performance up to when compounds were >30% outside of the applicability domain (AD). Regarding the potential for regulatory application, a revised version of DeepAmes with a much-improved sensitivity of 0.87 from 0.47. In conclusion, DeepAmes provides a DL-powered Ames test predictive model for predicting the results of Ames tests; with its defined AD and clear context of use, DeepAmes has potential for utility in regulatory application.

De novo design of bioactive phenol and chromone derivatives for inhibitors of Spike glycoprotein of SARS-CoV-2 in silico

This work presents the synthesis of 12 phenol and chromone derivatives, prepared by the analogs, and the possibility of conducting an in silico study of its derivatives as a therapeutic alternative to combat the SARS-CoV-2, pathogen responsible for COVID-19 pandemic, using its S-glycoprotein as a macromolecular target. After the initial screening for the ranking of the products, it was chosen which structure presented the best energy bond with the target. As a result, derivative 4 was submitted to a molecular growth study using artificial intelligence, where 8436 initial structures were obtained that passed through the interaction filters and similarity to the active glycoprotein pocket through the MolAICal computational package. Thus, 557 Hits with active configuration were generated, which is very promising compared to the BLA reference link for inhibiting the biological target. Molecular dynamics also simulated these compounds to verify their stability within the active protein site to seek new therapeutic propositions to fight against the pandemic. The Hit 48 and 250 are the most active compounds against SARS-CoV-2. In summary, the results show that the Hit 250 would be more active than the natural compound, which could be further developed for further testing against SARS-CoV-2. The study employs the de novo approach to design new drugs, combining artificial intelligence and molecular dynamics simulations to create efficient molecular structures. This research aims to contribute to the development of effective therapeutic strategies against the pandemic.

MolFPG: Multi-level fingerprint-based Graph Transformer for accurate and robust drug toxicity prediction

Drug toxicity prediction is essential to drug development, which can help screen compounds with potential toxicity and reduce the cost and risk of animal experiments and clinical trials. However, traditional handcrafted feature-based and molecular-graph-based approaches are insufficient for molecular representation learning. To address the problem, we developed an innovative molecular fingerprint Graph Transformer framework (MolFPG) with a global-aware module for interpretable toxicity prediction. Our approach encodes compounds using multiple molecular fingerprinting techniques and integrates Graph Transformer-based molecular representation for feature learning and toxic prediction. Experimental results show that our proposed approach has high accuracy and reliability in predicting drug toxicity. In addition, we explored the relationship between drug features and toxicity through an interpretive analysis approach, which improved the interpretability of the approach. Our results highlight the potential of Graph Transformers and multi-level fingerprints for accelerating the drug discovery process by reliably, effectively alarming drug safety. We believe that our study will provide vital support and reference for further development in the field of drug development and toxicity assessment.

GC-MS chemical profiling, antioxidant, anti-diabetic, and anti-inflammatory activities of ethyl acetate fraction of Spilanthes filicaulis (Schumach. and Thonn.) C.D. Adams leaves: experimental and computational studies

