Applicability Domains for Classification Problems: Benchmarking of Distance to Models for Ames Mutagenicity Set

The present state and challenges of active learning in drug discovery

Active learning (AL) is an iterative feedback process that efficiently identifies valuable data within vast chemical space, even with limited labeled data. This characteristic renders it a valuable approach to tackle the ongoing challenges faced in drug discovery, such as the ever-expanding explore space and the limitations of labeled data. Consequently, AL is increasingly gaining prominence in the field of drug development. In this paper, we comprehensively review the application of AL at all stages of drug discovery, including compounds-target interaction prediction, virtual screening, molecular generation and optimization, as well as molecular properties prediction. Additionally, we discuss the challenges and prospects associated with the current applications of AL in drug discovery.

Multitargeted molecular docking and dynamics simulation studies of flavonoids and volatile components from the peel of Citrus sinensis L. (Osbeck) against specific tumor protein markers

Citrus sinensis (L.) Osbeck (Rutaceae), commonly known as the sweet orange, is a popular and widely consumed fruit with several medicinal properties. The present study aimed to perform the in silico screening of 18 flavonoids and eight volatile components from the peel of C. sinensis against apoptotic and inflammatory proteins, metalloprotease, and tumor suppressor markers. Flavonoids obtained higher probabilities than volatile components against selected anti-cancer drug targets. Hence, the data from the binding energies against the essential apoptotic and cell proliferation proteins substantiate that they may be promising compounds in developing effective candidates to block cell growth, proliferation, and induced cell death by activating the apoptotic pathway. Further, the binding stability of the selected targets and the corresponding molecules were analyzed by 100 ns molecular dynamics (MD) simulations. Chlorogenic acid has the most binding affinity against the important anti-cancer targets iNOS, MMP-9, and p53. The congruent binding mode to different drug targets focused on cancer shown by chlorogenic acid suggests that it may be a compound with significant therapeutic potential. Moreover, the binding energy predictions indicated that the compound had stable electrostatic and van der Waal energies. Thus, our data reinforce the medicinal importance of flavonoids from C. sinensis and expand the need for more studies, seeking to optimize results and amplify the impacts of further in vitro and in vivo studies. Communicated by Ramaswamy H. Sarma.

Degradation of carbofuran and acetamiprid in wolfberry by dielectric barrier discharge plasma: Kinetics, pathways, toxicity and molecular dynamics simulation

Carbofuran and acetamiprid pose the highest residual risk among pesticides found in wolfberries. This study aimed to degrade these pesticides in wolfberries using a multi-array dielectric barrier discharge plasma (DBD), evaluate the impact on safety and quality and explore their degradation mechanism. The results showed that DBD treatment achieved 90.6% and 80.9% degradation rates for carbofuran and acetamiprid, respectively, following a first-order kinetic reaction. The 120 s treatment successfully reduced pesticide contamination to levels below maximum residue limits. Treatment up to 180 s did not adversely affect the quality of wolfberries. QTOF/MS identification and degradation pathway analysis revealed that DBD broke down the furan ring and carbamate group of carbofuran, while replacing the chlorine atom and oxidizing the side chain of acetamiprid, leading to degradation. The toxicological evaluation showed that the degradation products were less toxic than undegraded pesticides. Molecular dynamics simulations revealed the reactive oxygen species (ROS) facilitated the degradation of pesticides through dehydrogenation and radical addition reactions. ROS type and dosage significantly affected the breakage of chemical bonds associated with toxicity (C4-O5 and C2-Cl1). These findings deepen insights into the plasma chemical degradation of pesticides.

In Silico prediction of inhibitors for multiple transporters via machine learning methods

Transporters play an indispensable role in facilitating the transport of nutrients, signaling molecules and the elimination of metabolites and toxins in human cells. Contemporary computational methods have been employed in the prediction of transporter inhibitors. However, these methods often focus on isolated endpoints, overlooking the interactions between transporters and lacking good interpretation. In this study, we integrated a comprehensive dataset and constructed models to assess the inhibitory effects on seven transporters. Both conventional machine learning and multi-task deep learning methods were employed. The results demonstrated that the MLT-GAT model achieved superior performance with an average AUC value of 0.882. It is noteworthy that our model excels not only in prediction performance but also in achieving robust interpretability, aided by GNN-Explainer. It provided valuable insights into transporter inhibition. The reliability of our model's predictions positioned it as a promising and valuable tool in the field of transporter inhibition research. Related data and code are available at https://gitee.com/wutiantian99/transporter_code.git.

TISBE: A Public Web Platform for the Consensus-Based Explainable Prediction of Developmental Toxicity

Despite being extremely relevant for the protection of prenatal and neonatal health, the developmental toxicity (Dev Tox) is a highly complex endpoint whose molecular rationale is still largely unknown. The lack of availability of high-quality data as well as robust nontesting methods makes its understanding even more difficult. Thus, the application of new explainable alternative methods is of utmost importance, with Dev Tox being one of the most animal-intensive research themes of regulatory toxicology. Descending from TIRESIA (Toxicology Intelligence and Regulatory Evaluations for Scientific and Industry Applications), the present work describes TISBE (TIRESIA Improved on Structure-Based Explainability), a new public web platform implementing four fundamental advancements for in silico analyses: a three times larger dataset, a transparent XAI (explainable artificial intelligence) framework employing a fragment-based fingerprint coding, a novel consensus classifier based on five independent machine learning models, and a new applicability domain (AD) method based on a double top-down approach for better estimating the prediction reliability. The training set (TS) includes as many as 1008 chemicals annotated with experimental toxicity values. Based on a 5-fold cross-validation, a median value of 0.410 for the Matthews correlation coefficient was calculated; TISBE was very effective, with a median value of sensitivity and specificity equal to 0.984 and 0.274, respectively. TISBE was applied on two external pools made of 1484 bioactive compounds and 85 pediatric drugs taken from ChEMBL (Chemical European Molecular Biology Laboratory) and TEDDY (Task-Force in Europe for Drug Development in the Young) repositories, respectively. Notably, TISBE gives users the option to clearly spot the molecular fragments responsible for the toxicity or the safety of a given chemical query and is available for free at https://prometheus.farmacia.uniba.it/tisbe.

Theoretical insight of reactive oxygen species scavenging mechanism in lignin waste depolymerization products

Apart from natural products and synthesis, phenolic compounds can be produced from the depolymerization of lignin, a major waste in biofuel and paper production. This process yields a plethora of aryl propanoid phenolic derivatives with broad biological activities, especially antioxidant properties. Due to its versatility, our study focuses on investigating the antioxidant mechanisms of several phenolic compounds obtained from renewable and abundant resources, namely, syringol (Hs), 4-allylsyringol (HAs), 4-propenylsyringol (HPns), and 4-propylsyringol (HPs). Employing the density functional theory (DFT) approach in conjunction with the QM-ORSA protocol, we aim to explore the reactivity of these compounds in neutralizing hydroperoxyl radicals in physiological and non-polar media. Kinetic and thermodynamic parameter calculations on the antioxidant activity of these compounds were also included in this study. Additionally, our research utilizes the activation strain model (ASM) for the first time to explain the reactivity of the HT and RAF mechanisms in the peroxyl radical scavenging process. It is predicted that HPs has the best rate constant in both media (1.13 × 108 M-1 s-1 and 1.75 × 108 M-1 s-1, respectively). Through ASM analysis, it is observed that the increase in the interaction energy due to the formation of intermolecular hydrogen bonds during the reaction is an important feature for accelerating the hydrogen transfer process. Furthermore, by examining the physicochemical and toxicity parameters, only Hs is not suitable for further investigation as a therapeutic agent because of potential toxicity and mutagenicity. However, overall, all compounds are considered potent HOO˙ scavengers in lipid-rich environments compared to previously studied antioxidants.

