MouseTox: An online toxicity assessment tool for small molecules through Enalos Cloud platform

Development of a Robust Read-Across Model for the Prediction of Biological Potency of Novel Peroxisome Proliferator-Activated Receptor Delta Agonists

A robust predictive model was developed using 136 novel peroxisome proliferator-activated receptor delta (PPARδ) agonists, a distinct subtype of lipid-activated transcription factors of the nuclear receptor superfamily that regulate target genes by binding to characteristic sequences of DNA bases. The model employs various structural descriptors and docking calculations and provides predictions of the biological activity of PPARδ agonists, following the criteria of the Organization for Economic Co-operation and Development (OECD) for the development and validation of quantitative structure-activity relationship (QSAR) models. Specifically focused on small molecules, the model facilitates the identification of highly potent and selective PPARδ agonists and offers a read-across concept by providing the chemical neighbours of the compound under study. The model development process was conducted on Isalos Analytics Software (v. 0.1.17) which provides an intuitive environment for machine-learning applications. The final model was released as a user-friendly web tool and can be accessed through the Enalos Cloud platform's graphical user interface (GUI).

Gap-Δenergy, a New Metric of the Bond Energy State, Assisting to Predict Molecular Toxicity

Molecular toxicity is a critical feature of drug development. It is thus very important to develop computational models to evaluate the toxicity of small molecules. The accuracy of toxicity prediction largely depends on the quality of molecular representation; however, current methods for this purpose do not address this issue well. Here, we introduce a new metric, gap-Δenergy, which is designed to quantify the intermolecular bond energy difference with atom distance. We next find significant variations in the gap-Δenergy distribution among different types of molecules. Moreover, we show that this metric is able to distinguish the toxic small molecules. We collected data sets of toxic and exogenous small molecules and presented a novel index, namely, global toxicity, to evaluate the overall toxicity of molecules. Based on molecular descriptors and the proposed gap-Δenergy metric, we further constructed machine learning models that were trained with 7816 small molecules. The XGBoost-based model achieved the best performance with an AUC score of 0.965 and an F1 score of 0.849 on the test set (1954 small molecules), which outperformed the model that did not use gap-Δenergy features, with a sensitivity score increase of 3.2%.

In Silico and In Vitro Identification of 1,8-Dihydroxy-4,5-dinitroanthraquinone as a New Antibacterial Agent against Staphylococcus aureus and Enterococcus faecalis

Increasing rates of bacterial resistance to antibiotics are a growing concern worldwide. The search for potential new antibiotics has included several natural products such as anthraquinones. However, comparatively less attention has been given to anthraquinones that exhibit functional groups that are uncommon in nature. In this work, 114 anthraquinones were evaluated using in silico methods to identify inhibitors of the enzyme phosphopantetheine adenylyltransferase (PPAT) of Staphylococcus aureus, Enterococcus faecalis, and Escherichia coli. Virtual screenings based on molecular docking and the pharmacophore model, molecular dynamics simulations, and free energy calculations pointed to 1,8-dihydroxy-4,5-dinitroanthraquinone (DHDNA) as the most promising inhibitor. In addition, these analyses highlighted the contribution of the nitro group to the affinity of this anthraquinone for the nucleotide-binding site of PPAT. Furthermore, DHDNA was active in vitro towards Gram-positive bacteria with minimum inhibitory concentration (MIC) values of 31.25 µg/mL for S. aureus and 62.5 µg/mL for E. faecalis against both antibiotic-resistant isolates and reference strains but was ineffective against E. coli. Experiments on kill-time kinetics indicated that, at the tested concentrations, DHDNA produced bacteriostatic effects on both Gram-positive bacteria. Overall, our results present DHDNA as a potential PPAT inhibitor, showing antibacterial activity against antibiotic-resistant isolates of S. aureus and E. faecalis, findings that point to nitro groups as key to explaining these results.

