Current development of integrated web servers for preclinical safety and pharmacokinetics assessments in drug development

hERG toxicity prediction in early drug discovery using extreme gradient boosting and isometric stratified ensemble mapping

Blockade of the human Ether-à-go-go Related Gene (hERG) potassium channel by small molecules can prolong the QT interval, leading to fatal cardiotoxicity. Numerous drugs have been withdrawn from the market due to cardiac side effects, underscoring the need for early identification of hERG toxicity. Despite several classification machine learning (ML) models having been developed to this end, robustness, class imbalance, and interpretability are still challenges. Using the largest public database of hERG inhibition, this work integrates eXtreme Gradient Boosting (XGBoost) with Isometric Stratified Ensemble (ISE) mapping (XGB + ISE map) to enhance hERG prediction. An XGBoost consensus model was developed using balanced training sets and diverse variable subsets, resulting in robust models less affected by class imbalance. The model demonstrated competitive predictive performance, achieving a balance between sensitivity (SE = 0.83) and specificity (SP = 0.90) through exhaustive validation. ISE mapping estimated the model applicability domain and improved prediction confidence evaluation and compound selection by stratifying data. Refined variable selection procedures enhanced model interpretability. Variable importance analysis highlights key molecular determinants (peoe_VSA8, ESOL, SdssC, MaxssO, nRNR2, MATS1i, nRNHR, nRNH2) associated with hERG inhibition. The XGB + ISE map strategy provides an effective approach to identifying promising molecules in drug discovery campaigns with reduced hERG inhibition risk.

Artificial intelligence to predict inhibitors of drug-metabolizing enzymes and transporters for safer drug design

INTRODUCTION

Drug-metabolizing enzymes (DMEs) and transporters (DTs) play integral roles in drug metabolism and drug-drug interactions (DDIs) which directly impact drug efficacy and safety. It is well-established that inhibition of DMEs and DTs often leads to adverse drug reactions (ADRs) and therapeutic failure. As such, early prediction of such inhibitors is vital in drug development. In this context, the limitations of the traditional in vitro assays and QSAR models methods have been addressed by harnessing artificial intelligence (AI) techniques.

AREAS COVERED

This narrative review presents the insights gained from the application of AI for predicting DME and DT inhibitors over the past decade. Several case studies demonstrate successful AI applications in enzyme-transporter interaction prediction, and the authors discuss workflows for integrating these predictions into drug design and regulatory frameworks.

EXPERT OPINION

The application of AI in predicting DME and DT inhibitors has demonstrated significant potential toward enhancing drug safety and effectiveness. However, critical challenges involve the data quality, biases, and model transparency. The availability of diverse, high-quality datasets alongside the integration of pharmacokinetic and genomic data are essential. Lastly, the collaboration among computational scientists, pharmacologists, and regulatory bodies is pyramidal in tailoring AI tools for personalized medicine and safer drug development.

Natural Pangenotypic Hepatitis C Virus Preventives- a Hope With Terfenadine

Hepatitis C viral infections are a leading cause of long-term liver disease and hepatocellular cancer in drug addicts and other at-risk people, including those who require frequent blood transfusions. Current drug treatments are expensive and have numerous side effects; hence, more affordable treatment plans are required. Furthermore, there are not any preventives in the market. HCV comes in 8 genotypes, and the prevalence of genotypes varies worldwide. Hence, affordable, pangenotypic preventives are needed. In this study, we selected 83 natural compounds that have been shown to have anti-HCV properties. In silico screening was done via docking assays to check whether any of these compounds could potentially bind to the receptor interaction site of surface protein-E2 of genotypes 1a, 1b, 2a, 3a, and 3b. From the binding energies, five compounds were selected that exhibited pangenotypic effects. MD simulation was conducted on these compounds to assess their interaction properties. ADME properties and further drug likeliness revealed one of the compounds, Terfenadine, to be similar to an anti-allergy drug, Fexofenadine. To assess in vitro validation, a binding assay was set up using E2-GFP expressed in HEK293T cells with the primary receptor CD81. It was observed that Terfenadine (used as the derivative Fexofenadine) could inhibit an interaction between CD81 and E2. We conclude that there is potential for Terfenadine/fexofenadine to be effective as a preventive against the HCV genotypes 1a, 1b, 2a, 3a, and 3b. Further clinical validation is required to confirm these findings.