Introduction: This study aimed to investigate the chemical profile of GC-MS, antioxidant, anti-diabetic, and anti-inflammatory activities of the ethyl acetate fraction of Spilanthes filicaulis leaves (EFSFL) via experimental and computational studies. Methods: After inducing oxidative damage with FeSO4, we treated the tissues with different concentrations of EFSFL. An in-vitro analysis of EFSFL was carried out to determine its potential for antioxidant, anti-diabetic, and anti-inflammatory activities. We also measured the levels of CAT, SOD, GSH, and MDA. Results and discussion: EFSFL exhibited anti-inflammatory properties through membrane stabilizing properties (IC50 = 572.79 μg/ml), proteinase inhibition (IC50 = 319.90 μg/ml), and inhibition of protein denaturation (IC50 = 409.88 μg/ml). Furthermore, EFSFL inhibited α-amylase (IC50 = 169.77 μg/ml), α-glucosidase (IC50 = 293.12 μg/ml) and DPP-IV (IC50 = 380.94 μg/ml) activities, respectively. Our results indicated that induction of tissue damage reduced the levels of GSH, SOD, and CAT activities, and increased MDA levels. However, EFSFL treatment restores these levels to near normal. GC-MS profiling shows that EFSFL contains 13 compounds, with piperine being the most abundant. In silico interaction of the phytoconstituents using molecular and ensembled-based docking revealed strong binding tendencies of two hit compounds to DPP IV (alpha-caryophyllene and piperine with a binding affinity of -7.8 and -7.8 Kcal/mol), α-glucosidase (alpha-caryophyllene and piperine with a binding affinity of -9.6 and -8.9 Kcal/mol), and to α-amylase (piperine and Benzocycloheptano[2,3,4-I,j]isoquinoline, 4,5,6,6a-tetrahydro-1,9-dihydroxy-2,10-dimethoxy-5-methyl with a binding affinity of -7.8 and -7.9 Kcal/mol), respectively. These compounds also presented druggable properties with favorable ADMET. Conclusively, the antioxidant, antidiabetic, and anti-inflammatory activities of EFSFL could be due to the presence of secondary metabolites.

Targeting EGFR, RSK1, RAF1, PARP2 and LIN28B for several cancer type therapies with newly synthesized pyrazole derivatives via a computational study

Cancer remains the leading cause of death in the world despite the significant advancements made in anticancer drug discovery. This study is aimed to computationally evaluate the efficacy of 63 in-house synthesized pyrazole derivatives targeted to bind with prominent cancer targets namely EGFR, RSK1, RAF1, PARP2 and LIN28B known to be expressed, respectively, in lung, colon, skin, ovarian and pancreatic cancer cells. Initially, we perform the molecular docking investigations for all pyrazole compounds with a comparison to known standard drugs for each target. Docking studies have revealed that some pyrazole compounds possess better binding affinity scores than standard drug compounds. Thereafter, a long-range of 1 μs molecular dynamic (MD) simulation study for top ranked docked compounds with all respective proteins was carried out to assess the interaction stability in a dynamic environment. The results suggested that the top ranked complexes showed a stable interaction profile for a longer period of time. The outcome of this study suggests that pyrazole compounds, M33, M36, M76 and M77, are promising molecular candidates that can modulate the studied target proteins significantly in comparison to their known inhibitor based on their selective binding interactions profile. Furthermore, ADME-T profile has been explored to check for the drug-likeness and pharmacokinetics profiles and found that all proposed compounds exhibited acceptable values for being a potential drug-like candidate with non-toxic characteristics. Overall, extensive computational investigations indicate that the four proposed pyrazole inhibitors/modulators studied against each respective target protein will be helpful for future cancer therapeutic developments.Communicated by Ramaswamy H. Sarma.

Chemistry-intuitive explanation of graph neural networks for molecular property prediction with substructure masking

Graph neural networks (GNNs) have been widely used in molecular property prediction, but explaining their black-box predictions is still a challenge. Most existing explanation methods for GNNs in chemistry focus on attributing model predictions to individual nodes, edges or fragments that are not necessarily derived from a chemically meaningful segmentation of molecules. To address this challenge, we propose a method named substructure mask explanation (SME). SME is based on well-established molecular segmentation methods and provides an interpretation that aligns with the understanding of chemists. We apply SME to elucidate how GNNs learn to predict aqueous solubility, genotoxicity, cardiotoxicity and blood-brain barrier permeation for small molecules. SME provides interpretation that is consistent with the understanding of chemists, alerts them to unreliable performance, and guides them in structural optimization for target properties. Hence, we believe that SME empowers chemists to confidently mine structure-activity relationship (SAR) from reliable GNNs through a transparent inspection on how GNNs pick up useful signals when learning from data.