Artificial Intelligence in ADME Property Prediction

Absorption, distribution, metabolism, excretion (ADME) are key properties of a small molecule that govern pharmacokinetic profiles and impact its efficacy and safety. Computational methods such as machine learning and artificial intelligence have gained significant interest in both academic and industrial settings to predict pharmacokinetic properties of small molecules. These methods are applied in drug discovery to optimize chemical libraries, prioritize hits from biological screens, and optimize ADME properties of lead molecules. In the recent years, the drug discovery community witnessed the use of a range of neural network architectures such as deep neural networks, recurrent neural networks, graph neural networks, and transformer neural networks, which marked a paradigm shift in computer-aided drug design and development. This chapter discusses recent developments with an emphasis on their application to predict ADME properties.

Machine Learning Enables Accurate Prediction of Quinone Formation during Drug Metabolism

Metabolism helps in the elimination of drugs from the human body by making them more hydrophilic. Sometimes, drugs can be bioactivated to highly reactive metabolites or intermediates during metabolism. These reactive metabolites are often responsible for the toxicities associated with the drugs. Identification of reactive metabolites of drug candidates can be very helpful in the initial stages of drug discovery. Quinones are soft electrophiles that are generated as reactive intermediates during metabolism. Quinones make up more than 40% of the reactive metabolites. In this work, a reliable data set of 510 molecules was used to develop machine learning and deep learning-based predictive models to predict the formation of quinone-type metabolites. For representing molecules, two-dimensional (2D) descriptors, PubChem fingerprints, electro-topological state (E-state) fingerprints, and metabolic reactivity-based descriptors were used. Developed models were compared to the existing Xenosite web server using the untouched test set of 102 molecules. The best model achieved an accuracy of 86.27%, while the Xenosite server could achieve an accuracy of only 52.94% on the test set. Descriptor analysis revealed that the presence of greater numbers of polar moieties in a molecule can prevent the formation of quinone-type metabolites. In addition, the presence of a nitrogen atom in an aromatic ring and the presence of metabolophores V51, V52, and V53 (SMARTCyp descriptors) decrease the probability of quinone formation. Finally, a tool based on the best machine learning models was developed, which is accessible at http://14.139.57.41/quinonepred/.

In Silico Prediction of Human Organ Toxicity via Artificial Intelligence Methods

Unpredicted human organ level toxicity remains one of the major reasons for drug clinical failure. There is a critical need for cost-efficient strategies in the early stages of drug development for human toxicity assessment. At present, artificial intelligence methods are popularly regarded as a promising solution in chemical toxicology. Thus, we provided comprehensive in silico prediction models for eight significant human organ level toxicity end points using machine learning, deep learning, and transfer learning algorithms. In this work, our results showed that the graph-based deep learning approach was generally better than the conventional machine learning models, and good performances were observed for most of the human organ level toxicity end points in this study. In addition, we found that the transfer learning algorithm could improve model performance for skin sensitization end point using source domain of in vivo acute toxicity data and in vitro data of the Tox21 project. It can be concluded that our models can provide useful guidance for the rapid identification of the compounds with human organ level toxicity for drug discovery.

Prioritization of mycotoxins based on mutagenicity and carcinogenicity evaluation using combined in silico QSAR methods

Mycotoxins and their metabolites are a family of compounds that contains a great diversity of both structure and biological properties. Information on their toxicity is spread within several databases and in scientific literature. Due to the number of molecules and their structure diversity, the cost and time required for hazard evaluation of each compound is unrealistic. In that purpose, new approach methodologies (NAMs) can be applied to evaluate such large set of molecules. Among them, quantitative structure-activity relationship (QSAR) in silico models could be useful to predict the mutagenic and carcinogenic properties of mycotoxins. First, a complete list of 904 mycotoxins and metabolites was built. Then, some known mycotoxins were used to determine the best QSAR tools for mutagenicity and carcinogenicity predictions. The best tool was further applied to the whole list of 904 mycotoxins. At the end, 95 mycotoxins were identified as both mutagen and carcinogen and should be prioritized for further evaluation.

Combining machine learning and structure-based approaches to develop oncogene PIM kinase inhibitors

Introduction: PIM kinases are targets for therapeutic intervention since they are associated with a number of malignancies by boosting cell survival and proliferation. Over the past years, the rate of new PIM inhibitors discovery has increased significantly, however, new generation of potent molecules with the right pharmacologic profiles were in demand that can probably lead to the development of Pim kinase inhibitors that are effective against human cancer.

Method: In the current study, a machine learning and structure based approaches were used to generate novel and effective chemical therapeutics for PIM-1 kinase. Four different machine learning methods, namely, support vector machine, random forest, k-nearest neighbour and XGBoost have been used for the development of models. Total, 54 Descriptors have been selected using the Boruta method.

Results: SVM, Random Forest and XGBoost shows better performance as compared to k-NN. An ensemble approach was implemented and, finally, four potential molecules (CHEMBL303779, CHEMBL690270, MHC07198, and CHEMBL748285) were found to be effective for the modulation of PIM-1 activity. Molecular docking and molecular dynamic simulation corroborated the potentiality of the selected molecules. The molecular dynamics (MD) simulation study indicated the stability between protein and ligands.

Discussion: Our findings suggest that the selected models are robust and can be potentially useful for facilitating the discovery against PIM kinase.

Artificial neural network models driven novel virtual screening workflow for the identification and biological evaluation of BACE1 inhibitors

Beta-site amyloid-β precursor protein-cleaving enzyme 1 (BACE1) is a transmembrane aspartic protease and has shown potential as a possible therapeutic target for Alzheimer's disease. This aggravating disease involves the aberrant production of β amyloid plaques by BACE1 which catalyzes the rate-limiting step by cleaving the amyloid precursor protein (APP), generating the neurotoxic amyloid β protein that aggregates to form plaques leading to neurodegeneration. Therefore, it is indispensable to inhibit BACE1, thus modulating the APP processing. In this study, we present a workflow that utilizes a multi-stage virtual screening protocol for identifying potential BACE1 inhibitors by employing multiple artificial neural network-based models. Collectively, all the hyperparameter tuned models were assigned a task to virtually screen Maybridge library, thus yielding a consensus of 41 hits. The majority of these hits exhibited optimal pharmacokinetic properties confirmed by high central nervous system multiparameter optimization (CNS-MPO) scores. Further shortlisting of 8 compounds by molecular docking into the active site of BACE1 and their subsequent in-vitro evaluation identified 4 compounds as potent BACE1 inhibitors with IC50 values falling in the range 0.028-0.052 μM and can be further optimized with medicinal chemistry efforts to improve their activity.