A Comprehensive Evaluation of the Neuropharmacological Potential of Methanolic Leaf Extract of Acanthus ebracteatus (Vahl.) Using Experimental and In Silico Approaches

This study was undertaken to assess the antioxidant and neuropharmacological potentials of the methanol leaf extract of Acanthus ebracteatus (MAEL) through experimental and in silico methods. The phytochemical screening (PS) and GC-MS (gas chromatography-mass spectrometry) identified 28 phytochemicals with different classes in nature in MAEL. The MAEL revealed better antioxidant activity through various in vitro antioxidant assays. Additionally, in the tail suspension test (TST) and forced swimming test (FST), a dose-dependent reduction in immobility time was observed indicating antidepressant activity. In the elevated plus maze test (EPM), MAEL led to increased time spent and more entries in the open arms. At the same time, the hole board test (HBT) demonstrated an increase in head dipping compared to the control, both indicating anxiolytic activity. Moreover, a dose-dependent reduction in locomotor activities was observed in both the open field test (OFT) and hole cross test (HCT). Molecular docking showed better binding affinities of two compounds, CID-518982 and CID-236641. ADME/T analysis revealed good drug likeliness with no toxicity. Finally, the simulation demonstrated better structural stability with no significant fluctuations of the compounds with the selected receptors. In this study, compounds CID-518982 and CID-236641 might serve as drug candidates for treating anxiety and depression.

New Approach as Inhibitor Against Head-Neck Cancer by In silico, DFT, FMOs, Docking, Molecular Dynamic, and ADMET of Euphorbia tirucalli (Pencil Cactus)

BACKGROUND

Head and neck cancer (HNC) is on the rise worldwide, endangering lives and straining healthcare systems in both developing and developed nations. Despite the availability of a number of therapy options, the success rate for treating and controlling head and neck cancer remains dismal. To combat the aggressiveness and drug resistance of Epstein-Barr virus (EBV)-positive Head-Neck cancer cells, this study looks into the potential of Euphorbia tirucalli (pencil cactus) leaf extract.

OBJECTIVES

The goal of this study is to identify prospective therapeutic candidates from the extract of Euphorbia tirucalli (pencil cactus) leaves, which have the ability to inhibit Epstein-Barr virus (EBV)-positive Head- Neck cancer cells.

MATERIALS AND METHODS

The thirteen most important chemical components found in Euphorbia tirucalli (pencil cactus) leaves were analyzed by means of molecular modeling techniques such as Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET), Quantum Mechanics (QM) calculation, molecular docking, and molecular dynamics (MD) simulations. Using the Prediction of Activity Spectra for Substances (PASS) model, we assess the potency of these compounds. Important molecular properties such as chemical potential, electronegativity, hardness, and softness can be determined with the use of quantum chemical calculations employing HOMO-LUMO analysis. These drugs' safety and toxicological characteristics are better understood to assessments of their pharmacokinetics and ADMET. Finally, molecular dynamics simulations are employed to verify binding interactions and assess the stability of docked complexes.

RESULTS

The molecular docking analysis identifies ligands (01), (02), and (10) as strong competitors, with strong binding affinity for the Epstein-Barr virus (EBV)-positive Head-Neck cancer cell line. Not only do the ligands (01), (02), and (10) match the criteria for a potential new inhibitor of head-neck cancer, but they also outperform the present FDA-approved treatment.