In silico study of novel niclosamide derivatives, SARS-CoV-2 nonstructural proteins catalytic residue-targeting small molecules drug candidates

The severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2)-mediated coronavirus disease 2019 (COVID-19) infection remains a global pandemic and health emergency with overwhelming social and economic impacts throughout the world. Therapeutics for COVID-19 are limited to only remdesivir; therefore, there is a need for combined, multidisciplinary efforts to develop new therapeutic molecules and explore the effectiveness of existing drugs against SARS-CoV-2. In the present study, we reported eight (SCOV-L-02, SCOV-L-09, SCOV-L-10, SCOV-L-11, SCOV-L-15, SCOV-L-18, SCOV-L-22, and SCOV-L-23) novel structurally related small-molecule derivatives of niclosamide (SCOV-L series) for their targeting potential against angiotensin-converting enzyme-2 (ACE2), type II transmembrane serine protease (TMPRSS2), and SARS-COV-2 nonstructural proteins (NSPs) including NSP5 (3CLpro), NSP3 (PLpro), and RdRp. Our correlation analysis suggested that ACE2 and TMPRSS2 modulate host immune response via regulation of immune-infiltrating cells at the site of tissue/organs entries. In addition, we identified some TMPRSS2 and ACE2 microRNAs target regulatory networks in SARS-CoV-2 infection and thus open up a new window for microRNAs-based therapy for the treatment of SARS-CoV-2 infection. Our in vitro study revealed that with the exception of SCOV-L-11 and SCOV-L-23 which were non-active, the SCOV-L series exhibited strict antiproliferative activities and non-cytotoxic effects against ACE2- and TMPRSS2-expressing cells. Our molecular docking for the analysis of receptor-ligand interactions revealed that SCOV-L series demonstrated high ligand binding efficacies (at higher levels than clinical drugs) against the ACE2, TMPRSS2, and SARS-COV-2 NSPs. SCOV-L-18, SCOV-L-15, and SCOV-L-09 were particularly found to exhibit strong binding affinities with three key SARS-CoV-2's proteins: 3CLpro, PLpro, and RdRp. These compounds bind to the several catalytic residues of the proteins, and satisfied the criteria of drug-like candidates, having good adsorption, distribution, metabolism, excretion, and toxicity (ADMET) pharmacokinetic profile. Altogether, the present study suggests the therapeutic potential of SCOV-L series for preventing and managing SARs-COV-2 infection and are currently under detailed investigation in our lab.

Artificial Intelligence in Drug Toxicity Prediction: Recent Advances, Challenges, and Future Perspectives

Toxicity prediction is a critical step in the drug discovery process that helps identify and prioritize compounds with the greatest potential for safe and effective use in humans, while also reducing the risk of costly late-stage failures. It is estimated that over 30% of drug candidates are discarded owing to toxicity. Recently, artificial intelligence (AI) has been used to improve drug toxicity prediction as it provides more accurate and efficient methods for identifying the potentially toxic effects of new compounds before they are tested in human clinical trials, thus saving time and money. In this review, we present an overview of recent advances in AI-based drug toxicity prediction, including the use of various machine learning algorithms and deep learning architectures, of six major toxicity properties and Tox21 assay end points. Additionally, we provide a list of public data sources and useful toxicity prediction tools for the research community and highlight the challenges that must be addressed to enhance model performance. Finally, we discuss future perspectives for AI-based drug toxicity prediction. This review can aid researchers in understanding toxicity prediction and pave the way for new methods of drug discovery.

Chemical rules for optimization of chemical mutagenicity via matched molecular pairs analysis and machine learning methods