In Silico and In Vitro Approach for Evaluation of the Anti-Inflammatory and Antioxidant Potential of Mygalin

There is a great interest in describing new molecules to be used as therapeutic targets in various diseases, particularly those that play a role in inflammatory responses and infection control. Mygalin is a synthetic analogue of spermidine, and previous studies have demonstrated its bactericidal effect against Escherichia coli, as well as its ability to modulate the inflammatory response of macrophages against lipopolysaccharide (LPS). However, the mechanisms through which mygalin regulates this inflammatory response remain poorly characterized. A set of platforms using molecular docking analysis was employed to analyze various properties of mygalin, including toxicity, biodistribution, absorption, and the prediction of its anti-inflammatory properties. In in vitro assays, we evaluated the potential of mygalin to interact with products of the inflammatory response, such as reactive oxygen species (ROS) and antioxidant activity, using the BMDM cell. The in silico analyses indicated that mygalin is not toxic, and can interact with proteins from the kinase group, and enzymes and receptors in eukaryotic cells. Molecular docking analysis showed interactions with key amino acid residues of COX-2, iNOS and 5-LOX enzymes. In vitro, assays demonstrated a significant reduction in the expression of iNOS and COX-2 induced by LPS, along with a decrease in the oxidative stress caused by the treatment with PMA, all without altering cell viability. Mygalin exhibited robust antioxidant activity in DPPH assays, regardless of the dose used, and inhibited heat-induced hemolysis. These studies suggest that mygalin holds promise for further investigation as a new molecule with anti-inflammatory and antioxidant properties.

Reducing health & environmental impacts of chemical warfare agents: Computational chemistry contributions

This review article summarizes advances in computational chemistry and cheminformatics methods and techniques that are used or have potential for use in reducing health and environmental impacts of Chemical Warfare Agents (CWA). These methods, include, but are not limited to, predictive modeling, data mining and virtual screening, similarity searching, molecular docking and dynamics and are briefly presented here. Applications of these in silico approaches, specifically for the protection of personnel and civilians against CWA, but also beyond, are discussed. CWA include toxic chemicals that can cause death, injury, or temporary incapacitation through their chemical action. CWA impose a significant worldwide threat and as such, destruction, remediation as well as protection measurements need to be carefully designed. Towards this goal computational chemistry and cheminformatics can play a key role specifically as far as decontamination, risk assessment and risk management are concerned. Among the wide range of in silico techniques applied for CWA, specific previously published paradigms are presented, including toxicity and property prediction, CWA simulant identification and CWA detoxification. Beyond CWA research, other applications with military interest are briefly presented and emerging trends of potential relevance noted.

Machine learning and graph neural network for finding potential drugs related to multiple myeloma

An innovative voting mechanism for virtual drug screening.

Predicting Serum Levels of Lithium-Treated Patients: A Supervised Machine Learning Approach

Routine monitoring of lithium levels is common clinical practice. This is because the lithium prediction strategies available developed by previous studies are still limited due to insufficient prediction performance. Thus, we used machine learning approaches to predict lithium concentration in a large real-world dataset. Real-world data from multicenter electronic medical records were used in different machine learning algorithms to predict: (1) whether the serum level was 0.6–1.2 mmol/L or 0.0–0.6 mmol/L (binary prediction), and (2) its concentration value (continuous prediction). We developed models from 1505 samples through 5-fold cross-validation and used 204 independent samples to test their performance by evaluating their accuracy. Moreover, we ranked the most important clinical features in different models and reconstructed three reduced models with fewer clinical features. For binary and continuous predictions, the average accuracy of these models was 0.70–0.73 and 0.68–0.75, respectively. Seven features were listed as important features related to serum lithium levels of 0.6–1.2 mmol/L or higher lithium concentration, namely older age, lower systolic blood pressure, higher daily and last doses of lithium prescription, concomitant psychotropic drugs with valproic acid and -pine drugs, and comorbid substance-related disorders. After reducing the features in the three new predictive models, the binary or continuous models still had an average accuracy of 0.67–0.74. Machine learning processes complex clinical data and provides a potential tool for predicting lithium concentration. This may help in clinical decision-making and reduce the frequency of serum level monitoring.

