NERDD: a web portal providing access to in silico tools for drug discovery

Discovery of orally bioavailable phenyltetrazolium derivatives for the acute treatment and the secondary prevention of ischemic stroke

The potential for secondary stroke prevention, which can significantly reduce the risk of recurrent strokes by almost 90%, underscores its critical importance. N-butylphthalide (NBP) has emerged as a promising treatment for acute cerebral ischemia, yet its efficacy for secondary stroke prevention is hindered by inadequate pharmacokinetic properties. This study, driven by a comprehensive structural analysis, the iterative process of structure optimization culminated in the identification of compound B4, which demonstrated exceptional neuroprotective efficacy and remarkable oral exposure and oral bioavailability. Notably, in an in vivo transient middle cerebral artery occlusion (tMCAO) model, B4 substantially attenuated infarct volumes, surpassing the effectiveness of NBP. While oral treatment with B4 exhibited stronger prevention potency than NBP in photothrombotic (PT) model. In summary, compound B4, with its impressive oral bioavailability and potent neuroprotective effects, offers promise for both acute ischemic stroke treatment and secondary stroke prevention.

In vitro, in vivo metabolism and quantification of the novel synthetic opioid N-piperidinyl etonitazene (etonitazepipne)

OBJECTIVES

N-piperidinyl etonitazene (etonitazepipne) is a newly synthesized opioid related to the 2-benzylbenzimidazole analog class. Etonitazepipne has been formally notified and placed under intensive monitoring in Europe in January 2022. Nitazenes have high affinity at µ-opioid receptor (MOR). Etonitazepipne, specifically shows a EC50 of 2.49 nM, suggesting about 50 times higher potency combined with higher efficacy compared to morphine. Antinociceptive potency l ('hot plate test' with rats) was 192-fold greater than that of morphine.

METHODS

Here we report on a post-mortem case involving etonitazepipne and its quantification using a standard addition method (SAM) through liquid chromatography tandem mass spectrometry (LC-MS/MS). In addition, characterization and identification of phase I human metabolites using in vitro assay based on pooled human liver microsomes (pHLM) was performed along with the analysis of authentic urine samples by means of high-performance liquid chromatography high-resolution tandem mass spectrometry (LC-HRMS/MS).

RESULTS

The concentration of etonitazepipne in post-mortem blood and urine was 8.3 and 11 ng/mL, respectively. SAM was validated by assessing the following parameters: intraday and interday repeatability, matrix effect and recovery rate in post-mortem blood. A total of 20 and 14 metabolites were identified after pHLM incubation and urine analysis, respectively. Most pronounced in vitro and in vivo transformations were O-deethylation, hydroxylation, ketone reduction, and combinations thereof.

CONCLUSIONS

Considering small traces of the parent drug often found in real cases, the identification of metabolic biomarkers is crucial to identify exposure to this drug. O-deethylated, oxidated metabolites, and combination thereof are proposed as urinary biomarkers along with the parent compound.

Identification and quantification of acetyl tributyl citrate (ATBC) metabolites using human liver microsomes and human urine

Plasticizers are chemicals that make plastics flexible, and phthalates are commonly used. Due to the toxic effects of phthalates, there is increasing use of non-phthalate plasticizers like acetyl tributyl citrate (ATBC). ATBC has emerged as a safer alternative, yet concerns about its long-term safety persist due to its high leachability and potential endocrine-disrupting effects. This study aims to identify ATBC metabolites using human liver microsomes and suspect screening methods, and to explore potential urinary biomarkers for ATBC exposure. Using ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry, we identified ATBC metabolites, including acetyl dibutyl citrate (ADBC), tributyl citrate (TBC), and dibutyl citrate (DBC). Urine samples from 15 participants revealed the presence of ADBC in 5, TBC in 11, and DBC in all samples, with DBC concentrations pointedly higher than the other metabolites. These metabolites show promise as biomarkers for ATBC exposure, though further validation with human data is required. Our results underscore the need for comprehensive studies on ATBC metabolism, exposure pathways, and urinary excretion to accurately assess human exposure levels.

Polypharmacology prediction: the long road toward comprehensively anticipating small-molecule selectivity to de-risk drug discovery

INTRODUCTION

Small molecules often bind to multiple targets, a behavior termed polypharmacology. Anticipating polypharmacology is essential for drug discovery since unknown off-targets can modulate safety and efficacy - profoundly affecting drug discovery success. Unfortunately, experimental methods to assess selectivity present significant limitations and drugs still fail in the clinic due to unanticipated off-targets. Computational methods are a cost-effective, complementary approach to predict polypharmacology.

AREAS COVERED

This review aims to provide a comprehensive overview of the state of polypharmacology prediction and discuss its strengths and limitations, covering both classical cheminformatics methods and bioinformatic approaches. The authors review available data sources, paying close attention to their different coverage. The authors then discuss major algorithms grouped by the types of data that they exploit using selected examples.

EXPERT OPINION

Polypharmacology prediction has made impressive progress over the last decades and contributed to identify many off-targets. However, data incompleteness currently limits most approaches to comprehensively predict selectivity. Moreover, our limited agreement on model assessment challenges the identification of the best algorithms - which at present show modest performance in prospective real-world applications. Despite these limitations, the exponential increase of multidisciplinary Big Data and AI hold much potential to better polypharmacology prediction and de-risk drug discovery.