Chemical mutagenicity is a serious issue that needs to be addressed in early drug discovery. Over a long period of time, medicinal chemists have manually summarized a series of empirical rules for the optimization of chemical mutagenicity. However, given the rising amount of data, it is getting more difficult for medicinal chemists to identify more comprehensive chemical rules behind the biochemical data. Herein, we integrated a large Ames mutagenicity data set with 8576 compounds to derive mutagenicity transformation rules for reversing Ames mutagenicity via matched molecular pairs analysis. A well-trained consensus model with a reasonable applicability domain was constructed, which showed favorable performance in the external validation set with an accuracy of 0.815. The model was used to assess the generalizability and validity of these mutagenicity transformation rules. The results demonstrated that these rules were of great value and could provide inspiration for the structural modifications of compounds with potential mutagenic effects. We also found that the local chemical environment of the attachment points of rules was critical for successful transformation. To facilitate the use of these mutagenicity transformation rules, we integrated them into ADMETopt2 ( http://lmmd.ecust.edu.cn/admetsar2/admetopt2/ ), a free web server for optimization of chemical ADMET properties. The above-mentioned approach would be extended to the optimization of other toxicity endpoints.

Ecotoxicological QSAR modelling of the acute toxicity of fused and non-fused polycyclic aromatic hydrocarbons (FNFPAHs) against two aquatic organisms: Consensus modelling and comparison with ECOSAR

Fused and non-fused polycyclic aromatic hydrocarbons (FNFPAHs) are a type of organic compounds widely occurring in the environment that pose a potential hazard to ecosystem and public health, and thus receive extensive attention from various regulatory agencies. Here, quantitative structure-activity relationship (QSAR) models were constructed to model the ecotoxicity of FNFPAHs against two aquatic species, Daphnia magna and Oncorhynchus mykiss. According to the stringent OECD guidelines, we used genetic algorithm (GA) plus multiple linear regression (MLR) approach to establish QSAR models of the two aquatic toxicity endpoints: D. magna (48 h LC₅₀) and O. mykiss (96 h LC₅₀). The models were established using simple 2D descriptors with explicit physicochemical significance and evaluated using various internal/external validation metrics. The results clearly show that both models are statistically robust (Q_LOO² = 0.7834 for D. magna and Q_LOO² = 0.8162 for O. mykiss), have good internal fitness (R² = 0.8159 for D. magna and R² = 0.8626 for O. mykiss and external predictive ability (D. magna: R_test² = 0.8259, Q_Fn² = 0.7640∼0.8140, CCC_test = 0.8972; O. mykiss:R_test² = 0.8077, Q_Fn² = 0.7615∼0.7722, CCC_test = 0.8910). To prove the predictive performance of the developed models, an additional comparison with the standard ECOSAR tool obviously shows that our models have lower RMSE values. Subsequently, we utilized the best models to predict the true external set compounds collected from the PPDB database to further fill the toxicity data gap. In addition, consensus models (CMs) that integrate all validated individual models (IMs) were more externally predictive than IMs, of which CM2 has the best prediction performance towards the two aquatic species. Overall, the models presented here could be used to evaluate unknown FNFPAHs inside the domain of applicability (AD), thus being very important for environmental risk assessment under current regulatory frameworks.

The identification of metabolites from gut microbiota in NAFLD via network pharmacology

The metabolites of gut microbiota show favorable therapeutic effects on nonalcoholic fatty liver disease (NAFLD), but the active metabolites and mechanisms against NAFLD have not been documented. The aim of the study was to investigate the active metabolites and mechanisms of gut microbiota against NAFLD by network pharmacology. We obtained a total of 208 metabolites from the gutMgene database and retrieved 1256 targets from similarity ensemble approach (SEA) and 947 targets from the SwissTargetPrediction (STP) database. In the SEA and STP databases, we identified 668 overlapping targets and obtained 237 targets for NAFLD. Thirty-eight targets were identified out of those 237 and 223 targets retrieved from the gutMgene database, and were considered the final NAFLD targets of metabolites from the microbiome. The results of molecular docking tests suggest that, of the 38 targets, mitogen-activated protein kinase 8-compound K and glycogen synthase kinase-3 beta-myricetin complexes might inhibit the Wnt signaling pathway. The microbiota-signaling pathways-targets-metabolites network analysis reveals that Firmicutes, Fusobacteria, the Toll-like receptor signaling pathway, mitogen-activated protein kinase 1, and phenylacetylglutamine are notable components of NAFLD and therefore to understanding its processes and possible therapeutic approaches. The key components and potential mechanisms of metabolites from gut microbiota against NAFLD were explored utilizing network pharmacology analyses. This study provides scientific evidence to support the therapeutic efficacy of metabolites for NAFLD and suggests holistic insights on which to base further research.