CONCLUSION

Taraxerol, euphol, and ephorginol, three phytochemicals isolated from the leaves of the Euphorbia tirucalli (pencil cactus), have been identified as effective anti-cancer agents with the potential to serve as a foundation for novel head-neck cancer therapies, particularly those targeting the Epstein-Barr virus (EBV)-overexpressing subtype of this disease. An effective, individualized treatment plan for head-neck cancer is a long way off, but this study is a major step forward that could change the lives of patients and reduce the global burden of this disease.

Multimodal Representation Learning via Graph Isomorphism Network for Toxicity Multitask Learning

Toxicity is paramount for comprehending compound properties, particularly in the early stages of drug design. Due to the diversity and complexity of toxic effects, it became a challenge to compute compound toxicity tasks. To address this issue, we propose a multimodal representation learning model, termed multimodal graph isomorphism network (MMGIN), to address this challenge for compound toxicity multitask learning. Based on fingerprints and molecular graphs of compounds, our MMGIN model incorporates a multimodal representation learning model to acquire a comprehensive compound representation. This model adopts a two-channel structure to independently learn fingerprint representation and molecular graph representation. Subsequently, two feedforward neural networks utilize the learned multimodal compound representation to perform multitask learning, encompassing compound toxicity classification and multiple compound category classification simultaneously. To test the effectiveness of our model, we constructed a novel data set, termed the compound toxicity multitask learning (CTMTL) data set, derived from the TOXRIC data set. We compare our MMGIN model with other representative machine learning and deep learning models on the CTMTL and Tox21 data sets. The experimental results demonstrate significant advancements achieved by our MMGIN model. Furthermore, the ablation study underscores the effectiveness of the introduced fingerprints, molecular graphs, the multimodal representation learning model, and the multitask learning model, showcasing the model's superior predictive capability and robustness.

In silico analysis of potential inhibitors for breast cancer targeting 17beta-hydroxysteroid dehydrogenase type 1 (17beta-HSD1) catalyses

Breast cancer (BC) is still one of the major issues in world health, especially for women, which necessitates innovative therapeutic strategies. In this study, we investigated the efficacy of retinoic acid derivatives as inhibitors of 17beta-hydroxysteroid dehydrogenase type 1 (17beta-HSD1), which plays a crucial role in the biosynthesis and metabolism of oestrogen and thereby influences the progression of BC and, the main objective of this investigation is to identify the possible drug candidate against BC through computational drug design approach including PASS prediction, molecular docking, ADMET profiling, molecular dynamics simulations (MD) and density functional theory (DFT) calculations. The result has reported that total eight derivatives with high binding affinity and promising pharmacokinetic properties among 115 derivatives. In particular, ligands 04 and 07 exhibited a higher binding affinity with values of -9.9 kcal/mol and -9.1 kcal/mol, respectively, than the standard drug epirubicin hydrochloride, which had a binding affinity of -8.2 kcal/mol. The stability of the ligand-protein complexes was further confirmed by MD simulations over a 100-ns trajectory, which included assessments of hydrogen bonds, root mean square deviation (RMSD), root mean square Fluctuation (RMSF), dynamic cross-correlation matric (DCCM) and principal component analysis. The study emphasizes the need for experimental validation to confirm the therapeutic utility of these compounds. This study enhances the computational search for new BC drugs and establishes a solid foundation for subsequent experimental and clinical research.

Enhancing Molecular Property Prediction through Task-Oriented Transfer Learning: Integrating Universal Structural Insights and Domain-Specific Knowledge

Precisely predicting molecular properties is crucial in drug discovery, but the scarcity of labeled data poses a challenge for applying deep learning methods. While large-scale self-supervised pretraining has proven an effective solution, it often neglects domain-specific knowledge. To tackle this issue, we introduce Task-Oriented Multilevel Learning based on BERT (TOML-BERT), a dual-level pretraining framework that considers both structural patterns and domain knowledge of molecules. TOML-BERT achieved state-of-the-art prediction performance on 10 pharmaceutical datasets. It has the capability to mine contextual information within molecular structures and extract domain knowledge from massive pseudo-labeled data. The dual-level pretraining accomplished significant positive transfer, with its two components making complementary contributions. Interpretive analysis elucidated that the effectiveness of the dual-level pretraining lies in the prior learning of a task-related molecular representation. Overall, TOML-BERT demonstrates the potential of combining multiple pretraining tasks to extract task-oriented knowledge, advancing molecular property prediction in drug discovery.

