Recent progresses in the exploration of machine learning methods as in-silico ADME prediction tools

Comparative Analysis of the Structure and Pharmacological Properties of Some Piperidines and Host-Guest Complexes of β-Cyclodextrin

Pain and anesthesia are a problem for all physicians. Scientists from different countries are constantly searching for new anesthetic agents and methods of general anesthesia. In anesthesiology, the role and importance of local anesthesia always remain topical. In the present work, a comparative analysis of the results of pharmacological studies on models of the conduction and terminal anesthesia, as well as acute toxicity studies of the inclusion complex of 1-methyl-4-ethynyl-4-hydroxypiperidine (MEP) with β-cyclodextrin, was carried out. A virtual screening and comparative analysis of pharmacological activity were also performed on a number of the prepared piperidine derivatives and their host-guest complexes of β-cyclodextrin to identify the structure-activity relationship. Various programs were used to study biological activity in silico. For comparative analysis of chemical and pharmacological properties, data from previous works were used. For some piperidine derivatives, new dosage forms were prepared as beta-cyclodextrin host-guest complexes. Some compounds were recognized as promising local anesthetics. Pharmacological studies have shown that KFCD-7 is more active than reference drugs in terms of local anesthetic activity and acute toxicity but is less active than host-guest complexes, based on other piperidines. This fact is in good agreement with the predicted results of biological activity.

Human Brain Penetration Prediction Using Scaling Approach from Animal Machine Learning Models

Machine learning (ML) approaches have been applied to predicting drug pharmacokinetic properties. Previously, we predicted rat unbound brain-to-plasma ratio (Kpuu,brain) by ML models. In this study, we aimed to predict human Kpuu,brain through animal ML models. First, we re-evaluated ML models for rat Kpuu,brain prediction by using trendy open-source packages. We then developed ML models for monkey Kpuu,brain prediction. Leave-one-out cross validation was utilized to rationally build models using a relatively small dataset. After establishing the monkey and rat ML models, human Kpuu,brain prediction was achieved by implementing the animal models considering appropriate scaling methods. Mechanistic NeuroPK models for the identical monkey and human dataset were treated as the criteria for comparison. Results showed that rat Kpuu,brain predictivity was successfully replicated. The optimal ML model for monkey Kpuu,brain prediction was superior to the NeuroPK model, where accuracy within 2-fold error was 78% (R2 = 0.76). For human Kpuu,brain prediction, rat model using relative expression factor (REF), scaled transporter efflux ratios (ERs), and monkey model using in vitro ERs can provide comparable predictivity to the NeuroPK model, where accuracy within 2-fold error was 71% and 64% (R2 = 0.30 and 0.52), respectively. We demonstrated that ML models can deliver promising Kpuu,brain prediction with several advantages: (1) predict reasonable animal Kpuu,brain; (2) prospectively predict human Kpuu,brain from animal models; and (3) can skip expensive monkey studies for human prediction by using the rat model. As a result, ML models can be a powerful tool for drug Kpuu,brain prediction in the discovery stage.

Simultaneous biomonitoring of volatile organic compounds' metabolites in human urine samples using a novel in-syringe based fast urinary metabolites extraction (FaUMEx) technique coupled with UHPLC-MS/MS analysis

Assessing the impact of human exposure to environmental toxicants is often crucial to biomonitoring the exposed dose. In this work, we report a novel fast urinary metabolites extraction (FaUMEx) technique coupled with UHPLC-MS/MS analysis for the highly sensitive and simultaneous biomonitoring of the five major urinary metabolites (thiodiglycolic acid, s-phenylmercapturic acid, t,t-muconic acid, mandelic acid, and phenyl glyoxylic acid) of common volatile organic compounds' (VOCs) exposure (vinyl chloride, benzene, styrene, and ethylbenzene) in human. FaUMEx technique comprises of two-steps, liquid-liquid microextraction was performed first in an extraction syringe using 1 mL of methanol (pH 3) as an extraction solvent and then, the extractant was passed through a clean-up syringe (pre-packed-with various sorbents including 500 mg anhydrous MgSO₄, 50 mg C18, and 50 mg SiO₂) to obtain the high order of matrice clean-up and preconcentration efficiency. The developed method displayed excellent linearity, and the correlation coefficients were >0.998 for all the target metabolites with detection and quantification limits of 0.02-0.24 ng mL^-1 and 0.05-0.72 ng mL^-1, respectively. Furthermore, the matrix effects were < ±5%, and inter and intra-day precision were <9%. Moreover, the presented method was applied and validated to real sample analysis for biomonitoring of VOC's exposure levels. The results showed that the developed FaUMEx-UHPLC-MS/MS method is fast, simple, low-cost, low-solvent consumption, high sensitivity with good accuracy and precision for five targeted urinary VOCs' metabolites. Therefore, the presented dual-syringe mode FaUMEx strategy with UHPLC-MS/MS technique can be applied to biomonitoring of various urinary metabolites to assess human exposure to environmental toxicants.