Discovery of Biphenyl-Sulfonamides as Novel β-N-Acetyl-d-Hexosaminidase Inhibitors via Structure-Based Virtual Screening

Novel insecticidal targets are always in demand due to the development of resistance. OfHex1, a β-N-acetyl-d-hexosaminidase identified in Ostrinia furnacalis (Asian corn borer), is involved in insect chitin catabolism and has proven an ideal target for insecticide development. In this study, structure-based virtual screening, structure simplification, and biological evaluation are used to show that compounds with a biphenyl-sulfonamide skeleton have great potential as OfHex1 inhibitors. Specifically, compounds 10k, 10u, and 10v have K_i values of 4.30, 3.72, and 4.56 μM, respectively, and thus, they are more potent than some reported nonglycosyl-based inhibitors such as phlegmacin B₁ (K_i = 26 μM), berberine (K_i = 12 μM), 2 (K_i = 11.2 μM), and 3 (K_i = 28.9 μM). Furthermore, inhibitory kinetic assessments reveal that the target compounds are competitive inhibitors with respect substrate, and based on toxicity predictions, most of them have potent drug properties. The obtained results indicate that the biphenyl-sulfonamide skeleton characterized by simple chemical structure, synthetic tractability, potent activity, and low toxicity has potential for further development in pest management targeting OfHex1.

Filling data gap for nicotinic acid, nicotinate esters and nicotinamide for the determination of permitted daily exposure by a category approach

This study aimed to obtain necessary toxicological data using experimental and computational methods for the calculation of a common permitted daily exposure (PDE) which can be relevant for nicotinic acid and its esters and nicotinamide according to European Medicines Agency Guideline on setting health-based exposure limits. PDE calculation is mainly based on critical toxicological endpoints. During this procedure, critical toxicological endpoints data of an active pharmaceutical ingredient (API) may not be able to find satisfactorily. Hence, using toxicological data for another API that has a similar chemical structure can be a useful way. In this study, toxicological endpoints of nicotinic acid and its esters and nicotinamide were evaluated. Then, the data gaps in the toxicological endpoints were filledusing the read-across approach. Based on the current existing data, nicotinic acid and its esters and also nicotinamide are not genotoxic and do not have skin sensitization potential. These compounds do not present a concern for carcinogenicity and developmental/reproductive toxicity. Based on these critical endpoints and available experimental data, the final PDE of 10 mg/day was calculated for all category members. Our study showed the utility of the read-across for PDE calculation of APIs with experimental toxicological data gap.

Virtual screening, synthesis, and bioactivity evaluation for the discovery of β-N-acetyl-D-hexosaminidase inhibitors

BACKGROUND

Molting is an essential insect developmental process, in which a variety of enzymes are involved. The inhibition of these enzymes effect normal insect growth and development and may even cause death. OfHex1, one of the β-N-acetyl-D-hexosaminidases, is a key enzyme involved in the molting process of the Asian corn borer (Ostrinia furnacalis), and is deemed a potential insecticidal target.

RESULTS

Based on the crystal structure of OfHex1, virtual screening was carried out to obtain a novel class of OfHex1 inhibitors, of which, 28 compounds were subjected to bioactivity evaluation. The compound 3, N-(3-cyano-4,5,6,7-tetrahydrobenzo[b]thiophen-2-yl)benzamide, showed good inhibition against OfHex1 with a K_i value of 11.2 μM. Structure optimization and molecular docking were applied for the structure-activity relationship analysis. The results also showed that the cyano group of this compound was essential for the maintenance of its inhibitory activity against OfHex1. Additionally, the interaction between this compound and Trp490, Glu328, Tyr475 and Trp524 were important for inhibitory activity.

CONCLUSION

The advantages of the derivatives of 2-amino-4,5,6,7-tetrahydrobenzo[b]thiophene-3-carbonitrile, which have simple chemical structures and are easily synthesized, suggests them to be developed further as potential OfHex1 inhibitors for pest control. © 2020 Society of Chemical Industry.