Computational screening of natural MtbDXR inhibitors for novel anti-tuberculosis compound discovery

DXR (1-deoxy-d-xylulose-5-phosphate reductoisomerase) is an essential enzyme in the Methylerythritol 4-phosphate (MEP) pathway, which is used by M. tuberculosis and a few other pathogens. This essential enzyme in the isoprenoid synthesis pathway has been previously reported as an important target for antibiotic drug design. However, till now, there is no record of any drug-like safe molecule to inhibit MtbDXR. Numerous plant species have been traditionally used for tuberculosis therapies. In this study, we selected six plant species with anti-tubercular properties. The chemoinformatic screening was performed on 352 phytochemicals from those plants against the MtbDXR protein. After molecular docking analysis, we filtered the top five compounds, CID: 5280443 (Apigenin), CID: 3220 (Emodin), CID: 5280863 (Kaempferol), CID: 5280445 (Luteolin), and CID: 6101979 (beta-Hydroxychalcone), based on binding affinity. Molecular dynamics simulations disclosed the stability of the compounds at the active site of the proteins. Finally, in silico ADME and toxicity evaluations confirmed the compounds to be effective and safe for oral administration. Thus, our findings identified three drug-like safe molecules- Apigenin, Kaempferol, and beta-Hydroxychalcone, that showed good stability in the protein's active site. The results of this computational approach may act as an initial instruction for future in vitro and in vivo testing to identify natural drug-like compounds to treat tuberculosis.Communicated by Ramaswamy H. Sarma.

In silico targeting breast cancer biomarkers by applying rambutan (Nephelium lappaceum) phytocompounds

Worldwide, breast cancer is the leading type of cancer among women. Overexpression of various prognostic indicators, including nuclear receptors, is linked to breast cancer features. To date, no effective drug has been discovered to block the proliferation of breast cancer cells. This study has been designed to discover target-based small molecular-like natural drug candidates that have anti-cancer potential without causing any serious side effects. A comprehensive substrate-based drug design was carried out to discover the potential plant compounds against the target breast cancer biomarkers including phytochemicals screening, active site identification, molecular docking, pharmacokinetic (PK) properties prediction, toxicity prediction, and molecular dynamics (MD) simulation approaches. Twenty plant compounds extracted from the rambutan (Nephelium lappaceum) were obtained from PubChem Database; and screened against the breast cancer biomarkers including estrogen receptor (ER), progesterone receptor (PR), and androgen receptor (AR). The best docking interaction was chosen based on the higher binding affinity. Analyzing the pharmacokinetic properties and toxicity prediction results indicated that the fifteen selected plant compounds have good potency without toxicity and are safe for humans. Four phytochemicals with a higher binding affinity were chosen for each breast cancer biomarker to study their stability in interaction with the target proteins using MD simulation. Among the above compounds, Ellagic acid showed the high binding affinity against all three breast cancer biomarkers.Communicated by Ramaswamy H. Sarma.

Recent Studies of Artificial Intelligence on In Silico Drug Absorption

Absorption is an important area of research in pharmacochemistry and drug development, because the drug has to be absorbed before any drug effects can occur. Furthermore, the ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) profile of drugs can be directly and considerably altered by modulating factors affecting absorption. Many drugs in development fail because of poor absorption. The research and continuous efforts of researchers in recent years have brought many successes and promises in drug absorption property prediction, especially in silico, which helps to reduce the time and cost significantly for screening undesirable drug candidates. In this report, we explicitly provide an overview of recent in silico studies on predicting absorption properties, especially from 2019 to the present, using artificial intelligence. Additionally, we have collected and investigated public databases that support absorption prediction research. On those grounds, we also proposed the challenges and development directions of absorption prediction in the future. We hope this review can provide researchers with valuable guidelines on absorption prediction to facilitate the development of newer approaches in drug discovery.