Transformer-based deep learning method for optimizing ADMET properties of lead compounds

A successful drug needs to exhibit both effective pharmacodynamics (PD) and safe pharmacokinetics (PK). However, the coordinated optimization of PD and PK properties in molecule generation tasks remains a great challenge for most existing methods, especially when they focus on the pursuit of affinity and selectivity for the lead compound. Thus, molecular optimization for PK properties is a critical step in the drug discovery pipeline, in which absorption, distribution, metabolism, excretion and toxicity (ADMET) property predictive models play an increasingly important role by providing an effective method to assess multiple PK properties of compounds. Here, we proposed a Graph Bert-based ADMET prediction model that achieves state-of-the-art performance on the public dataset Therapeutics Data Commons (TDC) by combining molecular graph features and descriptor features, with 11 tasks ranked first and 20 tasks ranked in the top 3. Based on this prediction model, we trained a Transformer model with multiple properties as constraints for learning the structural transformations involved in MMP and the accompanying property changes. The experimental results show that the trained Constraints-Transformer can implement targeted modifications to the starting molecule, while preserving the core scaffold. Moreover, molecular docking and binding mode analysis demonstrate that the optimized molecules still retain the activity and selectivity for biological targets. Therefore, the proposed method accounts for biological activity and ADMET properties simultaneously. Finally, a webserver containing ADMET property prediction and molecular optimization functions is provided, enabling chemists to improve the properties of starting molecules individually.

Anticancer potential of phytochemicals from Oroxylum indicum targeting Lactate Dehydrogenase A through bioinformatic approach

In recent years, small molecule inhibition of LDHA (Lactate Dehydrogenase A) has evolved as an appealing option for anticancer therapy. LDHA catalyzes the interconversion of pyruvate and lactate in the glycolysis pathway to play a crucial role in aerobic glycolysis. Therefore, in the current investigation LDHA was targeted with bioactive phytochemicals of an ethnomedicinally important plant species Oroxylum indicum (L.) Kurz. A total of 52 phytochemicals were screened against LDHA protein through molecular docking, ADMET (Absorption, Distribution, Metabolism, Excretion and Toxicity) assay and molecular dynamics simulation to reveal three potential lead compounds such as Chrysin-7-O-glucuronide (-8.2 kcal/mol), Oroxindin (-8.1 kcal/mol) and Oroxin A (-8.0 kcal/mol). ADMET assay unveiled favorable pharmacokinetic, pharmacodynamic and toxicity properties for all the lead compounds. Molecular dynamics simulation exhibited significant conformational stability and compactness. MM/GBSA free binding energy calculations further corroborated the selection of top candidates where Oroxindin (-46.47 kcal/mol) was found to be better than Chrysin-7-O-glucuronide (-45.72 kcal/mol) and Oroxin A (-37.25 kcal/mol). Aldolase reductase and Xanthine dehydrogenase enzymes were found as potential drug targets and Esculin, the FDA approved drug was identified as structurally analogous to Oroxindin. These results could drive in establishing novel medications targeting LDHA to fight cancer.

Integrated bioinformatics-cheminformatics approach toward locating pseudo-potential antiviral marine alkaloids against SARS-CoV-2-Mpro

The emergence of the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) with the most contagious variants, alpha (B.1.1.7), beta (B.1.351), delta (B.1.617.2), and Omicron (B.1.1.529) has continuously added a higher number of morbidity and mortality, globally. The present integrated bioinformatics-cheminformatics approach was employed to locate potent antiviral marine alkaloids that could be used against SARS-CoV-2. Initially, 57 antiviral marine alkaloids and two repurposing drugs were selected from an extensive literature review. Then, the putative target enzyme SARS-CoV-2 main protease (SARS-CoV-2-Mpro) was retrieved from the protein data bank and carried out a virtual screening-cum-molecular docking study with all candidates using PyRx 0.8 and AutoDock 4.2 software. Further, the molecular dynamics (MD) simulation of the two most potential alkaloids and a drug docking complex at 100 ns (with two ligand topology files from PRODRG and ATB server, separately), the molecular mechanics/Poisson-Boltzmann surface area (MM/PBSA) free energy, and contributions of entropy were investigated. Then, the physicochemical-toxicity-pharmacokinetics-drug-likeness profiles, the frontier molecular orbitals energies (highest occupied molecular orbital, lowest unoccupied molecular orbital, and ΔE), and structural-activity relationship were assessed and analyzed. Based on binding energy, 8-hydroxymanzamine (-10.5 kcal/mol) and manzamine A (-10.1 kcal/mol) from all alkaloids with darunavir (-7.9 kcal/mol) and lopinavir (-7.4 kcal/mol) against SARS-CoV-2-Mpro were recorded. The MD simulation (RMSD, RMSF, Rg, H-bond, MM/PBSA binding energy) illustrated that the 8-hydroxymanzamine exhibits a static thermodynamic feature than the other two complexes. The predicted physicochemical, toxicity, pharmacokinetics, and drug-likeness profiles also revealed that the 8-hydroxymanzamine could be used as a potential lead candidate individually and/or synergistically with darunavir or lopinavir to combat SARS-CoV-2 infection after some pharmacological validation.

Integrated RNA-sequencing and network pharmacology approach reveals the protection of Yiqi Huoxue formula against idiopathic pulmonary fibrosis by interfering with core transcription factors

BACKGROUND

Idiopathic pulmonary fibrosis (IPF) is a refractory disease. Therefore, developing effective therapies for IPF is the need of the hour.

PURPOSE

Yiqi Huoxue Formula (YQHX) is an herbal formula comprising three herbal medicines: Ligusticum chuanxiong Hort. (Chuanxiong Rhizoma, CR), Panax notoginseng (Burk.) F. H. Chen (Notoginseng Radix Et Rhizoma, NR) and Panax ginseng C. A. Mey. (Ginseng Radix Et Rhizoma, GR). This study aims to determine the anti-pulmonary fibrosis effect of YQHX and explore its mechanism of action.