Assessing chemical hazard and unraveling binding affinity of priority pollutants to lignin modifying enzymes for environmental remediation

Lignin-modifying enzymes (LMEs) are impactful biocatalysts in environmental remediation applications. However, LMEs-assisted experimental degradation neglects the molecular basis of pollutant degradation. Furthermore, throughout the remediation process, the inherent hazards of environmental pollutants remain untapped for in-depth toxicological endpoints. In this investigation, a predictive toxicological framework and a computational framework adopting LMEs were employed to assess the hazards of Priority Pollutants (PP) and its possible LMEs-assisted catalytic screening. The potential hazardous outcomes of PP were assessed using Quantitative structure-activity relationship (QSARs)-based techniques including Toxtree, ECOSAR, and T.E.S.T. tools. Toxicological findings revealed positive outcomes in a multitude of endpoints for all PP. The PP compound 2,3,7,8-TCDD (dioxin) was found to exhibit the lowest concentration of aquatic toxicity implementing aquatic model systems; LC₅₀ as 0.01, 0.01, 0.04 (mg L^-1) for Fish (96 H), Daphnid (48 H), Green algae (96 H) respectively. T.E.S.T. results revealed that chloroform, and 2-chlorophenol both seem to be developmental toxicants. Subsequently, LMEs-assisted docking procedure was employed in predictive mitigation of PP. The docking approach as predicted degradation revealed the far lowest docking energy score for Versatile peroxidase (VP)- 2,3,7,8-TCDD docked complex with a binding energy of -9.2 (kcal mol^-1), involved PHE-46, PRO-139, PRO-141, ILE-148, LEU-165, HIS-169, LEU-228, MET-262, and MET-265 as key interacting amino acid residues. Second most ranked but lesser than VP, Lignin peroxidase (LiP)- 2,3,7,8-TCDD docked complex exhibited a rather lower binding affinity score (-8.8 kcal mol^-1). Predictive degradation screening employing comparative docking revealed varying binding affinities, portraying that each LMEs member has independent feasibility to bind PP as substrate. Predictive findings endorsed the hazardous nature of associated PP in a multitude of endpoints, which could be attenuated by undertaking LMEs as a predictive approach to protect the environment and implement it in regulatory considerations.

Computational Studies Applied to Linalool and Citronellal Derivatives Against Alzheimer's and Parkinson's Disorders: A Review with Experimental Approach

Alzheimer's and Parkinson's are neurodegenerative disorders that affect a great number of people around the world, seriously compromising the quality of life of individuals, due to motor and cognitive damage. In these diseases, pharmacological treatment is used only to alleviate symptoms. This emphasizes the need to discover alternative molecules for use in prevention. Using Molecular Docking, this review aimed to evaluate the anti-Alzheimer's and anti-Parkinson's activity of linalool and citronellal, as well as their derivatives. Before performing Molecular Docking simulations, the compounds' pharmacokinetic characteristics were evaluated. For Molecular Docking, 7 chemical compounds derived from citronellal, and 10 compounds derived from linalool, and molecular targets involved in Alzheimer's and Parkinson's pathophysiology were selected. According to the Lipinski rules, the compounds under study presented good oral absorption and bioavailability. For toxicity, some tissue irritability was observed. For Parkinson-related targets, the citronellal and linalool derived compounds revealed excellent energetic affinity for α-Synuclein, Adenosine Receptors, Monoamine Oxidase (MAO), and Dopamine D1 receptor proteins. For Alzheimer disease targets, only linalool and its derivatives presented promise against BACE enzyme activity. The compounds studied presented high probability of modulatory activity against the disease targets under study, and are potential candidates for future drugs.

Pharmacophore Synergism in Diverse Scaffold Clinches in Aurora Kinase B

Aurora kinase B (AKB) is a crucial signaling kinase with an important role in cell division. Therefore, inhibition of AKB is an attractive approach to the treatment of cancer. In the present work, extensive quantitative structure-activity relationships (QSAR) analysis has been performed using a set of 561 structurally diverse aurora kinase B inhibitors. The Organization for Economic Cooperation and Development (OECD) guidelines were used to develop a QSAR model that has high statistical performance (R²_tr = 0.815, Q²_LMO = 0.808, R²_ex = 0.814, CCC_ex = 0.899). The seven-variable-based newly developed QSAR model has an excellent balance of external predictive ability (Predictive QSAR) and mechanistic interpretation (Mechanistic QSAR). The QSAR analysis successfully identifies not only the visible pharmacophoric features but also the hidden features. The analysis indicates that the lipophilic and polar groups-especially the H-bond capable groups-must be present at a specific distance from each other. Moreover, the ring nitrogen and ring carbon atoms play important roles in determining the inhibitory activity for AKB. The analysis effectively captures reported as well as unreported pharmacophoric features. The results of the present analysis are also supported by the reported crystal structures of inhibitors bound to AKB.

A universal similarity based approach for predictive uncertainty quantification in materials science

Immense effort has been exerted in the materials informatics community towards enhancing the accuracy of machine learning (ML) models; however, the uncertainty quantification (UQ) of state-of-the-art algorithms also demands further development. Most prominent UQ methods are model-specific or are related to the ensembles of models; therefore, there is a need to develop a universal technique that can be readily applied to a single model from a diverse set of ML algorithms. In this study, we suggest a new UQ measure known as the Δ-metric to address this issue. The presented quantitative criterion was inspired by the k-nearest neighbor approach adopted for applicability domain estimation in chemoinformatics. It surpasses several UQ methods in accurately ranking the predictive errors and could be considered a low-cost option for a more advanced deep ensemble strategy. We also evaluated the performance of the presented UQ measure on various classes of materials, ML algorithms, and types of input features, thus demonstrating its universality.

Identification of 2-(4-N,N-Dimethylaminophenyl)-5-methyl-1-phenethyl-1H-benzimidazole targeting HIV-1 CA capsid protein and inhibiting HIV-1 replication in cellulo

The capsid (CA) subunit of the HIV-1 Gag polyprotein is involved in several steps of the viral cycle, from the assembly of new viral particles to the protection of the viral genome until it enters into the nucleus of newly infected cells. As such, it represents an interesting therapeutic target to tackle HIV infection. In this study, we screened hundreds of compounds with a low cost of synthesis for their ability to interfere with Gag assembly in vitro. Representatives of the most promising families of compounds were then tested for their ability to inhibit HIV-1 replication in cellulo. From these molecules, a hit compound from the benzimidazole family with high metabolic stability and low toxicity, 2-(4-N,N-dimethylaminophenyl)-5-methyl-1-phenethyl-1H-benzimidazole (696), appeared to block HIV-1 replication with an IC50 of 3 µM. Quantitative PCR experiments demonstrated that 696 does not block HIV-1 infection before the end of reverse transcription, and molecular docking confirmed that 696 is likely to bind at the interface between two monomers of CA and interfere with capsid oligomerization. Altogether, 696 represents a promising lead molecule for the development of a new series of HIV-1 inhibitors.

Development of Predictive Classification Models for Whole Cell Antimycobacterial Activity of Benzothiazinones

Nitrobenzothiazinones (BTZs) are a very potent class of antibiotics against Mycobacterium tuberculosis. However, relationships between their structural properties and whole cell activity remain poorly predictable. Herein, we present the synthesis and antimycobacterial evaluation of a diverse set of BTZs. High potency was predominantly achieved by piperidine and piperazine substitutions, whereupon three compounds were identified as promising candidates, showing preferable metabolic stability. Lack of correlation between potency and calculated binding energies suggested that target inhibition is not the only requirement to obtain suitable antimycobacterial agents. In contrast, prediction of whole cell activity class was successfully accomplished by extensively validated machine learning models. The performance of the superior model was further verified by >70% correct class predictions for a large set of reported BTZs. Our generated model is thus a key prerequisite to streamline lead optimization endeavors, particularly regarding the improvement of overall hit rates in whole cell antimycobacterial assays.

Machine learning-based predictive models for identifying high active compounds against HIV-1 integrase

HIV-integrase is an important drug target because it catalyzes chromosomal integration of proviral DNA towards establishing latent infection. Computer-aided drug design has immensely contributed to identifying and developing novel antiviral drugs. We have developed various machine learning-based predictive models for identifying high activity compounds against HIV-integrase. Multiclass models were built using support vector machine with reasonable accuracy on the test and evaluation sets. The developed models were evaluated by rigorous validation approaches and the best features were selected by Boruta method. As compared to the model developed from all descriptors set, a slight improvement was observed among the selected descriptors. Validated models were further used for virtual screening of potential compounds from ChemBridge library. Of the six high active compounds predicted from selected models, compounds 9103124, 6642917 and 9082952 showed the most reasonable binding-affinity and stable-interaction with HIV-integrase active-site residues Asp64, Glu152 and Asn155. This was in agreement with previous reports on the essentiality of these residues against a wide range of inhibitors. We therefore highlight the rigorosity of validated classification models for accurate prediction and ranking of high active lead drugs against HIV-integrase.