Revolutionizing anti-cancer drug discovery against breast cancer and lung cancer by modification of natural genistein: an advanced computational and drug design approach

Breast and lung cancer are two of the most lethal forms of cancer, responsible for a disproportionately high number of deaths worldwide. Both doctors and cancer patients express alarm about the rising incidence of the disease globally. Although targeted treatment has achieved enormous advancements, it is not without its drawbacks. Numerous medicines and chemotherapeutic drugs have been authorized by the FDA; nevertheless, they can be quite costly and often fall short of completely curing the condition. Therefore, this investigation has been conducted to identify a potential medication against breast and lung cancer through structural modification of genistein. Genistein is the active compound in Glycyrrhiza glabra (licorice), and it exhibits solid anticancer efficiency against various cancers, including breast cancer, lung cancer, and brain cancer. Hence, the design of its analogs with the interchange of five functional groups-COOH, NH2 and OCH3, Benzene, and NH-CH2-CH2-OH-have been employed to enhance affinities compared to primary genistein. Additionally, advanced computational studies such as PASS prediction, molecular docking, ADMET, and molecular dynamics simulation were conducted. Firstly, the PASS prediction spectrum was analyzed, revealing that the designed genistein analogs exhibit improved antineoplastic activity. In the prediction data, breast and lung cancer were selected as primary targets. Subsequently, other computational investigations were gradually conducted. The mentioned compounds have shown acceptable results for in silico ADME, AMES toxicity, and hepatotoxicity estimations, which are fundamental for their oral medication. It is noteworthy that the initial binding affinity was only -8.7 kcal/mol against the breast cancer targeted protein (PDB ID: 3HB5). However, after the modification of the functional group, when calculating the binding affinities, it becomes apparent that the binding affinities increase gradually, reaching a maximum of -11.0 and -10.0 kcal/mol. Similarly, the initial binding affinity was only -8.0 kcal/mol against lung cancer (PDB ID: 2P85), but after the addition of binding affinity, it reached -9.5 kcal/mol. Finally, a molecular dynamics simulation was conducted to study the molecular models over 100 ns and examine the stability of the docked complexes. The results indicate that the selected complexes remain highly stable throughout the 100-ns molecular dynamics simulation runs, displaying strong correlations with the binding of targeted ligands within the active site of the selected protein. It is important to further investigate and proceed to clinical or wet lab experiments to determine the practical value of the proposed compounds.

Herbal concoction Unveiled: A computational analysis of phytochemicals' pharmacokinetic and toxicological profiles using novel approach methodologies (NAMs)

Herbal medications have an extensive history of use in treating various diseases, attributed to their perceived efficacy and safety. Traditional medicine practitioners and contemporary healthcare providers have shown particular interest in herbal syrups, especially for respiratory illnesses associated with the SARS-CoV-2 virus. However, the current understanding of the pharmacokinetic and toxicological properties of phytochemicals in these herbal mixtures is limited. This study presents a comprehensive computational analysis utilizing novel approach methodologies (NAMs) to investigate the pharmacokinetic and toxicological profiles of phytochemicals in herbal syrup, leveraging in-silico techniques and prediction tools such as PubChem, SwissADME, and Molsoft's database. Although molecular dynamics, docking, and broader system-wide analyses were not considered, future studies hold potential for further investigation in these areas. By combining drug-likeness with molecular simulation, researchers identify diverse phytochemicals suitable for complex medication development examining their pharmacokinetic-toxicological profiles in phytopharmaceutical syrup. The study focuses on herbal solutions for respiratory infections, with the goal of adding to the pool of all-natural treatments for such ailments. This research has the potential to revolutionize environmental and alternative medicine by leveraging in-silico models and innovative analytical techniques to identify novel phytochemicals with enhanced therapeutic benefits and explore network-based and systems biology approaches for a deeper understanding of their interactions with biological systems. Overall, our study offers valuable insights into the computational analysis of the pharmacokinetic and toxicological profiles of herbal concoction. This paves the way for advancements in environmental and alternative medicine. However, we acknowledge the need for future studies to address the aforementioned topics that were not adequately covered in this research.