Machine Learning Toxicity Prediction: Latest Advances by Toxicity End Point

Machine learning (ML) models to predict the toxicity of small molecules have garnered great attention and have become widely used in recent years. Computational toxicity prediction is particularly advantageous in the early stages of drug discovery in order to filter out molecules with high probability of failing in clinical trials. This has been helped by the increase in the number of large toxicology databases available. However, being an area of recent application, a greater understanding of the scope and applicability of ML methods is still necessary. There are various kinds of toxic end points that have been predicted in silico. Acute oral toxicity, hepatotoxicity, cardiotoxicity, mutagenicity, and the 12 Tox21 data end points are among the most commonly investigated. Machine learning methods exhibit different performances on different data sets due to dissimilar complexity, class distributions, or chemical space covered, which makes it hard to compare the performance of algorithms over different toxic end points. The general pipeline to predict toxicity using ML has already been analyzed in various reviews. In this contribution, we focus on the recent progress in the area and the outstanding challenges, making a detailed description of the state-of-the-art models implemented for each toxic end point. The type of molecular representation, the algorithm, and the evaluation metric used in each research work are explained and analyzed. A detailed description of end points that are usually predicted, their clinical relevance, the available databases, and the challenges they bring to the field are also highlighted.

Machine Learning Modeling and Insights into the Structural Characteristics of Drug-Induced Neurotoxicity

Neurotoxicity can be resulted from many diverse clinical drugs, which has been a cause of concern to human populations across the world. The detection of drug-induced neurotoxicity (DINeurot) potential with biological experimental methods always required a lot of budget and time. In addition, few studies have addressed the structural characteristics of neurotoxic chemicals. In this study, we focused on the computational modeling for drug-induced neurotoxicity with machine learning methods and the insights into the structural characteristics of neurotoxic chemicals. Based on the clinical drug data with neurotoxicity effects, we developed 35 different classifiers by combining five different machine learning methods and seven fingerprint packages. The best-performing model achieved good results on both 5-fold cross-validation (balanced accuracy of 76.51%, AUC value of 0.83, and MCC value of 0.52) and external validation (balanced accuracy of 83.63%, AUC value of 0.87, and MCC value of 0.67). The model can be freely accessed on the web server DINeuroTpredictor (http://dineurot.sapredictor.cn/). We also analyzed the distribution of several key molecular properties between neurotoxic and non-neurotoxic structures. The results indicated that several physicochemical properties were significantly different between the neurotoxic and non-neurotoxic compounds, including molecular polar surface area (MPSA), AlogP, the number of hydrogen bond acceptors (nHAcc) and donors (nHDon), the number of rotatable bonds (nRotB), and the number of aromatic rings (nAR). In addition, 18 structural alerts responsible for chemical neurotoxicity were identified. The structural alerts have been integrated with our web server SApredictor (http://www.sapredictor.cn). The results of this study could provide useful information for the understanding of the structural characteristics and computational prediction for chemical neurotoxicity.

Evaluating different web applications to assess the toxicity of plasticizers

Plasticizers increase the flexibility of plastics. As environmental leachates they lead to increased water and soil pollution, as well as to serious harm to human health. This study was set out to explore various web applications to predict the toxicological properties of plasticizers. Web-based tools (e.g., BOILED-Egg, LAZAR, PROTOX-II, CarcinoPred-EL) and VEGA were accessed via an 5th-10th generation computer in order to obtain toxicological predictions. Based on the LAZAR mutagenicity assessment was only bisphenol F predicted as mutagenic. The BBP and DBP in RF; DEHP in RF and XGBoost; DNOP in RF and XGBoost models were predicted as carcinogenic in the CarcinoPred-EL web application. From the bee predictive model (KNN/IRFMN) BPF, di-n-propyl phthalate, diallyl phthalate, dibutyl phthalate, and diisohexyl phthalate were predicted as strong bee toxicants. Acute toxicity for fish using the model Sarpy/IRFMN predicted 19 plasticizers as strong toxicants with LC50 values of less than 1 mg/L. This study also considered plasticizer effects on gastrointestinal absorption and other toxicological endpoints.