STUDY

Design and Methods: The chemical components in the GR, CR and NR extracts were identified by High Performance Liquid Chromatography. A TGF-β1-induced myofibroblast cell model was used to test the anti-fibrosis effect of GR, CR, NR and YQHX. RNA-sequencing was used to identify the differentially expressed genes (DEGs) after YQHX treatment. Subsequently, gene enrichment analysis and key transcription factors (TFs) prediction for YQHX-regulated DEGs was performed. The active constituents of GR, CR and NR were obtained from the Traditional Chinese Medicine Database and Analysis Platform. Targets of the active constituents were predicted using the similarity ensemble approach search server and Swiss Target Prediction tool. YQHX-targeted key TFs that transcribed the DEGs were screened out. Then, the effect of YQHX on the bleomycin-induced pulmonary fibrosis mouse model was studied. Finally, one of the predicted TFs, STAT3, was selected to validate the prediction accuracy.

RESULTS

Seven, two, and five compounds were identified in the GR, CR, and NR extracts, respectively. YQHX and its constituents-GR, CR and NR-inhibited the expression of fibrotic markers, including α -SMA and fibronectin, indicating that YQHX inhibited TGF-β1-induced myofibroblast activation. RNA-sequencing identified 291 genes that were up-regulated in the TGF-β1 group but down-regulated after YQHX treatment. In total, 55 key TFs that transcribed YQHX-regulated targets were predicted. A regulatory network of 24 active ingredients and 232 corresponding targets for YQHX was established. Among YQHX's predicted targets, 20 were TFs. On overlapping YQHX-targeted TFs and DEGs' key TFs, six key TFs, including HIF1A, STAT6, STAT3, PPARA, DDIT3 and AR, were identified as the targets of YQHX. Additionally, YQHX alleviated bleomycin-induced pulmonary fibrosis in a mouse model by inhibiting the phosphorylation of STAT3 in the lungs of pulmonary fibrosis mice.

CONCLUSIONS

This study provides pharmacological support for the use of YQHX in the treatment of IPF. The potential mechanism of action of YQHX is speculated to involve the modulation of core TFs and inhibition of pathogenetic gene expressions in IPF.

Emerging frontiers in virtual drug discovery: From quantum mechanical methods to deep learning approaches

Virtual screening-based approaches to discover initial hit and lead compounds have the potential to reduce both the cost and time of early drug discovery stages, as well as to find inhibitors for even challenging target sites such as protein-protein interfaces. Here in this review, we provide an overview of the progress that has been made in virtual screening methodology and technology on multiple fronts in recent years. The advent of ultra-large virtual screens, in which hundreds of millions to billions of compounds are screened, has proven to be a powerful approach to discover highly potent hit compounds. However, these developments are just the tip of the iceberg, with new technologies and methods emerging to propel the field forward. Examples include novel machine-learning approaches, which can reduce the computational costs of virtual screening dramatically, while progress in quantum-mechanical approaches can increase the accuracy of predictions of various small molecule properties.

Development and Validation of an ADME-Related Gene Signature for Survival, Treatment Outcome and Immune Cell Infiltration in Head and Neck Squamous Cell Carcinoma

ADME genes are a set of genes which are involved in drug absorption, distribution, metabolism, and excretion (ADME). However, prognostic value and function of ADME genes in head and neck squamous cell carcinoma (HNSCC) remain largely unclear. In this study, we established an ADME-related prognostic model through the least absolute shrinkage and selection operator (LASSO) analysis in the Cancer Genome Atla (TCGA) training cohort and its robustness was validated by TCGA internal validation cohort and a Gene Expression Omnibus (GEO) external cohort. The 14-gene signature stratified patients into high- or low-risk groups. Patients with high-risk scores exhibited significantly poorer overall survival (OS) and disease-free survival (DFS) than those with low-risk scores. Receiver operating characteristic (ROC) curve analysis was used to confirm the signature’s predictive efficacy for OS and DFS. Furthermore, gene ontology (GO) and Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway analyses showed that immune-related functions and pathways were enriched, such as lymphocyte activation, leukocyte cell-cell adhesion and T-helper cell differentiation. The Cell-type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT) and other analyses revealed that immune cell (especially B cell and T cell) infiltration levels were significantly higher in the low-risk group. Moreover, patients with low-risk scores were significantly associated with immunotherapy and chemotherapy treatment benefit. In conclusion, we constructed a novel ADME-related prognostic and therapeutic biomarker associated with immune cell infiltration of HNSCC patients.

Nano-bio interactions: A major principle in the dynamic biological processes of nano-assemblies

Controllable nano-assembly with stimuli-responsive groups is emerging as a powerful strategy to generate theranostic nanosystems that meet unique requirements in modern medicine. However, this prospective field is still in a proof-of-concept stage due to the gaps in our understanding of complex-(nano-assemblies)-complex-(biosystems) interactions. Indeed, stimuli-responsive assembly-disassembly is, in and of itself, a process of nano-bio interactions, the key steps for biological fate and functional activity of nano-assemblies. To provide a comprehensive understanding of these interactions in this review, we first propose a 4W1H principle (Where, When, What, Which and How) to delineate the relevant dynamic biological processes, behaviour and fate of nano-assemblies. We further summarize several key parameters that govern effective nano-bio interactions. The effects of these kinetic parameters on ADMET processes (absorption, distribution, metabolism, excretion and transformation) are then discussed. Furthermore, we provide an overview of the challenges facing the evaluation of nano-bio interactions of assembled nanodrugs. We finally conclude with future perspectives on safe-by-design and application-driven-design of nano-assemblies. This review will highlight the dynamic biological and physicochemical parameters of nano-bio interactions and bridge discrete concepts to build a full spectrum understanding of the biological outcomes of nano-assemblies. These principles are expected to pave the way for future development and clinical translation of precise, safe and effective nanomedicines with intelligent theranostic features.