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.

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

Drug metabolism and excretion play crucial roles in determining the efficacy and safety of drug candidates, and predicting these processes is an essential part of drug discovery and development. In recent years, artificial intelligence (AI) has emerged as a powerful tool for predicting drug metabolism and excretion, offering the potential to speed up drug development and improve clinical success rates. This review highlights recent advances in AI-based drug metabolism and excretion prediction, including deep learning and machine learning algorithms. We provide a list of public data sources and free prediction tools for the research community. We also discuss the challenges associated with the development of AI models for drug metabolism and excretion prediction and explore future perspectives in the field. We hope this will be a helpful resource for anyone who is researching in silico drug metabolism, excretion, and pharmacokinetic properties.

Recent Studies of Artificial Intelligence on In Silico Drug Distribution Prediction

Drug distribution is an important process in pharmacokinetics because it has the potential to influence both the amount of medicine reaching the active sites and the effectiveness as well as safety of the drug. The main causes of 90% of drug failures in clinical development are lack of efficacy and uncontrolled toxicity. In recent years, several advances and promising developments in drug distribution property prediction have been achieved, especially in silico, which helped to drastically reduce the time and expense of screening undesired drug candidates. In this study, we provide comprehensive knowledge of drug distribution background, influencing factors, and artificial intelligence-based distribution property prediction models from 2019 to the present. Additionally, we gathered and analyzed public databases and datasets commonly utilized by the scientific community for distribution prediction. The distribution property prediction performance of five large ADMET prediction tools is mentioned as a benchmark for future research. On this basis, we also offer future challenges in drug distribution prediction and research directions. We hope that this review will provide researchers with helpful insight into distribution prediction, thus facilitating the development of innovative approaches for drug discovery.

Pharmacoinformatics and molecular dynamics simulation-based phytochemical screening of neem plant (Azadiractha indica) against human cancer by targeting MCM7 protein

Minichromosome maintenance complex component 7 (MCM7) belongs to the minichromosome maintenance family that is important for the initiation of eukaryotic DNA replication. Overexpression of the MCM7 protein is relative to cellular proliferation and responsible for aggressive malignancy in various cancers. Mechanistically, inhibition of MCM7 significantly reduces the cellular proliferation associated with cancer. To date, no effective small molecular candidate has been identified that can block the progression of cancer induced by the MCM7 protein. Therefore, the study has been designed to identify small molecular-like natural drug candidates against aggressive malignancy associated with various cancers by targeting MCM7 protein. To identify potential compounds against the targeted protein a comprehensive in silico drug design including molecular docking, ADME (Absorption, Distribution, Metabolism and Excretion), toxicity, and molecular dynamics (MD) simulation approaches has been applied. Seventy phytochemicals isolated from the neem tree (Azadiractha indica) were retrieved and screened against MCM7 protein by using the molecular docking simulation method, where the top four compounds have been chosen for further evaluation based on their binding affinities. Analysis of ADME and toxicity properties reveals the efficacy and safety of the selected four compounds. To validate the stability of the protein-ligand complex structure MD simulations approach has also been performed to the protein-ligand complex structure, which confirmed the stability of the selected three compounds including CAS ID:105377-74-0, CID:12308716 and CID:10505484 to the binding site of the protein. In the study, a comprehensive data screening process has performed based on the docking, ADMET properties, and MD simulation approaches, which found a good value of the selected four compounds against the targeted MCM7 protein and indicates as a promising and effective human anticancer agent.