Chemoinformatic Design and Profiling of Derivatives of Dasabuvir, Efavirenz, and Tipranavir as Potential Inhibitors of Zika Virus RNA-Dependent RNA Polymerase and Methyltransferase

Zika virus (ZIKV) infection is one of the mosquito-borne flaviviruses of human importance with more than 2 million suspected cases and more than 1 million people infected in about 30 countries. There are reported inhibitors of the zika virus replication machinery, but no approved effective antiviral therapy including vaccines directed against the virus for treatment or prevention is currently available. The study investigated the chemoinformatic design and profiling of derivatives of dasabuvir, efavirenz, and tipranavir as potential inhibitors of the zika virus RNA-dependent RNA polymerase (RdRP) and/or methyltransferase (MTase). The three-dimensional (3D) coordinates of dasabuvir, efavirenz, and tipranavir were obtained from the PubChem database, and their respective derivatives were designed with DataWarrior-5.2.1 using an evolutionary algorithm. Derivatives that were not mutagenic, tumorigenic, or irritant were selected; docked into RdRP and MTase; and further subjected to absorption, distribution, metabolism, excretion, and toxicity (ADMET) evaluation with Swiss-ADME and pkCSM web tools. Some of the designed compounds are Lipinski's rule-of-five compliant, with good synthetic accessibilities. Compounds 20d, 21d, 22d, and 1e are nontoxic with the only limitation of CYP1A2, CYP2C19, and/or CYP2C9 inhibition. Replacements of -CH₃ and -NH- in the methanesulfonamide moiety of dasabuvir with -OH and -CH₂- or -CH₂CH₂-, respectively, improved the safety/toxicity profile. Hepatotoxicity in 5d, 4d, and 18d is likely due to -NH- in their methanesulfonamide/sulfamic acid moieties. These compounds are potent inhibitors of N-7 and 2'-methylation activities of ZIKV methyltransferase and/or RNA synthesis through interactions with amino acid residues in the priming loop/"N-pocket" in the virus RdRP. Synthesis of these compounds and wet laboratory validation against ZIKV are recommended.

In-Silico Analysis of Phytocompounds of Olea europaea as Potential Anti-Cancer Agents to Target PKM2 Protein

Globally, cancer is the second leading cause of mortality and morbidity. The growth and development of cancer are extremely complex. It is caused by a variety of pathways and involves various types of enzymes. Pyruvate kinase M2 (PKM2) is an isoform of pyruvate kinase, that catalyses the last steps of glycolysis to produce energy. PKM2 is relatively more expressed in tumour cells where it tends to exist in a dimer form. Various medicinal plants are available that contain a variety of micronutrients to combat against different cancers. The phytocompounds of the olive tree (Olea europaea) leaves play an important role in inhibiting the proliferation of several cancers. In this study, the phytocompounds of olive leaf extract (OLE) were studied using various in silico tools, such as pkCSM software to predict ADMET properties and PASS Online software to predict anticancer activity. However, the molecular docking study provided the binding energies and inhibition constant and confirmed the interaction between PKM2 and the ligands. The dynamic behaviour, conformational changes, and stability between PKM2 and the top three hit compounds (Verbascoside (Ver), Rutin (Rut), and Luteolin_7_O_glucoside (Lut)) are studied by MD simulations.