Interpretable-ADMET: a web service for ADMET prediction and optimization based on deep neural representation

MOTIVATION

In the process of discovery and optimization of lead compounds, it is difficult for non-expert pharmacologists to intuitively determine the contribution of substructure to a particular property of a molecule.

RESULTS

In this work, we develop a user-friendly web service, named interpretable-absorption, distribution, metabolism, excretion and toxicity (ADMET), which predict 59 ADMET-associated properties using 90 qualitative classification models and 28 quantitative regression models based on graph convolutional neural network and graph attention network algorithms. In interpretable-ADMET, there are 250 729 entries associated with 59 kinds of ADMET-associated properties for 80 167 chemical compounds. In addition to making predictions, interpretable-ADMET provides interpretation models based on gradient-weighted class activation map for identifying the substructure, which is important to the particular property. Interpretable-ADMET also provides an optimize module to automatically generate a set of novel virtual candidates based on matched molecular pair rules. We believe that interpretable-ADMET could serve as a useful tool for lead optimization in drug discovery.

AVAILABILITY AND IMPLEMENTATION

Interpretable-ADMET is available at http://cadd.pharmacy.nankai.edu.cn/interpretableadmet/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Trends in In Silico Approaches to the Prediction of Biologically Active Peptides in Meat and Meat Products as an Important Factor for Preventing Food-Related Chronic Diseases

The increasing awareness of modern consumers regarding the nutritional and health value of food has changed their preferences, as well their requirements, for food products, including meat and meat products. Expanding the knowledge on the impact of food on human health is currently one of the most important research areas for scientists worldwide, and it is also of interest to consumers who want to consciously compose their daily diets. New research methods, such as in silico techniques, offer solutions to these new challenges. These research methods are preferred over food evaluation, e.g., from meat, because of their advantages, such as low costs, shorter analysis times, and general availability (e.g., online databases), and are often used to design in vitro and, subsequently, in vivo tests. This review focuses on the possible use of in silico computerized methods to assess the potential of food as a source of these health-relevant biomolecules by using examples from the literature on meat and meat products. This review also provides information and important suggestions for analyzing peptides in terms of assessing their best sources, and screening those resistant to digestive factors and that show biological activity. The information provided in this review could contribute to the development of new sources of foods as biomolecules important for preventing or treating food-related chronic diseases, such as obesity, hypertension, and diabetes.

The rise of deep learning and transformations in bioactivity prediction power of molecular modeling tools

The search and design for the better use of bioactive compounds are used in many experiments to best mimic compounds' functions in the human body. However, finding a cost-effective and timesaving approach is a top priority in different disciplines. Nowadays, artificial intelligence (AI) and particularly deep learning (DL) methods are widely applied to improve the precision and accuracy of models used in the drug discovery process. DL approaches have been used to provide more opportunities for a faster, efficient, cost-effective, and reliable computer-aided drug discovery. Moreover, the increasing biomedical data volume in areas, like genome sequences, medical images, protein structures, etc., has made data mining algorithms very important in finding novel compounds that could be drugs, uncovering or repurposing drugs and improving the area of genetic markers-based personalized medicine. Furthermore, deep neural networks (DNNs) have been demonstrated to outperform other techniques such as random forests and SVMs for QSAR studies and ligand-based virtual screening. Despite this, in QSAR studies, the quality of different data sources and potential experimental errors has greatly affected the accuracy of QSAR predictions. Therefore, further researches are still needed to improve the accuracy, selectivity, and sensitivity of the DL approach in building the best models of drug discovery.

XGraphBoost: Extracting Graph Neural Network-Based Features for a Better Prediction of Molecular Properties

Determining the properties of chemical molecules is essential for screening candidates similar to a specific drug. These candidate molecules are further evaluated for their target binding affinities, side effects, target missing probabilities, etc. Conventional machine learning algorithms demonstrated satisfying prediction accuracies of molecular properties. A molecule cannot be directly loaded into a machine learning model, and a set of engineered features needs to be designed and calculated from a molecule. Such hand-crafted features rely heavily on the experiences of the investigating researchers. The concept of graph neural networks (GNNs) was recently introduced to describe the chemical molecules. The features may be automatically and objectively extracted from the molecules through various types of GNNs, e.g., GCN (graph convolution network), GGNN (gated graph neural network), DMPNN (directed message passing neural network), etc. However, the training of a stable GNN model requires a huge number of training samples and a large amount of computing power, compared with the conventional machine learning strategies. This study proposed the integrated framework XGraphBoost to extract the features using a GNN and build an accurate prediction model of molecular properties using the classifier XGBoost. The proposed framework XGraphBoost fully inherits the merits of the GNN-based automatic molecular feature extraction and XGBoost-based accurate prediction performance. Both classification and regression problems were evaluated using the framework XGraphBoost. The experimental results strongly suggest that XGraphBoost may facilitate the efficient and accurate predictions of various molecular properties. The source code is freely available to academic users at https://github.com/chenxiaowei-vincent/XGraphBoost.git.

Artificial intelligence and machine learning-aided drug discovery in central nervous system diseases: State-of-the-arts and future directions

Neurological disorders significantly outnumber diseases in other therapeutic areas. However, developing drugs for central nervous system (CNS) disorders remains the most challenging area in drug discovery, accompanied with the long timelines and high attrition rates. With the rapid growth of biomedical data enabled by advanced experimental technologies, artificial intelligence (AI) and machine learning (ML) have emerged as an indispensable tool to draw meaningful insights and improve decision making in drug discovery. Thanks to the advancements in AI and ML algorithms, now the AI/ML-driven solutions have an unprecedented potential to accelerate the process of CNS drug discovery with better success rate. In this review, we comprehensively summarize AI/ML-powered pharmaceutical discovery efforts and their implementations in the CNS area. After introducing the AI/ML models as well as the conceptualization and data preparation, we outline the applications of AI/ML technologies to several key procedures in drug discovery, including target identification, compound screening, hit/lead generation and optimization, drug response and synergy prediction, de novo drug design, and drug repurposing. We review the current state-of-the-art of AI/ML-guided CNS drug discovery, focusing on blood-brain barrier permeability prediction and implementation into therapeutic discovery for neurological diseases. Finally, we discuss the major challenges and limitations of current approaches and possible future directions that may provide resolutions to these difficulties.