Visualization of renal rotenone accumulation after oral administration and in situ detection of kidney injury biomarkers via MALDI mass spectrometry imaging

The examination of drug accumulation within complex biological systems offers valuable insights into the molecular aspects of drug metabolism and toxicity. Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) is an innovative methodology that enables the spatial visualization and quantification of biomolecules as well as drug and its metabolites in complex biological system. Hence, this method provides valuable insights into the metabolic profile and any molecular changes that may occur as a result of drug treatment. The renal system is particularly vulnerable to adverse effects of drug-induced harm and toxicity. In this study, MALDI MSI was utilized to examine the spatial distribution of drug and renal metabolites within kidney tissues subsequent to a single oral dosage of the anticancer compound rotenone. The integration of ion mobility spectrometry with MALDI MSI enhanced the data acquisition and analysis, resulting to improved mass resolution. Subsequently, the MS/MS fragment ions of rotenone reference drug were detected and characterized using MALDI HDMS/MS imaging. Notably, drug accumulation was observed in the cortical region of the representative kidney tissue sections treated with rotenone. The histological examination of treated kidney tissues did not reveal any observable changes. Differential ion intensity of renal endogenous metabolites was observed between untreated and rotenone-treated tissues. In the context of treated kidney tissues, the ion intensity level of sphingomyelin (D18:1/16:0), a sphingolipid indicator of glomerular cell injury and renal damage, was found to be elevated significantly compared to untreated kidney tissues. Conversely, the ion intensities of choline, glycero-3-phosphocholine (GPC), inosine, and a lysophosphatidylcholine LysoPC(18:0) exhibited a significant decrease. The results of this study demonstrate the potential of MALDI MSI as a novel technique for investigating the in situ spatial distribution of drugs and renal endogenous molecules while preserving the anatomical integrity of the kidney tissue. This technique can be used to study drug-induced metabolism and toxicity in a dynamic manner.

Ibuprofen degradation by mixed bacterial consortia: Metabolic pathway and microbial community analysis

Degradation of ibuprofen, one of the most consumed drugs globally, by a mixed bacterial consortium was investigated. A contaminated hospital soil was used to enrich a bacterial consortium possessing the ability to degrade 4 mg/L ibuprofen in 6 days, fed on 6 mM acetate as a supplementary carbon source. Maximum ibuprofen degradation achieved was 99.51%, and for optimum ibuprofen degradation modelled statistically, the initial ibuprofen concentration, and temperature were determined to be 0.515 mg/L and 35 °C, respectively. The bacterial community analyses demonstrated an enrichment of Pseudomonas, Achromobacter, Bacillus, and Enterococcus in the presence of ibuprofen, suggesting their probable association with the biodegradation process. The biodegradation pathway developed using open-source metabolite predictors, GLORYx and BioTransformer suggested multiple degradation routes. Hydroxylation and oxidation were found to be the major mechanisms in ibuprofen degradation. Mono-hydroxylated metabolites were identified as well as predicted by the bioinformatics-based packages. Oxidation, dehydrogenation, super-hydroxylation, and hydrolysis were some other identified mechanisms.

3-CMC, 4-CMC, and 4-BMC Human Metabolic Profiling: New Major Pathways to Document Consumption of Methcathinone Analogues?

Synthetic cathinones represent one of the largest and most abused new psychoactive substance classes, and have been involved in numerous intoxications and fatalities worldwide. Methcathinone analogues like 3-methylmethcathinone (3-MMC), 3-chloromethcathinone (3-CMC), and 4-CMC currently constitute most of synthetic cathinone seizures in Europe. Documenting their consumption in clinical/forensic casework is therefore essential to tackle this trend. Targeting metabolite markers is a go-to to document consumption in analytical toxicology, and metabolite profiling is crucial to support investigations. We sought to identify 3-CMC, 4-CMC, and 4-bromomethcathinone (4-BMC) human metabolites. The substances were incubated with human hepatocytes; incubates were screened by liquid chromatography-high-resolution tandem mass spectrometry and data were mined with Compound Discoverer (Themo Scientific). 3-CMC-positive blood, urine, and oral fluid and 4-CMC-positive urine and saliva from clinical/forensic casework were analyzed. Analyses were supported by metabolite predictions with GLORYx freeware. Twelve, ten, and ten metabolites were identified for 3-CMC, 4-CMC, and 4-BMC, respectively, with similar transformations occurring for the three cathinones. Major reactions included ketoreduction and N-demethylation. Surprisingly, predominant metabolites were produced by combination of N-demethylation and ω-carboxylation (main metabolite in 3-CMC-positive urine), and combination of β-ketoreduction, oxidative deamination, and O-glucuronidation (main metabolite in 4-CMC-positive urine). These latter metabolites were detected in negative-ionization mode only and their non-conjugated form was not detected after glucuronide hydrolysis; this metabolic pathway was never reported for any methcathinone analogue susceptible to undergo the same transformations. These results support the need for comprehensive screening strategies in metabolite identification studies, to avoid overlooking significant metabolites and major markers of consumption.

In vitro metabolism of the emerging contaminant 6PPD-quinone in human and rat liver microsomes: Kinetics, pathways, and mechanism

N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine-quinone (6PPD-Q) is an ozonation product of the rubber antioxidant N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine (6PPD). 6PPD-Q has recently been detected in various environmental media, which may enter the human body via inhalation and skin contact pathways. However, the human metabolism of 6PPD-Q has remained unknown. This study investigated the in vitro Cytochrome P450-mediated metabolism of 6PPD-Q in human and rat liver microsomes (HLMs and RLMs). 6PPD-Q was significantly metabolized at lower concentrations but slowed at high concentrations. The intrinsic clearance (CLint) of 6PPD-Q was 21.10 and 18.58 μL min-1 mg-1 protein of HLMs and RLMs, respectively, suggesting low metabolic ability compared with other reported pollutants. Seven metabolites and one intermediate were identified, and metabolites were predicted immunotoxic or mutagenic toxicity. Mono- and di-oxygenation reactions were the main phase I in vitro metabolic pathways. Enzyme inhibition experiments and molecular docking techniques were further used to reveal the metabolic mechanism. CYP1A2, 3A4, and 2C19, especially CYP1A2, play critical roles in 6PPD-Q metabolism in HLMs, whereas 6PPD-Q is extensively metabolized in RLMs. Our study is the first to demonstrate the in vitro metabolic profile of 6PPD-Q in HLMs and RLMs. The results will significantly contribute to future human health management targeting the emerging pollutant 6PPD-Q.