In silico analytical toolset for predictive degradation and toxicity of hazardous pollutants in water sources

Different phenolic compounds, including multimeric lignin derivatives in the β-O-4 form, are among the most prevalent compounds in wastewater, often generated from paper industries. Relatively small concentrations of lignin are hazardous to aquatic organisms and can trigger severe environmental hazards. Herein, we present a predictive toolset to insight the induced toxic hazards prediction, and their Lignin peroxidase (LiP)-assisted degradation mechanism of selected multimeric lignin model compounds. T.E.ST and Toxtree toolset were deployed for toxic hazards estimation in different endpoints. To minimize the concerning hazards, we screened multimeric compounds for binding affinity with LiP. The binding affinity was found to be significantly lower than the reference compound. An Extra precision (XP) Glide score of -6.796 kcal/mol was found for dimer (guaiacyl 4-O-5 guaiacyl) complex as lowest compared to reference compound (-4.007 kcal/mol). The active site residues ASP-153, HIP-226, VAL-227, ARG-244, GLU-215, 239, PHE-261 were identified as site-specific key binding AA residues actively involved with corresponding ligands, forming Hydrophobic, H-Bond, π-Stacking, π-π type interactions. The DESMOND-assisted molecular dynamics simulation's (MDS) trajectories of protein-ligand revealed the considerable binding behavior and attained stability and system equilibrium state. Such theoretical and predictive conclusions indicted the feasibility of LiP assisted sustainable mitigation of lignin-based compounds, and such could be used to protect the environment from the potential hazards posed by recognized similar pollutants.

Assessing the External Exposome Using Wearable Passive Samplers and High-Resolution Mass Spectrometry among South African Children Participating in the VHEMBE Study

Children in low- and middle-income countries are often exposed to higher levels of chemicals and are more vulnerable to the health effects of air pollution. Little is known about the diversity, toxicity, and dynamics of airborne chemical exposures at the molecular level. We developed a workflow employing state-of-the-art wearable passive sampling technology coupled with high-resolution mass spectrometry to comprehensively measure 147 children's personal exposures to airborne chemicals in Limpopo, South Africa, as part of the Venda Health Examination of Mothers, Babies, and Their Environment (VHEMBE). 637 environmental exposures were detected, many of which have never been measured in this population; of these 50 airborne chemical exposures of concern were detected, including pesticides, plasticizers, organophosphates, dyes, combustion products, and perfumes. Biocides detected in wristbands included p,p'-dichlorodiphenyltrichloroethane (p,p'-DDT), p,p'-dichlorodiphenyldichloroethane (p,p'-DDD), p,p'-dichlorodiphenyldichloroethylene (p,p'-DDE), propoxur, piperonyl butoxide, and triclosan. Exposures differed across the assessment period with 27% of detected chemicals observed to be either higher or lower in the wet or dry seasons.

In Silico Prediction of Chemically Induced Mutagenicity: A Weight of Evidence Approach Integrating Information from QSAR Models and Read-Across Predictions

Information on genotoxicity is an essential piece of information in the framework of several regulations aimed at evaluating chemical toxicity. In this context, QSAR models that can predict Ames genotoxicity can conveniently provide relevant information. Indeed, they can be straightforwardly and rapidly used for predicting the presence or absence of genotoxic hazards associated with the interactions of chemicals with DNA. Nevertheless, and despite their ease of use, the main interpretative challenge is related to a critical assessment of the information that can be gathered, thanks to these tools. This chapter provides guidance on how to use freely available QSAR and read-across tools provided by VEGA HUB and on how to interpret their predictions according to a weight-of-evidence approach.

Active Learning for Drug Design: A Case Study on the Plasma Exposure of Orally Administered Drugs

The success of artificial intelligence (AI) models has been limited by the requirement of large amounts of high-quality training data, which is just the opposite of the situation in most drug discovery pipelines. Active learning (AL) is a subfield of AI that focuses on algorithms that select the data they need to improve their models. Here, we propose a two-phase AL pipeline and apply it to the prediction of drug oral plasma exposure. In phase I, the AL-based model demonstrated a remarkable capability to sample informative data from a noisy data set, which used only 30% of the training data to yield a prediction capability with an accuracy of 0.856 on an independent test set. In phase II, the AL-based model explored a large diverse chemical space (855K samples) for experimental testing and feedback. Improved accuracy and new highly confident predictions (50K samples) were observed, which suggest that the model's applicability domain has been significantly expanded.

Semi-automated workflow for molecular pair analysis and QSAR-assisted transformation space expansion

In the process of drug discovery, the optimization of lead compounds has always been a challenge faced by pharmaceutical chemists. Matched molecular pair analysis (MMPA), a promising tool to efficiently extract and summarize the relationship between structural transformation and property change, is suitable for local structural optimization tasks. Especially, the integration of MMPA with QSAR modeling can further strengthen the utility of MMPA in molecular optimization navigation. In this study, a new semi-automated procedure based on KNIME was developed to support MMPA on both large- and small-scale datasets, including molecular preparation, QSAR model construction, applicability domain evaluation, and MMP calculation and application. Two examples covering regression and classification tasks were provided to gain a better understanding of the importance of MMPA, which has also shown the reliability and utility of this MMPA-by-QSAR pipeline.

Development of Robust Quantitative Structure-Activity Relationship Models for CYP2C9, CYP2D6, and CYP3A4 Catalysis and Inhibition

Cytochrome P450 enzymes are responsible for the metabolism of >75% of marketed drugs, making it essential to identify the contributions of individual cytochromes P450 to the total clearance of a new candidate drug. Overreliance on one cytochrome P450 for clearance levies a high risk of drug-drug interactions; and considering that several human cytochrome P450 enzymes are polymorphic, it can also lead to highly variable pharmacokinetics in the clinic. Thus, it would be advantageous to understand the likelihood of new chemical entities to interact with the major cytochrome P450 enzymes at an early stage in the drug discovery process. Typical screening assays using human liver microsomes do not provide sufficient information to distinguish the specific cytochromes P450 responsible for clearance. In this regard, we experimentally assessed the metabolic stability of ∼5000 compounds for the three most prominent xenobiotic metabolizing human cytochromes P450, i.e., CYP2C9, CYP2D6, and CYP3A4, and used the data sets to develop quantitative structure-activity relationship models for the prediction of high-clearance substrates for these enzymes. Screening library included the NCATS Pharmaceutical Collection, comprising clinically approved low-molecular-weight compounds, and an annotated library consisting of drug-like compounds. To identify inhibitors, the library was screened against a luminescence-based cytochrome P450 inhibition assay; and through crossreferencing hits from the two assays, we were able to distinguish substrates and inhibitors of these enzymes. The best substrate and inhibitor models (balanced accuracies ∼0.7), as well as the data used to develop these models, have been made publicly available (https://opendata.ncats.nih.gov/adme) to advance drug discovery across all research groups. SIGNIFICANCE STATEMENT: In drug discovery and development, drug candidates with indiscriminate cytochrome P450 metabolic profiles are considered advantageous, since they provide less risk of potential issues with cytochrome P450 polymorphisms and drug-drug interactions. This study developed robust substrate and inhibitor quantitative structure-activity relationship models for the three major xenobiotic metabolizing cytochromes P450, i.e., CYP2C9, CYP2D6, and CYP3A4. The use of these models early in drug discovery will enable project teams to strategize or pivot when necessary, thereby accelerating drug discovery research.