Biomimetic Chromatographic Studies Combined with the Computational Approach to Investigate the Ability of Triterpenoid Saponins of Plant Origin to Cross the Blood-Brain Barrier

Biomimetic (non-cell based in vitro) and computational (in silico) studies are commonly used as screening tests in laboratory practice in the first stages of an experiment on biologically active compounds (potential drugs) and constitute an important step in the research on the drug design process. The main aim of this study was to evaluate the ability of triterpenoid saponins of plant origin to cross the blood-brain barrier (BBB) using both computational methods, including QSAR methodology, and biomimetic chromatographic methods, i.e., High Performance Liquid Chromatography (HPLC) with Immobilized Artificial Membrane (IAM) and cholesterol (CHOL) stationary phases, as well as Bio-partitioning Micellar Chromatography (BMC). The tested compounds were as follows: arjunic acid (Terminalia arjuna), akebia saponin D (Akebia quinata), bacoside A (Bacopa monnieri) and platycodin D (Platycodon grandiflorum). The pharmacokinetic BBB parameters calculated in silico show that three of the four substances, i.e., arjunic acid, akebia saponin D, and bacoside A exhibit similar values of brain/plasma equilibration rate expressed as logPS_Fubrain (the average logPS_Fubrain: -5.03), whereas the logPS_Fubrain value for platycodin D is -9.0. Platycodin D also shows the highest value of the unbound fraction in the brain obtained using the examined compounds (0.98). In these studies, it was found out for the first time that the logarithm of the analyte-micelle association constant (logK_MA) calculated based on Foley's equation can describe the passage of substances through the BBB. The most similar logBB values were obtained for hydrophilic platycodin D, applying both biomimetic and computational methods. All of the obtained logBB values and physicochemical parameters of the molecule indicate that platycodin D does not cross the BBB (the average logBB: -1.681), even though the in silico estimated value of the fraction unbound in plasma is relatively high (0.52). As far as it is known, this is the first paper that shows the applicability of biomimetic chromatographic methods in predicting the penetration of triterpenoid saponins through the BBB.

Artificial intelligence in the early stages of drug discovery

Although the use of computational methods within the pharmaceutical industry is well established, there is an urgent need for new approaches that can improve and optimize the pipeline of drug discovery and development. In spite of the fact that there is no unique solution for this need for innovation, there has recently been a strong interest in the use of Artificial Intelligence for this purpose. As a matter of fact, not only there have been major contributions from the scientific community in this respect, but there has also been a growing partnership between the pharmaceutical industry and Artificial Intelligence companies. Beyond these contributions and efforts there is an underlying question, which we intend to discuss in this review: can the intrinsic difficulties within the drug discovery process be overcome with the implementation of Artificial Intelligence? While this is an open question, in this work we will focus on the advantages that these algorithms provide over the traditional methods in the context of early drug discovery.

Computational Approaches in Preclinical Studies on Drug Discovery and Development

Because undesirable pharmacokinetics and toxicity are significant reasons for the failure of drug development in the costly late stage, it has been widely recognized that drug ADMET properties should be considered as early as possible to reduce failure rates in the clinical phase of drug discovery. Concurrently, drug recalls have become increasingly common in recent years, prompting pharmaceutical companies to increase attention toward the safety evaluation of preclinical drugs. In vitro and in vivo drug evaluation techniques are currently more mature in preclinical applications, but these technologies are costly. In recent years, with the rapid development of computer science, in silico technology has been widely used to evaluate the relevant properties of drugs in the preclinical stage and has produced many software programs and in silico models, further promoting the study of ADMET in vitro. In this review, we first introduce the two ADMET prediction categories (molecular modeling and data modeling). Then, we perform a systematic classification and description of the databases and software commonly used for ADMET prediction. We focus on some widely studied ADMT properties as well as PBPK simulation, and we list some applications that are related to the prediction categories and web tools. Finally, we discuss challenges and limitations in the preclinical area and propose some suggestions and prospects for the future.

How Qualification of 3D Disease Models Cuts the Gordian Knot in Preclinical Drug Development

Preclinical research struggles with its predictive power for drug effects in patients. The clinical success of preclinically approved drug candidates ranges between 3% and 33%. Regardless of the approach, novel disease models and test methods need to prove their relevance and reliability for predicting drug effects in patients, which is usually achieved by method validation. Nevertheless, validating all models appears unrealistic due to the variety of diseases. Thus, novel concepts are needed to increase the quality of preclinical research.Herein, we introduce qualification as a minimal standard to establish the relevance of preclinical models and test methods. Qualification starts with prioritizing and translating scientific requirements into technical parameters by quality function deployment. Qualified models use authenticated cells, which resemble the corresponding cells in humans in morphology and drug target expression. Moreover, disease models differ from normal models in the expression of relevant biomarkers. As a result, qualified test methods can discriminate effects of treatment standards and the effects of weakly effective or ineffective substances. Observer-blind readout, adequate data documentation, dropout inclusion, and a priori power studies are as crucial as realistic dosage regimens for qualified approaches. Here, we showcase the implementation of qualification. Adjusting the level of model complexity and qualification to three defined phases of preclinical research assures the optimal level of certainty at each step.In conclusion, qualification strengthens the researchers' impact by defining basic requirements that novel approaches must fulfill while still allowing for scientific creativity. Qualification helps to improve the predictive power of preclinical research. Applied to human cell-based models, qualification reduces animal testing, since only effective drug candidates are subjected to final animal testing and subsequently to clinical trials.