Novel xanthone derivatives as potent sirtuin 2 inhibitors

Six amino derivatives of xanthone were obtained via chemical synthesis. Biochemical studies revealed their SIRT2 inhibitory activity ranging from 48.5 % (compound 4, 5-chloro-2-((4-(3-methoxyphenyl)piperazin-1-yl)methyl)-9H-xanthen-9-one hydrochloride) to 93.2 % (compound 3, 5-chloro-2-(((2-methoxyphenethyl)amino)methyl)-9H-xanthen-9-one hydrochloride). The structure-activity analysis showed favourable properties of secondary amines relative to tertiary piperazine derivatives. The tested compounds do not possess additional SIRT1 activating activity and no antioxidant activity (DPPH in vitro assay). Comprehensive analysis of the lipophilicity of the obtained compounds was also performed. For compound 3 potential molecular targets and similar active compounds were predicted in order to facilitate further research in this group of compounds.

In silico and in vitro human metabolism of IOX2, a performance-enhancing doping agent

IOX2 is a potent inhibitor of prolyl hydroxylase 2, a key enzyme in the regulation of hypoxia-inducible factor (HIF) and oxygen homeostasis. As such, it can be used to enhance athletic performance and is currently banned by the World Anti-Doping Agency (WADA). Detection of metabolites is critical to demonstrate drug use in doping. However, there is currently little data on IOX2 human metabolism. Our aim was to identify relevant biomarkers of IOX2 use in humans. For this purpose, IOX2 was incubated with 10-donor-pooled human hepatocytes for 3 h, incubates were analyzed by liquid chromatography-high-resolution tandem mass spectrometry (LC-HRMS/MS), and LC-HRMS/MS data were screened with Compound Discoverer (Thermo Scientific) for a comprehensive identification of IOX2 metabolites. Additionally, IOX2 human metabolites were predicted with GLORYx open-access software (University of Hamburg, Germany) to assist in the LC-HRMS/MS analysis and data mining. Thirteen metabolites were identified, oxidation at the quinolinyl group, O-glucuronidation, and combinations being predominant biotransformations. The results were consistent with previous animal studies and a single case of oral microdose administration. We suggest hydroxyquinolinyl-IOX2 as major biomarker of IOX2 use in biological samples, glucuronide hydrolysis being critical to increase IOX2 and hydroxyquinolinyl-IOX2 detectability in urine.

Design, synthesis, and lead optimization of piperazinyl-pyrimidine analogues as potent small molecules targeting the viral capping machinery of Chikungunya virus

The worldwide re-emerge of the Chikungunya virus (CHIKV), the high morbidity associated with it, and the lack of an available vaccine or antiviral treatment make the development of a potent CHIKV-inhibitor highly desirable. Therefore, an extensive lead optimization was performed based on the previously reported CHVB compound 1b and the reported synthesis route was optimized - improving the overall yield in remarkably shorter synthesis and work-up time. Hundred analogues were designed, synthesized, and investigated for their antiviral activity, physiochemistry, and toxicological profile. An extensive structure-activity relationship study (SAR) was performed, which focused mainly on the combination of scaffold changes and revealed the key chemical features for potent anti-CHIKV inhibition. Further, a thorough ADMET investigation of the compounds was carried out: the compounds were screened for their aqueous solubility, lipophilicity, their toxicity in CaCo-2 cells, and possible hERG channel interactions. Additionally, 55 analogues were assessed for their metabolic stability in human liver microsomes (HLMs), leading to a structure-metabolism relationship study (SMR). The compounds showed an excellent safety profile, favourable physicochemical characteristics, and the required metabolic stability. A cross-resistance study confirmed the viral capping machinery (nsP1) to be the viral target of these compounds. This study identified 31b and 34 as potent, safe, and stable lead compounds for further development as selective CHIKV inhibitors. Finally, the collected insight led to a successful scaffold hop (64b) for future antiviral research studies.

Unraveling biomarkers of exposure for tenuazonic acid through urinary metabolomics

Mycotoxins are secondary metabolites produced by fungi such as Aspergillus, Alternaria, and Penicillium, affecting nearly 80% of global food crops. Tenuazonic acid (TeA) is the major mycotoxin produced by Alternaria alternata, a prevalent pathogen affecting plants, fruits, and vegetables. TeA is notably prevalent in European diets, however, TeA biomarkers of exposure and metabolites remain unknown. This research aims to bridge this knowledge-gap by gaining insights about human TeA exposure and metabolization. Nine subjects were divided into two groups. The first group received a single bolus of TeA at the Threshold of Toxicological Concern (TTC) to investigate the presence of TeA urinary biomarkers, while the second group served as a control. Sixty-nine urinary samples were prepared and analyzed using UPLC-Xevo TQ-XS for TeA quantification and UPLC-Orbitrap Exploris for polar metabolome acquisition. TeA was rapidly excreted during the first 13 h and the fraction extracted was 0.39 ± 0.22. The polar metabolome compounds effectively discriminating the two groups were filtered using Orthogonal Partial Least Squares-Discriminant Analysis and subsequently annotated (n = 122) at confidence level 4. Finally, the urinary metabolome was compared to in silico predicted TeA metabolites. Nine metabolites, including oxidized, N-alkylated, desaturated, glucuronidated, and sulfonated forms of TeA were detected.