On modeling and utilizing chemical compound information with deep learning technologies: A task-oriented approach

A large number of chemical compounds are available in databases such as PubChem and ZINC. However, currently known compounds, though large, represent only a fraction of possible compounds, which is known as chemical space. Many of these compounds in the databases are annotated with properties and assay data that can be used for drug discovery efforts. For this goal, a number of machine learning algorithms have been developed and recent deep learning technologies can be effectively used to navigate chemical space, especially for unknown chemical compounds, in terms of drug-related tasks. In this article, we survey how deep learning technologies can model and utilize chemical compound information in a task-oriented way by exploiting annotated properties and assay data in the chemical compounds databases. We first compile what kind of tasks are trying to be accomplished by machine learning methods. Then, we survey deep learning technologies to show their modeling power and current applications for accomplishing drug related tasks. Next, we survey deep learning techniques to address the insufficiency issue of annotated data for more effective navigation of chemical space. Chemical compound information alone may not be powerful enough for drug related tasks, thus we survey what kind of information, such as assay and gene expression data, can be used to improve the prediction power of deep learning models. Finally, we conclude this survey with four important newly developed technologies that are yet to be fully incorporated into computational analysis of chemical information.

Network pharmacology-based study to identify the significant pathways of Lentinula edodes against cancer

Lentinula edodes (LE) is known as a good food source with potent anticancer efficacy, but its active chemical compounds and pathways against cancer have not been revealed. This study was to uncover the active chemical constituents and pathways of LE against cancer through network pharmacology. The chemical compositions were recognized by gas chromatography-mass spectrometry (GC-MS) and filtered drug-like compounds (DLCs) by SwissADME. Targets related to filtered compounds were recognized by two public databases and the final overlapping targets were identified by Venn diagram. Then, protein-protein interaction (PPI) and pathway-target-compound (PTC) networks were built by RStudio. Ultimately, we recognized the key compounds and targets via molecular docking test (MDT). A total of 33 compounds from LE were accepted by Lipinski's rule were selected as DLCs. The 33 compounds were associated with 108 targets and a key target (cyclooxygenase2 [COX2]) was identified through PPI networks. Most significantly, inactivation of pathways in cancer and activation of peroxisome proliferator activated receptor signaling pathway were significant pathways of LE. On MDT, we identified a key compound (Indole, 2-methyl-3-phenyl) on COX2 related to inactivation of athways in cancer, additionally, the number of 6 ergostane steroids was associated with the two pathways might be dual efficacy to alleviate inflammation against cancer. Overall, 13 targets, 11 compounds, and 2 key pathways of LE were identified as the significant elements to treat cancer. Hence, this study shows therapeutic evidence to verify the promising clinical effect of LE on cancer, suggesting that LE might be an important mushroom against cancer. PRACTICAL APPLICATIONS: Lentinula edodes (LE) has been used widely in cuisine as well as alternative medicines, especially, for anticancer. The LE has rich nutritional compounds including proteins, vitamins, polyphenols, and glucans, however, most of which have a critical hurdle as poor bioavailability not to be applicable for pharmaceuticals. Its main cause is very hydrophilic property. Thus, we adopted GC-MS analysis to identify lipophilic compounds to enhance cell permeability involved in bioavailability. The compounds selected from LE were confirmed by Lipinski's rule for drug-like-compounds (DLCs). Then, we retrieved targets associated with DLCs, and multiple pathways, multiple targets, and multiple compounds against cancer on network-based analysis. In summary, our study reveals the medicinal value of LE on cancer based on the multicomponents. Overall, the aim of this work is to represent the pharmacological evidence to reveal the therapeutic efficacy of AC on cancer, suggesting that DLCs from AC might be alleviators to dampen cancer.