ChemBioSim: Enhancing Conformal Prediction of In Vivo Toxicity by Use of Predicted Bioactivities

Computational methods such as machine learning approaches have a strong track record of success in predicting the outcomes of in vitro assays. In contrast, their ability to predict in vivo endpoints is more limited due to the high number of parameters and processes that may influence the outcome. Recent studies have shown that the combination of chemical and biological data can yield better models for in vivo endpoints. The ChemBioSim approach presented in this work aims to enhance the performance of conformal prediction models for in vivo endpoints by combining chemical information with (predicted) bioactivity assay outcomes. Three in vivo toxicological endpoints, capturing genotoxic (MNT), hepatic (DILI), and cardiological (DICC) issues, were selected for this study due to their high relevance for the registration and authorization of new compounds. Since the sparsity of available biological assay data is challenging for predictive modeling, predicted bioactivity descriptors were introduced instead. Thus, a machine learning model for each of the 373 collected biological assays was trained and applied on the compounds of the in vivo toxicity data sets. Besides the chemical descriptors (molecular fingerprints and physicochemical properties), these predicted bioactivities served as descriptors for the models of the three in vivo endpoints. For this study, a workflow based on a conformal prediction framework (a method for confidence estimation) built on random forest models was developed. Furthermore, the most relevant chemical and bioactivity descriptors for each in vivo endpoint were preselected with lasso models. The incorporation of bioactivity descriptors increased the mean F1 scores of the MNT model from 0.61 to 0.70 and for the DICC model from 0.72 to 0.82 while the mean efficiencies increased by roughly 0.10 for both endpoints. In contrast, for the DILI endpoint, no significant improvement in model performance was observed. Besides pure performance improvements, an analysis of the most important bioactivity features allowed detection of novel and less intuitive relationships between the predicted biological assay outcomes used as descriptors and the in vivo endpoints. This study presents how the prediction of in vivo toxicity endpoints can be improved by the incorporation of biological information-which is not necessarily captured by chemical descriptors-in an automated workflow without the need for adding experimental workload for the generation of bioactivity descriptors as predicted outcomes of bioactivity assays were utilized. All bioactivity CP models for deriving the predicted bioactivities, as well as the in vivo toxicity CP models, can be freely downloaded from https://doi.org/10.5281/zenodo.4761225.

Development of Machine Learning Models for Accurately Predicting and Ranking the Activity of Lead Molecules to Inhibit PRC2 Dependent Cancer

Disruption of epigenetic processes to eradicate tumor cells is among the most promising interventions for cancer control. EZH2 (Enhancer of zeste homolog 2), a catalytic component of polycomb repressive complex 2 (PRC2), methylates lysine 27 of histone H3 to promote transcriptional silencing and is an important drug target for controlling cancer via epigenetic processes. In the present study, we have developed various predictive models for modeling the inhibitory activity of EZH2. Binary and multiclass models were built using SVM, random forest and XGBoost methods. Rigorous validation approaches including predictiveness curve, Y-randomization and applicability domain (AD) were employed for evaluation of the developed models. Eighteen descriptors selected from Boruta methods have been used for modeling. For binary classification, random forest and XGBoost achieved an accuracy of 0.80 and 0.82, respectively, on external test set. Contrastingly, for multiclass models, random forest and XGBoost achieved an accuracy of 0.73 and 0.75, respectively. 500 Y-randomization runs demonstrate that the models were robust and the correlations were not by chance. Evaluation metrics from predictiveness curve show that the selected eighteen descriptors predict active compounds with total gain (TG) of 0.79 and 0.59 for XGBoost and random forest, respectively. Validated models were further used for virtual screening and molecular docking in search of potential hits. A total of 221 compounds were commonly predicted as active with above the set probability threshold and also under the AD of training set. Molecular docking revealed that three compounds have reasonable binding energy and favorable interactions with critical residues in the active site of EZH2. In conclusion, we highlighted the potential of rigorously validated models for accurately predicting and ranking the activities of lead molecules against cancer epigenetic targets. The models presented in this study represent the platform for development of EZH2 inhibitors.

Preclinical Studies in Anti-Trypanosomatidae Drug Development

The trypanosomatid parasites Trypanosoma brucei, Trypanosoma cruzi and Leishmania are the causative agents of human African trypanosomiasis, Chagas Disease and Leishmaniasis, respectively. These infections primarily affect poor, rural communities in the developing world, and are responsible for trapping sufferers and their families in a disease/poverty cycle. The development of new chemotherapies is a priority given that existing drug treatments are problematic. In our search for novel anti-trypanosomatid agents, we assess the growth-inhibitory properties of >450 compounds from in-house and/or "Pathogen Box" (PBox) libraries against L. infantum, L. amazonensis, L.braziliensis, T. cruzi and T. brucei and evaluate the toxicities of the most promising agents towards murine macrophages. Screens using the in-house series identified 17 structures with activity against and selective toward Leishmania: Compounds displayed 50% inhibitory concentrations between 0.09 and 25 μM and had selectivity index values >10. For the PBox library, ~20% of chemicals exhibited anti-parasitic properties including five structures whose activity against L. infantum had not been reported before. These five compounds displayed no toxicity towards murine macrophages over the range tested with three being active in an in vivo murine model of the cutaneous disease, with 100% survival of infected animals. Additionally, the oral combination of three of them in the in vivo Chagas disease murine model demonstrated full control of the parasitemia. Interestingly, phenotyping revealed that the reference strain responds differently to the five PBox-derived chemicals relative to parasites isolated from a dog. Together, our data identified one drug candidate that displays activity against Leishmania and other Trypanosomatidae in vitro and in vivo, while exhibiting low toxicity to cultured mammalian cells and low in vivo acute toxicity.

Prediction of an Organic Compound’s Biotransformation Time: A Study Using Avermectins

The current spread of the SARS-CoV-2 coronavirus is a challenge for the entire world. Ivermectin is a promising agent, which could be used to combat the SARS-CoV-2 coronavirus. It represents a complex of semisynthetic derivatives of natural avermectins that have been taken advantage of for a long time in medicine and agriculture as antiparasitic drugs. However, the experimental ecotoxicology assessment data for individual avermectins are still scarce. In relation to this, the aim of this study is to develop a mathematical model that would allow reliably predicting the biotransformation ability of natural and semisynthetic avermectins and identifying the structural fragments of avermectin molecules that have the largest impact on this biological activity. The base for the model construction was a structurally heterogeneous set including organic compounds with experimentally determined biotransformation half-life periods (KmHL). Using the OCHEM web platform (https://ochem.eu) with the implemented PyDescriptor plugin for the descriptor calculation and Random Forest and Transformer-CNN algorithms, a satisfactory (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$R_{{{\text{test}}}}^{2}$$\end{document} = 0.81) Quantitative Relationship Structure—Activity (QSAR) model was developed. The subsequent calculations have shown that natural avermectins undergo on average faster biotransformation in fish than the semisynthetic ones. In addition, structural fragments that increase and decrease the biotransformation rate are identified.