A practical guide to machine-learning scoring for structure-based virtual screening

Structure-based virtual screening (SBVS) via docking has been used to discover active molecules for a range of therapeutic targets. Chemical and protein data sets that contain integrated bioactivity information have increased both in number and in size. Artificial intelligence and, more concretely, its machine-learning (ML) branch, including deep learning, have effectively exploited these data sets to build scoring functions (SFs) for SBVS against targets with an atomic-resolution 3D model (e.g., generated by X-ray crystallography or predicted by AlphaFold2). Often outperforming their generic and non-ML counterparts, target-specific ML-based SFs represent the state of the art for SBVS. Here, we present a comprehensive and user-friendly protocol to build and rigorously evaluate these new SFs for SBVS. This protocol is organized into four sections: (i) using a public benchmark of a given target to evaluate an existing generic SF; (ii) preparing experimental data for a target from public repositories; (iii) partitioning data into a training set and a test set for subsequent target-specific ML modeling; and (iv) generating and evaluating target-specific ML SFs by using the prepared training-test partitions. All necessary code and input/output data related to three example targets (acetylcholinesterase, HMG-CoA reductase, and peroxisome proliferator-activated receptor-α) are available at https://github.com/vktrannguyen/MLSF-protocol , can be run by using a single computer within 1 week and make use of easily accessible software/programs (e.g., Smina, CNN-Score, RF-Score-VS and DeepCoy) and web resources. Our aim is to provide practical guidance on how to augment training data to enhance SBVS performance, how to identify the most suitable supervised learning algorithm for a data set, and how to build an SF with the highest likelihood of discovering target-active molecules within a given compound library.

Using in silico tools to predict flame retardant metabolites for more informative exposomics-based approaches

Introduction: The positive identification of xenobiotics and their metabolites in human biosamples is an integral aspect of exposomics research, yet challenges in compound annotation and identification continue to limit the feasibility of comprehensive identification of total chemical exposure. Nonetheless, the adoption of in silico tools such as metabolite prediction software, QSAR-ready structural conversion workflows, and molecular standards databases can aid in identifying novel compounds in untargeted mass spectral investigations, permitting the assessment of a more expansive pool of compounds for human health hazard. This strategy is particularly applicable when it comes to flame retardant chemicals. The population is ubiquitously exposed to flame retardants, and evidence implicates some of these compounds as developmental neurotoxicants, endocrine disruptors, reproductive toxicants, immunotoxicants, and carcinogens. However, many flame retardants are poorly characterized, have not been linked to a definitive mode of toxic action, and are known to share metabolic breakdown products which may themselves harbor toxicity. As U.S. regulatory bodies begin to pursue a subclass- based risk assessment of organohalogen flame retardants, little consideration has been paid to the role of potentially toxic metabolites, or to expanding the identification of parent flame retardants and their metabolic breakdown products in human biosamples to better inform the human health hazards imposed by these compounds. Methods: The purpose of this study is to utilize publicly available in silico tools to 1) characterize the structural and metabolic fates of proposed flame retardant classes, 2) predict first pass metabolites, 3) ascertain whether metabolic products segregate among parent flame retardant classification patterns, and 4) assess the existing coverage in of these compounds in mass spectral database. Results: We found that flame retardant classes as currently defined by the National Academies of Science, Engineering and Medicine (NASEM) are structurally diverse, with highly variable predicted pharmacokinetic properties and metabolic fates among member compounds. The vast majority of flame retardants (96%) and their predicted metabolites (99%) are not present in spectral databases, posing a challenge for identifying these compounds in human biosamples. However, we also demonstrate the utility of publicly available in silico methods in generating a fit for purpose synthetic spectral library for flame retardants and their metabolites that have yet to be identified in human biosamples. Discussion: In conclusion, exposomics studies making use of fit-for-purpose synthetic spectral databases will better resolve internal exposure and windows of vulnerability associated with complex exposures to flame retardant chemicals and perturbed neurodevelopmental, reproductive, and other associated apical human health impacts.

Metabolism Study of Anamorelin, a GHSR1a Receptor Agonist Potentially Misused in Sport, with Human Hepatocytes and LC-HRMS/MS

Anamorelin, developed for the treatment of cancer cachexia, is an orally active medication that improves appetite and food intake, thereby increasing body mass and physical functioning. It is classified as a growth hormone secretagogue and strictly monitored by the World Anti-Doping Agency (WADA), owing to its anabolic enhancing potential. Identifying anamorelin and/or metabolite biomarkers of consumption is critical in doping controls. However, there are currently no data available on anamorelin human metabolic fate. The aim of this study was to investigate and identify biomarkers characteristic of anamorelin intake using in silico metabolite predictions with GLORYx, in vitro incubation with 10-donor-pooled human hepatocytes, liquid chromatography-high-resolution tandem mass spectrometry (LC-HRMS/MS) analysis, and data processing with Thermo Scientific's Compound Discoverer. In silico prediction resulted in N-acetylation at the methylalanyl group as the main transformation (score, 88%). Others including hydroxylation at the indole substructure, and oxidation and N-demethylation at the trimethylhydrazino group were predicted (score, ≤36%). Hepatocyte incubations resulted in 14 phase I metabolites formed through N-demethylation at the trimethylhydrazino group, N-dealkylation at the piperidine ring, and oxidation at the indole and methylalanyl groups; and two phase II glucuronide conjugates occurring at the indole. We propose four metabolites detected as specific biomarkers for toxicological screening.

MBC and ECBL libraries: outstanding tools for drug discovery

Chemical libraries have become of utmost importance to boost drug discovery processes. It is widely accepted that the quality of a chemical library depends, among others, on its availability and chemical diversity which help in rising the chances of finding good hits. In this regard, our group has developed a source for useful chemicals named Medicinal and Biological Chemistry (MBC) library. It originates from more than 30 years of experience in drug design and discovery of our research group and has successfully provided effective hits for neurological, neurodegenerative and infectious diseases. Moreover, in the last years, the European research infrastructure for chemical biology EU-OPENSCREEN has generated the European Chemical Biology library (ECBL) to be used as a source of hits for drug discovery. Here we present and discuss the updated version of the MBC library (MBC v.2022), enriched with new scaffolds and containing more than 2,500 compounds together with ECBL that collects about 100,000 small molecules. To properly address the improved potentialities of the new version of our MBC library in drug discovery, up to 44 among physicochemical and pharmaceutical properties have been calculated and compared with those of other well-known publicly available libraries. For comparison, we have used ZINC20, DrugBank, ChEMBL library, ECBL and NuBBE along with an approved drug library. Final results allowed to confirm the competitive chemical space covered by MBC v.2022 and ECBL together with suitable drug-like properties. In all, we can affirm that these two libraries represent an interesting source of new hits for drug discovery.