HelixADMET: a robust and endpoint extensible ADMET system incorporating self-supervised knowledge transfer

MOTIVATION

Accurate ADMET (an abbreviation for "absorption, distribution, metabolism, excretion, and toxicity") predictions can efficiently screen out undesirable drug candidates in the early stage of drug discovery. In recent years, multiple comprehensive ADMET systems that adopt advanced machine learning models have been developed, providing services to estimate multiple endpoints. However, those ADMET systems usually suffer from weak extrapolation ability. First, due to the lack of labelled data for each endpoint, typical machine learning models perform frail for the molecules with unobserved scaffolds. Second, most systems only provide fixed built-in endpoints and cannot be customised to satisfy various research requirements. To this end, we develop a robust and endpoint extensible ADMET system, HelixADMET (H-ADMET). H-ADMET incorporates the concept of self-supervised learning to produce a robust pre-trained model. The model is then fine-tuned with a multi-task and multi-stage framework to transfer knowledge between ADMET endpoints, auxiliary tasks, and self-supervised tasks.

RESULTS

Our results demonstrate that H-ADMET achieves an overall improvement of 4%, compared with existing ADMET systems on comparable endpoints. Additionally, the pre-trained model provided by H-ADMET can be fine-tuned to generate new and customised ADMET endpoints, meeting various demands of drug research and development requirements.

AVAILABILITY

H-ADMET is freely accessible at https://paddlehelix.baidu.com/app/drug/admet/train.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

A feature transferring workflow between data-poor compounds in various tasks

Compound screening by in silico approaches has advantages in identifying high-activity leading compounds and can predict the safety of the drug. A key challenge is that the number of observations of drug activity and toxicity accumulation varies by target in different datasets, some of which are more understudied than others. Owing to an overall insufficiency and imbalance of drug data, it is hard to accurately predict drug activity and toxicity of multiple tasks by the existing models. To solve this problem, this paper proposed a two-stage transfer learning workflow to develop a novel prediction model, which can accurately predict drug activity and toxicity of the targets with insufficient observations. We built a balanced dataset based on the Tox21 dataset and developed a drug activity and toxicity prediction model based on Siamese networks and graph convolution to produce multitasking output. We also took advantage of transfer learning from data-rich targets to data-poor targets. We showed greater accuracy in predicting the activity and toxicity of compounds to targets with rich data and poor data. In Tox21, a relatively rich dataset, the prediction model accuracy for classification tasks was 0.877 AUROC. In the other five unbalanced datasets, we also found that transfer learning strategies brought the accuracy of models to a higher level in understudied targets. Our models can overcome the imbalance in target data and predict the compound activity and toxicity of understudied targets to help prioritize upcoming biological experiments.

Exploring nature's bounty: identification of Withania somnifera as a promising source of therapeutic agents against COVID-19 by virtual screening and in silico evaluation

Coronaviruses are etiological agents of extreme human and animal infection resulting in abnormalities primarily in the respiratory tract. Presently, there is no defined COVID-19 intervention and clinical trials of prospective therapeutic agents are still in the nascent stage. Withania somnifera (L.) Dunal (WS), is an important medicinal plant in Ayurveda. The present study aimed to evaluate the antiviral potential of selected WS phytoconstituents against the novel SARS-CoV-2 target proteins and human ACE2 receptor using in silico methods. Most of the phytoconstituents displayed good absorption and transport kinetics and were also found to display no associated mutagenic or adverse effect(s). Molecular docking analyses revealed that most of the WS phytoconstituents exhibited potent binding to human ACE2 receptor, SAR-CoV and SARS-CoV-2 spike glycoproteins as well as the two main SARS-CoV-2 proteases. Most of the phytoconstituents were predicted to undergo Phase-I metabolism prior to excretion. All phytoconstituents had favorable bioactivity scores with respect to various receptor proteins and target enzymes. SAR analysis revealed that the number of oxygen atoms in the withanolide backbone and structural rearrangements were crucial for effective binding. Molecular simulation analyses of SARS-CoV-2 spike protein and papain-like protease with Withanolides A and B, respectively, displayed a stability profile at 300 K and constant RMSDs of protein side chains and Cα atoms throughout the simulation run time. In a nutshell, WS phytoconstituents warrant further investigations in vitro and in vivo to unravel their molecular mechanism(s) and modes of action for their future development as novel antiviral agents against COVID-19.