Direct Comparison of Total Clearance Prediction: Computational Machine Learning Model versus Bottom-Up Approach Using In Vitro Assay

The in vitro-in vivo extrapolation (IVIVE) approach for predicting total plasma clearance (CL_tot) has been widely used to rank order compounds early in discovery. More recently, a computational machine learning approach utilizing physicochemical descriptors and fingerprints calculated from chemical structure information has emerged, enabling virtual predictions even earlier in discovery. Previously, this approach focused more on in vitro intrinsic clearance (CL_int) prediction. Herein, we directly compare these two approaches for predicting CL_tot in rats. A structurally diverse set of 1114 compounds with known in vivo CL_tot, in vitro CL_int, and plasma protein binding was used as the basis for this evaluation. The machine learning models were assessed by validation approaches using the time- and cluster-split training and test sets, and five-fold cross validation. Assessed by five-fold validation, the random forest regression (RF) and radial basis function (RBF) models demonstrated better prediction performance in eight attempted machine learning models. The CL_tot values predicted by the RF and RBF models were within two-fold of the observed values for 67.7 and 71.9% of cluster-split test set compounds, respectively, while the predictivity was worse in the time-split dataset. The predictivity of both models tended to be improved by incorporating in vitro parameters, unbound fraction in plasma (f_u,p), and CL_int. CL_tot prediction utilizing in vitro CL_int and the well-stirred model, correcting for the fraction unbound in blood, was substantially worse compared to machine learning approaches for the same cluster-split test set. The reason that CL_tot is underestimated by IVIVE is not fully explained by considering the calculated microsomal unbound fraction (cf_u,mic), extended clearance classification system (ECCS), and omitting high clearance compounds in excess of hepatic blood flow. The analysis suggests that in silico machine learning models may have the power to reduce reliance on or replace in vitro and in vivo studies for chemical structure optimization in early drug discovery.

How Sure Can We Be about ML Methods-Based Evaluation of Compound Activity: Incorporation of Information about Prediction Uncertainty Using Deep Learning Techniques

A great variety of computational approaches support drug design processes, helping in selection of new potentially active compounds, and optimization of their physicochemical and ADMET properties. Machine learning is a group of methods that are able to evaluate in relatively short time enormous amounts of data. However, the quality of machine-learning-based prediction depends on the data supplied for model training. In this study, we used deep neural networks for the task of compound activity prediction and developed dropout-based approaches for estimating prediction uncertainty. Several types of analyses were performed: the relationships between the prediction error, similarity to the training set, prediction uncertainty, number and standard deviation of activity values were examined. It was tested whether incorporation of information about prediction uncertainty influences compounds ranking based on predicted activity and prediction uncertainty was used to search for the potential errors in the ChEMBL database. The obtained outcome indicates that incorporation of information about uncertainty of compound activity prediction can be of great help during virtual screening experiments.

Virtual screening web servers: designing chemical probes and drug candidates in the cyberspace

The interplay between life sciences and advancing technology drives a continuous cycle of chemical data growth; these data are most often stored in open or partially open databases. In parallel, many different types of algorithms are being developed to manipulate these chemical objects and associated bioactivity data. Virtual screening methods are among the most popular computational approaches in pharmaceutical research. Today, user-friendly web-based tools are available to help scientists perform virtual screening experiments. This article provides an overview of internet resources enabling and supporting chemical biology and early drug discovery with a main emphasis on web servers dedicated to virtual ligand screening and small-molecule docking. This survey first introduces some key concepts and then presents recent and easily accessible virtual screening and related target-fishing tools as well as briefly discusses case studies enabled by some of these web services. Notwithstanding further improvements, already available web-based tools not only contribute to the design of bioactive molecules and assist drug repositioning but also help to generate new ideas and explore different hypotheses in a timely fashion while contributing to teaching in the field of drug development.

Challenges in Alzheimer's Disease Drug Discovery and Development: The Role of Modeling, Simulation, and Open Data

Alzheimer's disease (AD) is the leading cause of dementia worldwide. With 35 million people over 60 years of age with dementia, there is an urgent need to develop new treatments for AD. To streamline this process, it is imperative to apply insights and learnings from past failures to future drug development programs. In the present work, we focus on how modeling and simulation tools can leverage open data to address drug development challenges in AD.

Development of New Methods Needs Proper Evaluation-Benchmarking Sets for Machine Learning Experiments for Class A GPCRs

New computational approaches for virtual screening applications are constantly being developed. However, before a particular tool is used to search for new active compounds, its effectiveness in the type of task must be examined. In this study, we conducted a detailed analysis of various aspects of preparation of respective data sets for such an evaluation. We propose a protocol for fetching data from the ChEMBL database, examine various compound representations in terms of the possible bias resulting from the way they are generated, and define a new metric for comparing the structural similarity of compounds, which is in line with chemical intuition. The newly developed method is also used for the evaluation of various approaches for division of the data set into training and test set parts, which are also examined in detail in terms of being the source of possible results bias. Finally, machine learning methods are applied in cross-validation studies of data sets constructed within the paper, constituting benchmarks for the assessment of computational methods developed for virtual screening tasks. Additionally, analogous data sets for class A G protein-coupled receptors (100 targets with the highest number of records) were prepared. They are available at http://gmum.net/benchmarks/ , together with script enabling reproduction of all results available at https://github.com/lesniak43/ananas .