Design, synthesis, and pharmacological evaluation of [1, 3] dioxolo-chromeno[2,3-b]pyridines as anti-seizure agents

An efficient one-pot three-component reaction for the synthesis of [1,3]dioxolo[4',5':6,7]chromeno[2,3-b]pyridines 4(a-i) has been developed. Synthesis was achieved by reacting sesamol (1), aromatic aldehydes 2(a-i), and 2-aminopropene-1,1,3-tricarbonitrile (3) in the presence of triethylamine at 100 °C under neat reaction condition. Simple operational procedure, broad substrate scope, column chromatography free separations, and high yield of products make it an efficient and largely acceptable synthetic strategy. Synthesized compounds 4(a-i) were further screened for preliminary anticonvulsant activity using MES and scPTZ tests. These analogs were also checked for neurotoxicity and hepatotoxicity. Selected active compounds have been then screened quantitatively to determine ED50 and TD50 values. Analog 4h was found effective in both preclinical seizure models with significant therapeutic/toxicity profile (4h: ED50 = 34.7 mg/kg, MES test; ED50 = 37.9 mg/kg, scPTZ test; TD50 = 308.7 mg/kg). Molecular dynamic simulation for 100 ns of compound 4h-complexed with GABAA receptor revealed good thermodynamic behavior and fairly stable interactions (4h, Docking score =  - 10.94). In conclusion, effective synthetic strategy, significant anticonvulsant activity with good toxicity profile and detailed molecular modeling studies led us to anticipate the emergence of these analogs as valid leads for the development of future effective neurotherapeutic agents.

A Robust Machine Learning Framework Built Upon Molecular Representations Predicts CYP450 Inhibition: Toward Precision in Drug Repurposing

Human cytochrome P450 (CYP450) enzymes play a crucial role in drug metabolism and pharmacokinetics. CYP450 inhibition can lead to toxicity, in particular when drugs are co-administered with other drugs and xenobiotics or in the case of polypharmacy. Predicting CYP450 inhibition is also important for rational drug discovery and development, and precision in drug repurposing. In this overarching context, digital transformation of drug discovery and development, for example, using machine and deep learning approaches, offers prospects for prediction of CYP450 inhibition through computational models. We report here the development of a majority-voting machine learning framework to classify inhibitors and noninhibitors for seven major human liver CYP450 isoforms (CYP1A2, CYP2A6, CYP2B6, CYP2C9, CYP2C19, CYP2D6, and CYP3A4). For the machine learning models reported herein, we employed interaction fingerprints that were derived from molecular docking simulations, thus adding an additional layer of information for protein-ligand interactions. The proposed machine learning framework is based on the structure of the binding site of isoforms to produce predictions beyond previously reported approaches. Also, we carried out a comparative analysis so as to identify which representation of test compounds (molecular descriptors, molecular fingerprints, or protein-ligand interaction fingerprints) affects the predictive performance of the models. This work underlines the ways in which the structure of the enzyme catalytic site influences machine learning predictions and the need for robust frameworks toward better-informed predictions.

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

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

Kolocouris A. Synthesis and Testing of Analogs of the Tuberculosis Drug Candidate SQ109 against Bacteria and Protozoa: Identification of Lead Compounds against Mycobacterium abscessus and Malaria Parasites

SQ109 is a tuberculosis drug candidate that has high potency against Mycobacterium tuberculosis and is thought to function at least in part by blocking cell wall biosynthesis by inhibiting the MmpL3 transporter. It also has activity against bacteria and protozoan parasites that lack MmpL3, where it can act as an uncoupler, targeting lipid membranes and Ca2+ homeostasis. Here, we synthesized 18 analogs of SQ109 and tested them against M. smegmatis, M. tuberculosis, M. abscessus, Bacillus subtilis, and Escherichia coli, as well as against the protozoan parasites Trypanosoma brucei, T. cruzi, Leishmania donovani, L. mexicana, and Plasmodium falciparum. Activity against the mycobacteria was generally less than with SQ109 and was reduced by increasing the size of the alkyl adduct, but two analogs were ∼4-8-fold more active than SQ109 against M. abscessus, including a highly drug-resistant strain harboring an A309P mutation in MmpL3. There was also better activity than found with SQ109 with other bacteria and protozoa. Of particular interest, we found that the adamantyl C-2 ethyl, butyl, phenyl, and benzyl analogs had 4-10× increased activity against P. falciparum asexual blood stages, together with low toxicity to a human HepG2 cell line, making them of interest as new antimalarial drug leads. We also used surface plasmon resonance to investigate the binding of inhibitors to MmpL3 and differential scanning calorimetry to investigate binding to lipid membranes. There was no correlation between MmpL3 binding and M. tuberculosis or M. smegmatis cell activity, suggesting that MmpL3 is not a major target in mycobacteria. However, some of the more active species decreased lipid phase transition temperatures, indicating increased accumulation in membranes, which is expected to lead to enhanced uncoupler activity.