QSAR analysis of the acute toxicity of avermectins towards Tetrahymena pyriformis

Avermectins have been effectively used in medicine, veterinary medicine, and agriculture as antiparasitic agents for many years. However, there are still no reliable data on the main ecotoxicological characteristics of most individual avermectins. Although many QSAR models have been proposed to describe the acute toxicity of organic compounds towards Tetrahymena pyriformis (T. pyriformis), avermectins are outside the applicability domain of these models. The influence of the molecular structures of various organic compounds on the acute toxicity towards T. pyriformis was studied using the OCHEM web platform (https://ochem.eu). A data set of 1792 toxicants was used to create models. The QSAR (Quantitative Structure-Activity Relationship) models were developed using the molecular descriptors Dragon, ISIDA, CDK, PyDescriptor, alvaDesc, and SIRMS and machine learning methods, such as Least Squares Support Vector Machine and Transformer Convolutional Neural Network. The HYBOT descriptors and Random Forest were used for a comparative QSAR investigation. Since the best predictive ability was demonstrated by the Transformer Convolutional Neural Network model, it was used to predict the toxicity of individual avermectins towards T. pyriformis. During a structural interpretation of the developed QSAR model, we determined the significant molecular transformations that increase and decrease the acute toxicity of organic compounds.

Natural dye from Caesalpinia sappan L. heartwood for eco-friendly coloring of recycled paper based packing material and its in silico toxicity analysis

The uses of natural dyes are getting popularized due to the increased awareness regarding the toxicity of many chemical colorants. The chemical colorants are being replaced by the natural colorants for the various industrial applications. The plant-based natural colorants are considered eco-friendly and toxic free. In the present study, we report a natural dye from the heartwood of Caesalpinia sappan suitable for paper based packing materials. This forms the first report on the study of natural dye obtained from the heartwood of C. sappan on paper material. The extracted dye had a good photostability and able to make imprints on recycled paper bags. Moreover, a significant inhibition of bacterial growth was observed at a higher dye concentration of 100 μg mL^-1 against P. aeruginosa which was higher than the standard antibiotics. Growth inhibition was also observed in case of B. subtilis (22 ± 0.17 mm) and K. pneumonia (21 ± 0.53 mm) at 100 μg mL^-1. The dye could be used in making medicated packing materials and have many other bio-potential which was validated through in silico toxicity analysis. The application of such natural dyes in paper material value addition will help in a cleaner and sustainable process during paper recycling.

GGL-Tox: Geometric Graph Learning for Toxicity Prediction

Toxicity analysis is a major challenge in drug design and discovery. Recently significant progress has been made through machine learning due to its accuracy, efficiency, and lower cost. US Toxicology in the 21st Century (Tox21) screened a large library of compounds, including approximately 12 000 environmental chemicals and drugs, for different mechanisms responsible for eliciting toxic effects. The Tox21 Data Challenge offered a platform to evaluate different computational methods for toxicity predictions. Inspired by the success of multiscale weighted colored graph (MWCG) theory in protein-ligand binding affinity predictions, we consider MWCG theory for toxicity analysis. In the present work, we develop a geometric graph learning toxicity (GGL-Tox) model by integrating MWCG features and the gradient boosting decision tree (GBDT) algorithm. The benchmark tests of the Tox21 Data Challenge are employed to demonstrate the utility and usefulness of the proposed GGL-Tox model. An extensive comparison with other state-of-the-art models indicates that GGL-Tox is an accurate and efficient model for toxicity analysis and prediction.

Evolutionary algorithm-based generation of optimum peptide sequences with dengue virus inhibitory activity

Background: There is currently no effective dengue virus (DENV) therapeutic. We aim to develop a genetic algorithm-based framework for the design of peptides with possible DENV inhibitory activity. Methods & results: A Python-based tool (denominated AutoPepGEN) based on a DENV support vector machine classifier as the objective function was implemented. AutoPepGEN was applied to the design of three- to seven-amino acid sequences and ten peptides were selected. Peptide-protease (DENV) docking and Molecular Mechanics-Generalized Born Surface Area calculations were performed for the selected sequences and favorable binding energies were observed. Conclusion: It is hoped that AutoPepGEN will serve as an in silico alternative to the experimental design of positional scanning combinatorial libraries, known to be prone to a combinatorial explosion. AutoPepGEN is available at: https://github.com/sjbarigye/AutoPepGEN.

Epigenetic Target Fishing with Accurate Machine Learning Models

Epigenetic targets are of significant importance in drug discovery research, as demonstrated by the eight approved epigenetic drugs for treatment of cancer and the increasing availability of chemogenomic data related to epigenetics. This data represents many structure-activity relationships that have not been exploited thus far to develop predictive models to support medicinal chemistry efforts. Herein, we report the first large-scale study of 26 318 compounds with a quantitative measure of biological activity for 55 protein targets with epigenetic activity. We built predictive models with high accuracy for small molecules' epigenetic target profiling through a systematic comparison of the machine learning models trained on different molecular fingerprints. The models were thoroughly validated, showing mean precisions of up to 0.952 for the epigenetic target prediction task. Our results indicate that the models reported herein have considerable potential to identify small molecules with epigenetic activity. Therefore, our results were implemented as a freely accessible web application.

Head, Shoulders, Knees, and Toes: Placement of Wearable Passive Samplers Alters Exposure Profiles Observed

Chemical exposures are a major risk factor for many diseases. Comprehensive characterization of personal exposures is necessary to highlight chemicals of concern and factors that influence these chemical exposure dynamics. For this purpose, wearable passive samplers can be applied to assess longitudinal personal exposures to airborne contaminants. Questions remain regarding the impact of sampler placement at different locations of the body on the exposure profiles observed and how these placements affect the monitoring of seasonal dynamics in exposures. This study assessed personal air contaminant exposure using passive samplers worn in parallel across 32 participant's wrists, chest, and shoes over 24 h. Samplers were analyzed by thermal desorption gas chromatography high-resolution mass spectrometry. Personal exposure profiles were similar for about one-third of the 275 identified chemicals, irrespective of sampler placement. Signals of certain semivolatile organic compounds (SVOCs) were enhanced in shoes and, to a lesser extent, wrist samplers, as compared to those in chest samplers. Signals of volatile organic compounds were less impacted by sampler placement. Results showed that chest samplers predominantly captured more volatile exposures, as compared to those of particle-bound exposures, which may indicate predominant monitoring of chemicals via the inhalation route of exposure for chest samplers. In contrast, shoe samplers were more sensitive to particle-bound SVOCs. Seventy-one chemicals changed across participants between winter and summer in the same manner for two or more different sampler placements on the body, whereas 122 chemicals were observed to have seasonal differences in only one placement. Hence, the placement in certain cases significantly impacts exposure dynamics observed. This work shows that it is essential in epidemiological studies undertaking exposure assessment to consider the consequence of the placement of exposure monitors.

Pyrazol(in)e derivatives of curcumin analogs as a new class of anti-Trypanosoma cruzi agents

Aim: We report the synthesis and biological evaluation of a small library of 15 functionalized 3-styryl-2-pyrazolines and pyrazoles, derived from curcuminoids, as trypanosomicidal agents. Methods & results: The compounds were prepared via a cyclization reaction between the corresponding curcuminoids and the appropriate hydrazines. All of the derivatives synthesized were investigated for their trypanosomicidal activities. Compounds 4a and 4e showed significant activity against epimastigotes of Trypanosoma cruzi, with IC₅₀ values of 5.0 and 4.2 μM, respectively, accompanied by no toxicity to noncancerous mammalian cells. Compound 6b was found to effectively inhibit T. cruzi triosephosphate isomerase. Conclusion: The up to 16-fold higher potency of these derivatives compared with their curcuminoid precursors makes them a promising new family of T. cruzi inhibitors.