A drug-likeness toolbox facilitates ADMET study in drug discovery

Undesirable pharmacokinetic (PK) properties or unacceptable toxicity are the main causes of the failure of drug candidates at the clinical trial stage. Since the concept of drug-likeness was first proposed, it has become an important consideration in the selection of compounds with desirable bioavailability during the early phases of drug discovery. Over the past decade, online resources have effectively facilitated drug-likeness studies in an economical and time-efficient manner. Here, we provide a comprehensive summary and comparison of current accessible online resources, in terms of their key features, application fields, and performance for in silico drug-likeness studies. We hope that the assembled toolbox will provide useful guidance to facilitate future in silico drug-likeness research.

Optimizing drug development in oncology by clinical trial simulation: Why and how?

In therapeutic research, the safety and efficacy of pharmaceutical products are necessarily tested on humans via clinical trials after an extensive and expensive preclinical development period. Methodologies such as computer modeling and clinical trial simulation (CTS) might represent a valuable option to reduce animal and human assays. The relevance of these methods is well recognized in pharmacokinetics and pharmacodynamics from the preclinical phase to postmarketing. However, they are barely used and are poorly regarded for drug approval, despite Food and Drug Administration and European Medicines Agency recommendations. The generalization of CTS could be greatly facilitated by the availability of software for modeling biological systems, by clinical trial studies and hospital databases. Data sharing and data merging raise legal, policy and technical issues that will need to be addressed. Development of future molecules will have to use CTS for faster development and thus enable better patient management. Drug activity modeling coupled with disease modeling, optimal use of medical data and increased computing speed should allow this leap forward. The realization of CTS requires not only bioinformatics tools to allow interconnection and global integration of all clinical data but also a universal legal framework to protect the privacy of every patient. While recognizing that CTS can never replace 'real-life' trials, they should be implemented in future drug development schemes to provide quantitative support for decision-making. This in silico medicine opens the way to the P4 medicine: predictive, preventive, personalized and participatory.

Salvindolin elicits opioid system-mediated antinociceptive and antidepressant-like activities

BACKGROUND

Salvinorin A is known as a highly selective kappa opioid receptor agonist with antinociceptive but mostly pro-depressive effects.

AIMS

In this article, we present its new semisynthetic analog with preferential mu opioid affinity, and promising antinociceptive, as well as antidepressant-like activities.

METHODS

Competitive binding studies were performed for salvindolin with kappa opioid and mu opioid. The mouse model of nociception (acetic-acid-induced writhing, formalin, and hot plate tests), depression (forced swim and tail suspension tests), and the open field test, were used to evaluate antinociceptive, antidepressant-like, and locomotion effects, respectively, of salvindolin. We built a 3-D molecular model of the kappa opioid receptor, using a mu opioid X-ray crystal structure as a template, and docked salvindolin into the two proteins.

RESULTS/OUTCOMES

Salvindolin showed affinity towards kappa opioid and mu opioid receptors but with 100-fold mu opioid preference. Tests of salvindolin in mice revealed good oral bioavailability, antinociceptive, and antidepressive-like effects, without locomotor incoordination. Docking of salvindolin showed strong interactions with the mu opioid receptor which matched well with experimental binding data. Salvindolin-induced behavioral changes in the hot plate and forced swim tests were attenuated by naloxone (nonselective opioid receptor antagonist) and/or naloxonazine (selective mu opioid receptor antagonist) but not by nor-binaltorphimine (selective kappa opioid receptor antagonist). In addition, WAY100635 (a selective serotonin 1A receptor antagonist) blocked the antidepressant-like effect of salvindolin.

CONCLUSIONS/INTERPRETATION

By simple chemical modification, we were able to modulate the pharmacological profile of salvinorin A, a highly selective kappa opioid receptor agonist, to salvindolin, a ligand with preferential mu opioid receptor affinity and activity on the serotonin 1A receptor. With its significant antinociceptive and antidepressive-like activities, salvindolin has the potential to be an analgesic and/or antidepressant drug candidate.

Application of New Cellular and Microphysiological Systems to Drug Metabolism Optimization and Their Positioning Respective to In Silico Tools

New cellular model systems for drug metabolism applications, such as advanced 2D liver co-cultures, spheroids, and microphysiological systems (MPSs), offer exciting opportunities to reproduce human biology more closely in vitro with the aim of improving predictions of pharmacokinetics, drug-drug interactions, and efficacy. These advanced cellular systems have quickly become established for human intrinsic clearance determination and have been validated for several other absorption, distribution, metabolism, and excretion (ADME) applications. Adoption will be driven through the demonstration of clear added value, for instance, by more accurate and precise clearance predictions and by more reliable extrapolation of drug interaction potential leading to faster progression to pivotal proof-of-concept studies. New experimental systems are attractive when they can (1) increase experimental capacity, removing optimization bottlenecks; (2) improve measurement quality of ADME properties that impact pharmacokinetics; and (3) enable measurements to be made that were not previously possible, reducing risk in ADME prediction and candidate selection. As new systems become established, they will find their place in the repository of tools used at different stages of the research and development process, depending on the balance of value, throughput, and cost. In this article, we give a perspective on the integration of these new methodologies into ADME optimization during drug discovery, and the likely applications and impacts on drug development.

An Integrated Transfer Learning and Multitask Learning Approach for Pharmacokinetic Parameter Prediction

BACKGROUND

Pharmacokinetic evaluation is one of the key processes in drug discovery and development. However, current absorption, distribution, metabolism, and excretion prediction models still have limited accuracy.