Natural inhibitors for acetylcholinesterase and autophagy modulators as effective antagonists for tau and β-amyloid in Alzheimer's rat model

Background: Phytochemicals have amazing biological effects in relation to age-related illnesses and are increasingly being studied in clinical trials. The goal of this study was to examine the effectiveness of the aqueous extracts of Rosmarinus officinalis L. (Rosemary) and Crocus sativus L. (Saffron) and their combinations as tau and β-amyloid antagonists in an Alzheimer's rat model. Methods: AlCl₃ and D-galactose (150 & 300 mg/kg) were used to create the Alzheimer's neuroinflammation rat model. The animals were subsequently given the two extracts and their combinations (500 mg/kg) along 15 days. The cognitive impairment, oxidative stress, tau & amyloid neuroproteins, acetylcholine, acetylcholinesterase neurotransmitters, proinflammatory cytokines, LC3 as an autophagy marker, computational analysis, and morphological alterations were all assessed. Results: When compared to the conventional donepezil and normal groups, the treated groups showed a significant improvement in all calculated parameters. The cortex and hippocampus have a better morphological appearance. In silico analysis found that these extracts may have an affinity for and impede the activity of some proteins thought to be essential regulators of disease progression. Conclusion: Rosemary and Saffron extracts by the power of their constituents were able to alleviate the neurotoxicity of AlCl₃ & D-galactose and regulate the natural autophagy process.

In Silico and In Vitro Approach to Assess Direct Allosteric AMPK Activators from Nature

The 5'-adenosine monophosphate-activated protein kinase (AMPK) is an important metabolic regulator. Its allosteric drug and metabolite binding (ADaM) site was identified as an attractive target for direct AMPK activation and holds promise as a novel mechanism for the treatment of metabolic diseases. With the exception of lusianthridin and salicylic acid, no natural product (NP) is reported so far to directly target the ADaM site. For the streamlined assessment of direct AMPK activators from the pool of NPs, an integrated workflow using in silico and in vitro methods was applied. Virtual screening combining a 3D shape-based approach and docking identified 21 NPs and NP-like molecules that could potentially activate AMPK. The compounds were purchased and tested in an in vitro AMPK α 1 β 1 γ 1 kinase assay. Two NP-like virtual hits were identified, which, at 30 µM concentration, caused a 1.65-fold (± 0.24) and a 1.58-fold (± 0.17) activation of AMPK, respectively. Intriguingly, using two different evaluation methods, we could not confirm the bioactivity of the supposed AMPK activator lusianthridin, which rebuts earlier reports.

Human Hepatocyte 4-Acetoxy-N,N-Diisopropyltryptamine Metabolite Profiling by Reversed-Phase Liquid Chromatography Coupled with High-Resolution Tandem Mass Spectrometry

Tryptamine intoxications and fatalities are increasing, although these novel psychoactive substances (NPS) are not controlled in most countries. There are few data on the metabolic pathways and enzymes involved in tryptamine biotransformation. 4-acetoxy-N,N-diisopropyltryptamine (4-AcO-DiPT) is a synthetic tryptamine related to 4-hydroxy-N,N-diisopropyltryptamine (4-OH-DiPT), 4-acetyloxy-N,N-dipropyltryptamine (4-AcO-DPT), and 4-acetoxy-N,N-dimethyltryptamine (4-AcO-DMT). The aim of this study was to determine the best 4-AcO-DiPT metabolites to identify 4-AcO-DiPT consumption through human hepatocyte metabolism and high-resolution mass spectrometry. 4-AcO-DiPT metabolites were predicted in silico with GLORYx freeware to assist in metabolite identification. 4-AcO-DiPT was incubated with 10-donor-pooled human hepatocytes and sample analysis was performed with reversed-phase liquid chromatography coupled with high-resolution tandem mass spectrometry (LC-HRMS/MS) in positive- and negative-ion modes. Software-assisted LC-HRMS/MS raw data mining was performed. A total of 47 phase I and II metabolites were predicted, and six metabolites were identified after 3 h incubation following ester hydrolysis, O-glucuronidation, O-sulfation, N-oxidation, and N-dealkylation. All second-generation metabolites were derived from the only first-generation metabolite detected after ester hydrolysis (4-OH-DiPT). The metabolite with the second-most-intense signal was 4-OH-iPT-sulfate followed by 4-OH-DiPT-glucuronide, indicating that glucuronidation and sulfation are common in this tryptamine’s metabolic pathway. 4-OH-DiPT, 4-OH-iPT, and 4-OH-DiPT-N-oxide are suggested as optimal biomarkers to identify 4-AcO-DiPT consumption.

Discovery and Anticancer Activity of Novel 1,3,4-Thiadiazole- and Aziridine-Based Indolin-2-ones via In Silico Design Followed by Supramolecular Green Synthesis

Three crucial anticancer scaffolds, namely indolin-2-one, 1,3,4-thiadiazole, and aziridine, are explored to synthesize virtually screened target molecules based on the c-KIT kinase protein. The stem cell factor receptor c-KIT was selected as target because most U.S. FDA-approved receptor tyrosine kinase inhibitors bearing the indolin-2-one scaffold profoundly inhibit c-KIT. Molecular hybrids of indolin-2-one with 1,3,4-thiadiazole (IIIa-m) and aziridine (VIa and VIc) were afforded through a modified Schiff base green synthesis using β-cyclodextrin-SO₃H in water as a recyclable proton-donor catalyst. A computational study found that indolin-2,3-dione forms a supramolecular inclusion complex with β-cyclodextrin-SO₃H through noncovalent interactions. A molecular docking study of all the synthesized compounds was executed on the c-KIT kinase domain, and most compounds displayed binding affinities similar to that of Sunitinib. On the basis of the pharmacokinetic significance of the aryl thioether linkage in small molecules, 1,3,4-thiadiazole hybrids (IIIa-m) were extended to a new series of 3-((5-(phenylthio)-1,3,4-thiadiazol-2-yl)imino)indolin-2-ones (IVa-m) via thioetherification using bis(triphenylphosphine)palladium(II)dichloride as the catalyst for C-S bond formation. Target compounds were tested against NCI-60 human cancer cell lines for a single-dose concentration. Among all three series of indolin-2-ones, the majority of compounds demonstrated broad-spectrum activity toward various cancer cell lines. Compounds IVc and VIc were further evaluated for a five-dose anticancer study. Compound IVc showed a potent activity of IC₅₀ = 1.47 μM against a panel of breast cancer cell lines, whereas compound VIc exhibited the highest inhibition for a panel of colon cancer cell lines at IC₅₀ = 1.40 μM. In silico ADME property descriptors of all the target molecules are in an acceptable range. Machine learning algorithms were used to examine the metabolites and phase I and II regioselectivities of compounds IVc and VIc, and the results suggested that these two compounds could be potential leads for the treatment of cancer.