Large-Scale Modeling of Multispecies Acute Toxicity End Points Using Consensus of Multitask Deep Learning Methods

Computational methods to predict molecular properties regarding safety and toxicology represent alternative approaches to expedite drug development, screen environmental chemicals, and thus significantly reduce associated time and costs. There is a strong need and interest in the development of computational methods that yield reliable predictions of toxicity, and many approaches, including the recently introduced deep neural networks, have been leveraged towards this goal. Herein, we report on the collection, curation, and integration of data from the public data sets that were the source of the ChemIDplus database for systemic acute toxicity. These efforts generated the largest publicly available such data set comprising > 80,000 compounds measured against a total of 59 acute systemic toxicity end points. This data was used for developing multiple single- and multitask models utilizing random forest, deep neural networks, convolutional, and graph convolutional neural network approaches. For the first time, we also reported the consensus models based on different multitask approaches. To the best of our knowledge, prediction models for 36 of the 59 end points have never been published before. Furthermore, our results demonstrated a significantly better performance of the consensus model obtained from three multitask learning approaches that particularly predicted the 29 smaller tasks (less than 300 compounds) better than other models developed in the study. The curated data set and the developed models have been made publicly available at https://github.com/ncats/ld50-multitask, https://predictor.ncats.io/, and https://cactus.nci.nih.gov/download/acute-toxicity-db (data set only) to support regulatory and research applications.

SAR and QSAR research on tyrosinase inhibitors using machine learning methods

Tyrosinase is a key rate-limiting enzyme in the process of melanin synthesis, which is closely related to human pigmentation disorders. Tyrosinase inhibitors can down-regulate tyrosinase to effectively reduce melanin synthesis. In this work, we conducted structure-activity relationship (SAR) study on 1097 diverse mushroom tyrosinase inhibitors. We applied five kinds of machine learning methods to develop 15 classification models. Model 5B built by fully connected neural networks and ECFP4 fingerprints achieved the highest prediction accuracy of 91.36% and Matthews correlation coefficient (MCC) of 0.81 on the test set. The applicability domains (AD) of classification models were defined by d S T D - P R O method. Moreover, we clustered the 1097 inhibitors into eight subsets by K-Means to figure out inhibitors' structural features. In addition, 10 quantitative structure-activity relationship (QSAR) models were constructed by four machine learning methods based on 813 inhibitors. Model 6 J, the best QSAR model, was developed by fully connected neural networks with 50 RDKit descriptors. It resulted in a coefficient of determination (r ²) of 0.770 and a root mean squared error (RMSE) of 0.482 on the test set. The AD of Model 6 J was visualized by Williams plot. The models built in this study can be obtained from the authors.

Exploring the external exposome using wearable passive samplers - The China BAPE study

Environmental exposures are one of the greatest threats to human health, yet we lack tools to answer simple questions about our exposures: what are our personal exposure profiles and how do they change overtime (external exposome), how toxic are these chemicals, and what are the sources of these exposures? To capture variation in personal exposures to airborne chemicals in the gas and particulate phases and identify exposures which pose the greatest health risk, wearable exposure monitors can be deployed. In this study, we deployed passive air sampler wristbands with 84 healthy participants (aged 60-69 years) as part of the Biomarkers for Air Pollutants Exposure (China BAPE) study. Participants wore the wristband samplers for 3 days each month for five consecutive months. Passive samplers were analyzed using a novel gas chromatography high resolution mass spectrometry data-processing workflow to overcome the bottleneck of processing large datasets and improve confidence in the resulting identified features. The toxicity of chemicals observed frequently in personal exposures were predicted to identify exposures of potential concern via inhalation route or other routes of airborne contaminant exposure. Three exposures were highlighted based on elevated toxicity: dichlorvos from insecticides (mosquito/malaria control), naphthalene partly from mothballs, and 183 polyaromatic hydrocarbons from multiple sources. Other exposures explored in this study are linked to diet and personal care products, cigarette smoke, sunscreen, and antimicrobial soaps. We highlight the potential for this workflow employing wearable passive samplers for prioritizing chemicals of concern at both the community and individual level, and characterizing sources of exposures for follow up interventions.

QSAR-assisted-MMPA to expand chemical transformation space for lead optimization

Matched molecular pairs analysis (MMPA) has become a powerful tool for automatically and systematically identifying medicinal chemistry transformations from compound/property datasets. However, accurate determination of matched molecular pair (MMP) transformations largely depend on the size and quality of existing experimental data. Lack of high-quality experimental data heavily hampers the extraction of more effective medicinal chemistry knowledge. Here, we developed a new strategy called quantitative structure-activity relationship (QSAR)-assisted-MMPA to expand the number of chemical transformations and took the logD7.4 property endpoint as an example to demonstrate the reliability of the new method. A reliable logD7.4 consensus prediction model was firstly established, and its applicability domain was strictly assessed. By applying the reliable logD7.4 prediction model to screen two chemical databases, we obtained more high-quality logD7.4 data by defining a strict applicability domain threshold. Then, MMPA was performed on the predicted data and experimental data to derive more chemical rules. To validate the reliability of the chemical rules, we compared the magnitude and directionality of the property changes of the predicted rules with those of the measured rules. Then, we compared the novel chemical rules generated by our proposed approach with the published chemical rules, and found that the magnitude and directionality of the property changes were consistent, indicating that the proposed QSAR-assisted-MMPA approach has the potential to enrich the collection of rule types or even identify completely novel rules. Finally, we found that the number of the MMP rules derived from the experimental data could be amplified by the predicted data, which is helpful for us to analyze the medicinal chemical rules in local chemical environment. In summary, the proposed QSAR-assisted-MMPA approach could be regarded as a very promising strategy to expand the chemical transformation space for lead optimization, especially when no enough experimental data can support MMPA.

Structure-Activity Relationship Modeling and Experimental Validation of the Imidazolium and Pyridinium Based Ionic Liquids as Potential Antibacterials of MDR Acinetobacter Baumannii and Staphylococcus Aureus

Online Chemical Modeling Environment (OCHEM) was used for QSAR analysis of a set of ionic liquids (ILs) tested against multi-drug resistant (MDR) clinical isolate Acinetobacter baumannii and Staphylococcus aureus strains. The predictive accuracy of regression models has coefficient of determination q² = 0.66 - 0.79 with cross-validation and independent test sets. The models were used to screen a virtual chemical library of ILs, which was designed with targeted activity against MDR Acinetobacter baumannii and Staphylococcus aureus strains. Seven most promising ILs were selected, synthesized, and tested. Three ILs showed high activity against both these MDR clinical isolates.

Uncertainty quantification in drug design

Machine learning and artificial intelligence are increasingly being applied to the drug-design process as a result of the development of novel algorithms, growing access, the falling cost of computation and the development of novel technologies for generating chemically and biologically relevant data. There has been recent progress in fields such as molecular de novo generation, synthetic route prediction and, to some extent, property predictions. Despite this, most research in these fields has focused on improving the accuracy of the technologies, rather than on quantifying the uncertainty in the predictions. Uncertainty quantification will become a key component in autonomous decision making and will be crucial for integrating machine learning and chemistry automation to create an autonomous design-make-test-analyse cycle. This review covers the empirical, frequentist and Bayesian approaches to uncertainty quantification, and outlines how they can be used for drug design. We also outline the impact of uncertainty quantification on decision making.

Fingerprint-based computational models of 5-lipo-oxygenase activating protein inhibitors: Activity prediction and structure clustering

Inflammatory diseases can be treated by inhibiting 5-lipo-oxygenase activating protein (FLAP). In this study, a data set containing 2,112 FLAP inhibitors was collected. A total of 25 classification models were built by five machine learning algorithms with five different types of fingerprints. The best model, which was built by support vector machine algorithm with ECFP_4 fingerprint had an accuracy and a Matthews correlation coefficient of 0.862 and 0.722 on the test set, respectively. The predicted results were further evaluated by the application domain d_STD-PRO (a distance between one compound to models). Each compound had a d_STD-PRO value, which was calculated by the predicted probabilities obtained from all 25 models. The application domain results suggested that the reliability of predicted results depended mainly on the compounds themselves rather than algorithms or fingerprints. A group of customized 10-bit fingerprint was manually defined for clustering the molecular structures of 2,112 FLAP inhibitors into eight subsets by K-Means. According to the clustering results, most of inhibitors in two subsets (subsets 2 and 4) were highly active inhibitors. We found that aryl oxadiazole/oxazole alkanes, biaryl amino-heteroarenes, two aromatic rings (often N-containing) linked by a cyclobutene group, and 1,2,4-triazole group were typical fragments in highly active inhibitors.