Zeta-Potential Read-Across Model Utilizing Nanodescriptors Extracted via the NanoXtract Image Analysis Tool Available on the Enalos Nanoinformatics Cloud Platform

Zeta potential is one of the most critical properties of nanomaterials (NMs) which provides an estimation of the surface charge, and therefore electrostatic stability in medium and, in practical terms, influences the NM's tendency to form agglomerates and to interact with cellular membranes. This paper describes a robust and accurate read-across model to predict NM zeta potential utilizing as the input data a set of image descriptors derived from transmission electron microscopy (TEM) images of the NMs. The image descriptors are calculated using NanoXtract (http://enaloscloud.novamechanics.com/EnalosWebApps/NanoXtract/), a unique online tool that generates 18 image descriptors from the TEM images, which can then be explored by modeling to identify those most predictive of NM behavior and biological effects. NM TEM images are used to develop a model for prediction of zeta potential based on grouping of the NMs according to their nearest neighbors. The model provides interesting insights regarding the most important similarity features between NMs-in addition to core composition the main elongation emerged, which links to key drivers of NM toxicity such as aspect ratio. Both the NanoXtract image analysis tool and the validated model for zeta potential (http://enaloscloud.novamechanics.com/EnalosWebApps/ZetaPotential/) are freely available online through the Enalos Nanoinformatics platform.

Transcriptomics in Toxicogenomics, Part III: Data Modelling for Risk Assessment

Transcriptomics data are relevant to address a number of challenges in Toxicogenomics (TGx). After careful planning of exposure conditions and data preprocessing, the TGx data can be used in predictive toxicology, where more advanced modelling techniques are applied. The large volume of molecular profiles produced by omics-based technologies allows the development and application of artificial intelligence (AI) methods in TGx. Indeed, the publicly available omics datasets are constantly increasing together with a plethora of different methods that are made available to facilitate their analysis, interpretation and the generation of accurate and stable predictive models. In this review, we present the state-of-the-art of data modelling applied to transcriptomics data in TGx. We show how the benchmark dose (BMD) analysis can be applied to TGx data. We review read across and adverse outcome pathways (AOP) modelling methodologies. We discuss how network-based approaches can be successfully employed to clarify the mechanism of action (MOA) or specific biomarkers of exposure. We also describe the main AI methodologies applied to TGx data to create predictive classification and regression models and we address current challenges. Finally, we present a short description of deep learning (DL) and data integration methodologies applied in these contexts. Modelling of TGx data represents a valuable tool for more accurate chemical safety assessment. This review is the third part of a three-article series on Transcriptomics in Toxicogenomics.

In silico prediction of toxicity and its applications for chemicals at work

OBJECTIVE AND METHODS

This study reviewed the concept of in silico prediction of chemical toxicity for prevention of occupational cancer and future prospects in workers' health. In this review, a new approach to determine the credibility of in silico predictions with raw data is explored, and the method of determining the confidence level of evaluation based on the credibility of data is discussed. I searched various papers and books related to the in silico prediction of chemical toxicity and carcinogenicity. The intention was to utilize the most recent reports after 2015 regarding in silico prediction.

RESULTS AND CONCLUSION

The application of in silico methods is increasing with the prediction of toxic risks to human and the environment. The various toxic effects of industrial chemicals have triggered the recognition of the importance of using a combination of in silico models in the risk assessments. In silico occupational exposure models, industrial accidents, and occupational cancers are effectively managed and chemicals evaluated. It is important to identify and manage hazardous substances proactively through the rigorous evaluation of chemicals.

Computational Prediction of a New ADMET Endpoint for Small Molecules: Anticommensal Effect on Human Gut Microbiota

The human gut microbiota (HGM), which are evolutionarily commensal in the human gastrointestinal system, are crucial to our health. However, HGM can be broadly shaped by multifaceted factors such as intake of drugs. About one-quarter of the existing drugs for humans, which are designed to target human cells rather than HGM, can notably alter the composition of HGM. Therefore, the anticommensal effect of human drugs should be avoided to the maximum extent possible in the drug discovery and development process. Nevertheless, the anticommensal effect of small molecules is a new ADMET (absorption, distribution, metabolism, excretion, and toxicity) end point, which was never predicted with the computational method before. In this work, we present the first machine-learning based consensus classification model with the accuracy (0.811 ± 0.012), precision (0.759 ± 0.032), specificity (0.901 ± 0.019), sensitivity (0.628 ± 0.036), F1-score (0.687 ± 0.023), and AUC (0.814 ± 0.030) respectively on the test set. Furthermore, we develop an easy-to-use "e-Commensal" program for the automatic prediction. Based on this program, virtual-screening of the food-constituent database (FooDB) indicates that 5888 of 23 202 food-relevant compounds are forecasted to possess an anticommensal effect on HGM. Several top-ranked anticommensal compounds in our prediction are further scrutinized and confirmed by experiments in the existing literature. To the best of our knowledge, this is the first classification model and stand-alone software for the prediction of commensal or anticommensal compounds impacting HGM.