In silico, in vitro, and in vivo human metabolism of acetazolamide, a carbonic anhydrase inhibitor and common "diuretic and masking agent" in doping

Acetazolamide (ACZ) is a carbonic anhydrase inhibitor prescribed for the treatment of various pathologies. It is also used in doping and is prohibited in and out of sportive competitions. ACZ was reported not to undergo metabolization. However, the detection of ACZ metabolites may be critical for documenting ACZ use. We aimed to further investigate ACZ metabolic fate in humans. ACZ putative metabolites were generated in silico to assist in metabolite identification. ACZ was incubated with primary human hepatocytes to identify in vitro metabolites (10 µmol/l ACZ and 10⁶ cells/ml), and urine and plasma samples from patients receiving a single 5.0 mg/kg BW PO ACZ dose were analyzed to confirm the results in vivo. Analyses were performed with reversed-phase liquid chromatography and hydrophilic interaction chromatography coupled with high-resolution tandem mass spectrometry (RPLC-HRMS/MS and HILIC-HRMS/MS, respectively). Data were screened with a software-assisted targeted/untargeted workflow. ACZ was quantified in urine samples with creatinine normalization. We identified two metabolites in hepatocyte incubations and three additional metabolites in urine and plasma. Major transformations included cysteine conjugation, glucuronidation, and N-acetylation. All metabolites were detected in plasma, 1.5 h after intake. Major metabolites were detected in urine from 0.25 to 24 h (last collection) after intake. As opposed to the literature, ACZ does undergo metabolization in humans. We propose ACZ, ACZ-Cys, and N-acetyl-ACZ in urine, and ACZ and N-acetyl-ACZ in plasma as specific biomarkers of ACZ intake in doping.

Novel 1,3,5-Triazinyl Aminobenzenesulfonamides Incorporating Aminoalcohol, Aminochalcone and Aminostilbene Structural Motifs as Potent Anti-VRE Agents, and Carbonic Anhydrases I, II, VII, IX, and XII Inhibitors

A series of 1,3,5-triazinyl aminobenzenesulfonamides substituted by aminoalcohol, aminostilbene, and aminochalcone structural motifs was synthesized as potential human carbonic anhydrase (hCA) inhibitors. The compounds were evaluated on their inhibition of tumor-associated hCA IX and hCA XII, hCA VII isoenzyme present in the brain, and physiologically important hCA I and hCA II. While the test compounds had only a negligible effect on physiologically important isoenzymes, many of the studied compounds significantly affected the hCA IX isoenzyme. Several compounds showed activity against hCA XII; (E)-4-{2-[(4-[(2,3-dihydroxypropyl)amino]-6-[(4-styrylphenyl)amino]-1,3,5-triazin-2-yl)amino]ethyl}benzenesulfonamide (31) and (E)-4-{2-[(4-[(4-hydroxyphenyl)amino]-6-[(4-styrylphenyl)amino]-1,3,5-triazin-2-yl)amino]ethyl}benzenesulfonamide (32) were the most effective inhibitors with K_Is = 4.4 and 5.9 nM, respectively. In addition, the compounds were tested against vancomycin-resistant Enterococcus faecalis (VRE) isolates. (E)-4-[2-({4-[(4-cinnamoylphenyl)amino]-6-[(4-hydroxyphenyl)amino]-1,3,5-triazin-2-yl}amino)ethyl]benzenesulfonamide (21) (MIC = 26.33 µM) and derivative 32 (MIC range 13.80-55.20 µM) demonstrated the highest activity against all tested strains. The most active compounds were evaluated for their cytotoxicity against the Human Colorectal Tumor Cell Line (HCT116 p53 +/+). Only 4,4'-[(6-chloro-1,3,5-triazin-2,4-diyl)bis(iminomethylene)]dibenzenesulfonamide (7) and compound 32 demonstrated an IC₅₀ of ca. 6.5 μM; otherwise, the other selected derivatives did not show toxicity at concentrations up to 50 µM. The molecular modeling and docking of active compounds into various hCA isoenzymes, including bacterial carbonic anhydrase, specifically α-CA present in VRE, was performed to try to outline a possible mechanism of selective anti-VRE activity.

Novel DYRK1A Inhibitor Rescues Learning and Memory Deficits in a Mouse Model of Down Syndrome

Down syndrome (DS) is a complex genetic disorder associated with substantial physical, cognitive, and behavioral challenges. Due to better treatment options for the physical co-morbidities of DS, the life expectancy of individuals with DS is beginning to approach that of the general population. However, the cognitive deficits seen in individuals with DS still cannot be addressed pharmacologically. In young individuals with DS, the level of intellectual disability varies from mild to severe, but cognitive ability generally decreases with increasing age, and all individuals with DS have early onset Alzheimer’s disease (AD) pathology by the age of 40. The present study introduces a novel inhibitor for the protein kinase DYRK1A, a key controlling kinase whose encoding gene is located on chromosome 21. The novel inhibitor is well characterized for use in mouse models and thus represents a valuable tool compound for further DYRK1A research.