AIM

This study aims to construct an integrated transfer learning and multitask learning approach for developing quantitative structure-activity relationship models to predict four human pharmacokinetic parameters.

METHODS

A pharmacokinetic data set included 1104 U.S. FDA approved small molecule drugs. The data set included four human pharmacokinetic parameter subsets (oral bioavailability, plasma protein binding rate, apparent volume of distribution at steady-state, and elimination half-life). The pretrained model was trained on over 30 million bioactivity data entries. An integrated transfer learning and multitask learning approach was established to enhance the model generalization.

RESULTS

The pharmacokinetic data set was split into three parts (60:20:20) for training, validation, and testing by the improved maximum dissimilarity algorithm with the representative initial set selection algorithm and the weighted distance function. The multitask learning techniques enhanced the model predictive ability. The integrated transfer learning and multitask learning model demonstrated the best accuracies, because deep neural networks have the general feature extraction ability; transfer learning and multitask learning improve the model generalization.

CONCLUSIONS

The integrated transfer learning and multitask learning approach with the improved data set splitting algorithm was first introduced to predict the pharmacokinetic parameters. This method can be further employed in drug discovery and development.

ADMETlab: a platform for systematic ADMET evaluation based on a comprehensively collected ADMET database

Current pharmaceutical research and development (R&D) is a high-risk investment which is usually faced with some unexpected even disastrous failures in different stages of drug discovery. One main reason for R&D failures is the efficacy and safety deficiencies which are related largely to absorption, distribution, metabolism and excretion (ADME) properties and various toxicities (T). Therefore, rapid ADMET evaluation is urgently needed to minimize failures in the drug discovery process. Here, we developed a web-based platform called ADMETlab for systematic ADMET evaluation of chemicals based on a comprehensively collected ADMET database consisting of 288,967 entries. Four function modules in the platform enable users to conveniently perform six types of drug-likeness analysis (five rules and one prediction model), 31 ADMET endpoints prediction (basic property: 3, absorption: 6, distribution: 3, metabolism: 10, elimination: 2, toxicity: 7), systematic evaluation and database/similarity searching. We believe that this web platform will hopefully facilitate the drug discovery process by enabling early drug-likeness evaluation, rapid ADMET virtual screening or filtering and prioritization of chemical structures. The ADMETlab web platform is designed based on the Django framework in Python, and is freely accessible at http://admet.scbdd.com/ .

Sachem: a chemical cartridge for high-performance substructure search

BACKGROUND

Structure search is one of the valuable capabilities of small-molecule databases. Fingerprint-based screening methods are usually employed to enhance the search performance by reducing the number of calls to the verification procedure. In substructure search, fingerprints are designed to capture important structural aspects of the molecule to aid the decision about whether the molecule contains a given substructure. Currently available cartridges typically provide acceptable search performance for processing user queries, but do not scale satisfactorily with dataset size.

RESULTS

We present Sachem, a new open-source chemical cartridge that implements two substructure search methods: The first is a performance-oriented reimplementation of substructure indexing based on the OrChem fingerprint, and the second is a novel method that employs newly designed fingerprints stored in inverted indices. We assessed the performance of both methods on small, medium, and large datasets containing 1, 10, and 94 million compounds, respectively. Comparison of Sachem with other freely available cartridges revealed improvements in overall performance, scaling potential and screen-out efficiency.

CONCLUSIONS

The Sachem cartridge allows efficient substructure searches in databases of all sizes. The sublinear performance scaling of the second method and the ability to efficiently query large amounts of pre-extracted information may together open the door to new applications for substructure searches.

Computational modeling of human oral bioavailability: what will be next?

INTRODUCTION

The oral route is the most convenient way of administrating drugs. Therefore, accurate determination of oral bioavailability is paramount during drug discovery and development. Quantitative structure-property relationship (QSPR), rule-of-thumb (RoT) and physiologically based-pharmacokinetic (PBPK) approaches are promising alternatives to the early oral bioavailability prediction. Areas covered: The authors give insight into the factors affecting bioavailability, the fundamental theoretical framework and the practical aspects of computational methods for predicting this property. They also give their perspectives on future computational models for estimating oral bioavailability. Expert opinion: Oral bioavailability is a multi-factorial pharmacokinetic property with its accurate prediction challenging. For RoT and QSPR modeling, the reliability of datasets, the significance of molecular descriptor families and the diversity of chemometric tools used are important factors that define model predictability and interpretability. Likewise, for PBPK modeling the integrity of the pharmacokinetic data, the number of input parameters, the complexity of statistical analysis and the software packages used are relevant factors in bioavailability prediction. Although these approaches have been utilized independently, the tendency to use hybrid QSPR-PBPK approaches together with the exploration of ensemble and deep-learning systems for QSPR modeling of oral bioavailability has opened new avenues for development promising tools for oral bioavailability prediction.

Convolutional Embedding of Attributed Molecular Graphs for Physical Property Prediction

The task of learning an expressive molecular representation is central to developing quantitative structure-activity and property relationships. Traditional approaches rely on group additivity rules, empirical measurements or parameters, or generation of thousands of descriptors. In this paper, we employ a convolutional neural network for this embedding task by treating molecules as undirected graphs with attributed nodes and edges. Simple atom and bond attributes are used to construct atom-specific feature vectors that take into account the local chemical environment using different neighborhood radii. By working directly with the full molecular graph, there is a greater opportunity for models to identify important features relevant to a prediction task. Unlike other graph-based approaches, our atom featurization preserves molecule-level spatial information that significantly enhances model performance. Our models learn to identify important features of atom clusters for the prediction of aqueous solubility, octanol solubility, melting point, and toxicity. Extensions and limitations of this strategy are discussed.