A nanoinformatics decision support tool for the virtual screening of gold nanoparticle cellular association using protein corona fingerprints

The increasing use of nanoparticles (NPs) in a wide range of consumer and industrial applications has necessitated significant effort to address the challenge of characterizing and quantifying the underlying nanostructure - biological response relationships to ensure that these novel materials can be exploited responsibly and safely. Such efforts demand reliable experimental data not only in terms of the biological dose-response, but also regarding the physicochemical properties of the NPs and their interaction with the biological environment. The latter has not been extensively studied, as a large surface to bind biological macromolecules is a unique feature of NPs that is not relevant for chemicals or pharmaceuticals, and thus only limited data have been reported in the literature quantifying the protein corona formed when NPs interact with a biological medium and linking this with NP cellular association/uptake. In this work we report the development of a predictive model for the assessment of the biological response (cellular association, which can include both internalized NPs and those attached to the cell surface) of surface-modified gold NPs, based on their physicochemical properties and protein corona fingerprints, utilizing a dataset of 105 unique NPs. Cellular association was chosen as the end-point for the original experimental study due to its relevance to inflammatory responses, biodistribution, and toxicity in vivo. The validated predictive model is freely available online through the Enalos Cloud Platform ( http://enalos.insilicotox.com/NanoProteinCorona/ ) to be used as part of a regulatory or NP safe-by-design decision support system. This online tool will allow the virtual screening of NPs, based on a list of the significant NP descriptors, identifying those NPs that would warrant further toxicity testing on the basis of predicted NP cellular association.

e-Bitter: Bitterant Prediction by the Consensus Voting From the Machine-Learning Methods

In-silico bitterant prediction received the considerable attention due to the expensive and laborious experimental-screening of the bitterant. In this work, we collect the fully experimental dataset containing 707 bitterants and 592 non-bitterants, which is distinct from the fully or partially hypothetical non-bitterant dataset used in the previous works. Based on this experimental dataset, we harness the consensus votes from the multiple machine-learning methods (e.g., deep learning etc.) combined with the molecular fingerprint to build the bitter/bitterless classification models with five-fold cross-validation, which are further inspected by the Y-randomization test and applicability domain analysis. One of the best consensus models affords the accuracy, precision, specificity, sensitivity, F1-score, and Matthews correlation coefficient (MCC) of 0.929, 0.918, 0.898, 0.954, 0.936, and 0.856 respectively on our test set. For the automatic prediction of bitterant, a graphic program “e-Bitter” is developed for the convenience of users via the simple mouse click. To our best knowledge, it is for the first time to adopt the consensus model for the bitterant prediction and develop the first free stand-alone software for the experimental food scientist.

Enalos Suite: New Cheminformatics Platform for Drug Discovery and Computational Toxicology

In this chapter we present and discuss, with the aid of several representative case studies from drug discovery and computational toxicology, a new cheminformatics platform, Enalos Suite, that was developed with open source and freely available software. Enalos Suite ( http://enalossuite.novamechanics.com/ ) was designed and developed as a useful tool to address a variety of cheminformatics problems, given that it expedites tasks performed in predictive modeling and allows access, data mining and manipulation for multiple chemical databases (PubChem, UniChem, etc.). Enalos Suite was carefully designed to permit its extension and adjustment to the special field of interest of each user, including, for instance, nanoinformatics, biomedical, and other applications. To demonstrate the functionalities of Enalos Suite that are useful in different cheminformatics applications, we present indicative case studies that include the exploitation of chemical databases within a drug discovery project, the calculation of molecular descriptors, and finally the development of a predictive QSAR model validated according to OECD principles. We aspire that at the end of this chapter, the reader will capture the effectiveness of different functionalities included in the Enalos Suite that could be of significant value in a multitude of cheminformatics applications.