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.

Comprehensive Analysis of Applicability Domains of QSPR Models for Chemical Reactions

Nowadays, the problem of the model’s applicability domain (AD) definition is an active research topic in chemoinformatics. Although many various AD definitions for the models predicting properties of molecules (Quantitative Structure-Activity/Property Relationship (QSAR/QSPR) models) were described in the literature, no one for chemical reactions (Quantitative Reaction-Property Relationships (QRPR)) has been reported to date. The point is that a chemical reaction is a much more complex object than an individual molecule, and its yield, thermodynamic and kinetic characteristics depend not only on the structures of reactants and products but also on experimental conditions. The QRPR models’ performance largely depends on the way that chemical transformation is encoded. In this study, various AD definition methods extensively used in QSAR/QSPR studies of individual molecules, as well as several novel approaches suggested in this work for reactions, were benchmarked on several reaction datasets. The ability to exclude wrong reaction types, increase coverage, improve the model performance and detect Y-outliers were tested. As a result, several “best” AD definitions for the QRPR models predicting reaction characteristics have been revealed and tested on a previously published external dataset with a clear AD definition problem.

Enhanced kinetic performance of peroxymonosulfate/ZVI system with the addition of copper ions: Reactivity, mechanism, and degradation pathways

Advanced oxidation processes (AOPs) based on the bimetallic system has been demonstrated as a promising way to enhance the degradation of pollutants in the water. In this study, the degradation of Rhodamine B (RhB) in a zero-valent iron (ZVI)/ peroxymonosulfate system with Cu²⁺ was thoroughly investigated. RhB could be efficiently removed (99.3 %) in the optimal ZVI/PMS/Cu²⁺ system, while only 58.2 % of RhB could be degraded in the ZVI/PMS system. The influence of reaction parameters on the degradation of RhB was further investigated. Quenching experiments and electron paramagnetic resonance (EPR) tests revealed that various reactive oxygen species could be generated in the ternary system, of which, ¹O₂ and O₂^- were identified for the first time. The effect of various anions, NOM and different water matrix were also considered at different concentrations. A variety of byproducts and degradation pathways were identified using HPLC/MS/MS. Finally, the Quantitative Structure Activity Relationship (QSAR) method of Toxicity Estimation Software Tool (TEST) was applied to estimate the toxicity of the byproducts and the results indicated that the overall toxicity of the target was relatively reduced. This study demonstrated the potential for the removal of environmental reluctant pollutants in water via the combined radical and non-radical pathways.

Quantitative weight of evidence method for combining predictions of quantitative structure-activity relationship models

A method for combining statistical-based QSAR predictions of two or more binary classification models is presented. It was assumed that all models were independent. This facilitated the combination of positive and negative predictions using a quantitative weight of evidence (qWoE) procedure based on Bayesian statistics and the additivity of the logarithms of the likelihood ratios. Previous studies combined more than one prediction but used arbitrary strengths for positive and negative predictions. In our approach, the combined models were validated by determining the sensitivity and specificity values, which are performance metrics that are a point of departure for obtaining values that measure the weight of evidence of positive and negative predictions. The developed method was experimentally applied in the prediction of Ames mutagenicity. The method achieved a similar accuracy to that of the experimental Ames test for this endpoint when the overall prediction was determined using a combination of the individual predictions of more than one model. Calculating the qWoE value would reduce the requirement for expert knowledge and decrease the subjectivity of the prediction. This method could be applied to other endpoints such as developmental toxicity and skin sensitisation with binary classification models.

Computational design of substrate selective inhibition

Most enzymes act on more than a single substrate. There is frequently a need to block the production of a single pathogenic outcome of enzymatic activity on a substrate but to avoid blocking others of its catalytic actions. Full blocking might cause severe side effects because some products of that catalysis may be vital. Substrate selectivity is required but not possible to achieve by blocking the catalytic residues of an enzyme. That is the basis of the need for "Substrate Selective Inhibitors" (SSI), and there are several molecules characterized as SSI. However, none have yet been designed or discovered by computational methods. We demonstrate a computational approach to the discovery of Substrate Selective Inhibitors for one enzyme, Prolyl Oligopeptidase (POP) (E.C 3.4.21.26), a serine protease which cleaves small peptides between Pro and other amino acids. Among those are Thyrotropin Releasing Hormone (TRH) and Angiotensin-III (Ang-III), differing in both their binding (K_m) and in turnover (k_cat). We used our in-house "Iterative Stochastic Elimination" (ISE) algorithm and the structure-based "Pharmacophore" approach to construct two models for identifying SSI of POP. A dataset of ~1.8 million commercially available molecules was initially reduced to less than 12,000 which were screened by these models to a final set of 20 molecules which were sent for experimental validation (five random molecules were tested for comparison). Two molecules out of these 20, one with a high score in the ISE model, the other successful in the pharmacophore model, were confirmed by in vitro measurements. One is a competitive inhibitor of Ang-III (increases its K_m), but non-competitive towards TRH (decreases its V_max).

Many proteins are enzymes—"catalytic machines" performing chemical reactions on "substrates"–which may be small or large molecules. Evolution optimized the speed of enzyme reactions, but mutations or excessive enzyme production could lead to non-controlled, accelerated activity, which must be blocked to avoid a product that promotes disease. Many inhibitors of enzymatic activity became drugs which can block the production of the aberrant product, due to blocking the enzymatic "machinery", the amino acids involved in catalysis. Most enzymes have several substrates and so, those other substrates are blocked too. Those may be vital to the well-being of cells and life and total inhibition is prone to cause serious side effects. It is therefore essential to solve the need for inhibition of a single substrate without inhibiting others. We have thus developed computational methods to block specifically the "culprit" substrate while allowing the enzyme machine to act on other substrates. By applying these computational methods, we predicted candidates for inhibiting one out of two substrates ("substrate selective inhibition") of a well-known enzyme reaction. In collaboration with a research group that excels in studying that specific enzyme (prolyl oligopeptidase) we found that two candidates out of a set of twenty that we picked out of 1.8 million molecules by filtering through computer models—are indeed selective to one substrate vis-a-vis the other (five random molecules were tested for comparison). This may be the first example of a computational method leading to substrate selective inhibitor drugs which could avoid side effects.

Transformer-CNN: Swiss knife for QSAR modeling and interpretation

We present SMILES-embeddings derived from the internal encoder state of a Transformer [1] model trained to canonize SMILES as a Seq2Seq problem. Using a CharNN [2] architecture upon the embeddings results in higher quality interpretable QSAR/QSPR models on diverse benchmark datasets including regression and classification tasks. The proposed Transformer-CNN method uses SMILES augmentation for training and inference, and thus the prognosis is based on an internal consensus. That both the augmentation and transfer learning are based on embeddings allows the method to provide good results for small datasets. We discuss the reasons for such effectiveness and draft future directions for the development of the method. The source code and the embeddings needed to train a QSAR model are available on https://github.com/bigchem/transformer-cnn. The repository also has a standalone program for QSAR prognosis which calculates individual atoms contributions, thus interpreting the model’s result. OCHEM [3] environment (https://ochem.eu) hosts the on-line implementation of the method proposed.