Computer-Aided Design and Synthesis of a New Class of PEX14 Inhibitors: Substituted 2,3,4,5-Tetrahydrobenzo[F][1,4]oxazepines as Potential New Trypanocidal Agents

African and American trypanosomiases are estimated to affect several million people across the world, with effective treatments distinctly lacking. New, ideally oral, treatments with higher efficacy against these diseases are desperately needed. Peroxisomal import matrix (PEX) proteins represent a very interesting target for structure- and ligand-based drug design. The PEX5-PEX14 protein-protein interface in particular has been highlighted as a target, with inhibitors shown to disrupt essential cell processes in trypanosomes, leading to cell death. In this work, we present a drug development campaign that utilizes the synergy between structural biology, computer-aided drug design, and medicinal chemistry in the quest to discover and develop new potential compounds to treat trypanosomiasis by targeting the PEX14-PEX5 interaction. Using the structure of the known lead compounds discovered by Dawidowski et al. as the template for a chemically advanced template search (CATS) algorithm, we performed scaffold-hopping to obtain a new class of compounds with trypanocidal activity, based on 2,3,4,5-tetrahydrobenzo[f][1,4]oxazepines chemistry. The initial compounds obtained were taken forward to a first round of hit-to-lead optimization by synthesis of derivatives, which show activities in the range of low- to high-digit micromolar IC₅₀ in the in vitro tests. The NMR measurements confirm binding to PEX14 in solution, while immunofluorescent microscopy indicates disruption of protein import into the glycosomes, indicating that the PEX14-PEX5 protein-protein interface was successfully disrupted. These studies result in development of a novel scaffold for future lead optimization, while ADME testing gives an indication of further areas of improvement in the path from lead molecules toward a new drug active against trypanosomes.

Computational Applications in Secondary Metabolite Discovery (CAiSMD): an online workshop

We report the major conclusions of the online open-access workshop "Computational Applications in Secondary Metabolite Discovery (CAiSMD)" that took place from 08 to 10 March 2021. Invited speakers from academia and industry and about 200 registered participants from five continents (Africa, Asia, Europe, South America, and North America) took part in the workshop. The workshop highlighted the potential applications of computational methodologies in the search for secondary metabolites (SMs) or natural products (NPs) as potential drugs and drug leads. During 3 days, the participants of this online workshop received an overview of modern computer-based approaches for exploring NP discovery in the "omics" age. The invited experts gave keynote lectures, trained participants in hands-on sessions, and held round table discussions. This was followed by oral presentations with much interaction between the speakers and the audience. Selected applicants (early-career scientists) were offered the opportunity to give oral presentations (15 min) and present posters in the form of flash presentations (5 min) upon submission of an abstract. The final program available on the workshop website ( https://caismd.indiayouth.info/ ) comprised of 4 keynote lectures (KLs), 12 oral presentations (OPs), 2 round table discussions (RTDs), and 5 hands-on sessions (HSs). This meeting report also references internet resources for computational biology in the area of secondary metabolites that are of use outside of the workshop areas and will constitute a long-term valuable source for the community. The workshop concluded with an online survey form to be completed by speakers and participants for the goal of improving any subsequent editions.

Machine Learning in Drug Discovery: A Review

This review provides the feasible literature on drug discovery through ML tools and techniques that are enforced in every phase of drug development to accelerate the research process and deduce the risk and expenditure in clinical trials. Machine learning techniques improve the decision-making in pharmaceutical data across various applications like QSAR analysis, hit discoveries, de novo drug architectures to retrieve accurate outcomes. Target validation, prognostic biomarkers, digital pathology are considered under problem statements in this review. ML challenges must be applicable for the main cause of inadequacy in interpretability outcomes that may restrict the applications in drug discovery. In clinical trials, absolute and methodological data must be generated to tackle many puzzles in validating ML techniques, improving decision-making, promoting awareness in ML approaches, and deducing risk failures in drug discovery.

Metabolism-Coupled Cell-Independent Acetylcholinesterase Activity Assay for Evaluation of the Effects of Chlorination on Diazinon Toxicity

Drinking water quality guideline values for toxic compounds are determined based on their acceptable daily intake. The toxicological end point for determining the acceptable daily intake of most organophosphorus insecticides is inhibition of acetylcholinesterase (AChE). Although insecticides ingested with drinking water are partly metabolized by the liver before transport to the rest of the body, no current cell-independent AChE activity assay takes the effects of metabolism into account. Here, we incorporated metabolism into a cell-independent AChE activity assay and then evaluated the change in anti-AChE activity during chlorination of a solution containing the organophosphorus insecticide diazinon. The anti-AChE activities of solutions of diazinon or diazinon-oxon, the major transformation product of diazinon during chlorination, were dramatically changed by metabolism: the activity of diazinon solution was markedly increased, whereas that of diazinon-oxon solution was slightly decreased, clearly indicating the importance of incorporating metabolism into assays examining toxicity after oral ingestion. Upon chlorination, diazinon was completely transformed, in part to diazinon-oxon. Although diazinon solution without metabolism did not show anti-AChE activity before chlorination, it did after chlorination. In contrast, with metabolism, diazinon solution did show anti-AChE activity before chlorination, but chlorination gradually decreased this activity over time. The observed anti-AChE activities were attributable solely to diazinon and diazinon-oxon having been contained in the samples before metabolism, clearly suggesting that the presence not only of diazinon but also of diazinon-oxon should be monitored in drinking water. Further examination using a combination of tandem mass spectrometry and in silico site-of-metabolism analyses revealed the structure of a single metabolite that was responsible for the observed anti-AChE activity after metabolism. However, because this compound is produced via metabolism in the human body after oral ingestion of diazinon, its presence in drinking water need not be monitored and regulated.

CYPstrate: A Set of Machine Learning Models for the Accurate Classification of Cytochrome P450 Enzyme Substrates and Non-Substrates

The interaction of small organic molecules such as drugs, agrochemicals, and cosmetics with cytochrome P450 enzymes (CYPs) can lead to substantial changes in the bioavailability of active substances and hence consequences with respect to pharmacological efficacy and toxicity. Therefore, efficient means of predicting the interactions of small organic molecules with CYPs are of high importance to a host of different industries. In this work, we present a new set of machine learning models for the classification of xenobiotics into substrates and non-substrates of nine human CYP isozymes: CYPs 1A2, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1, and 3A4. The models are trained on an extended, high-quality collection of known substrates and non-substrates and have been subjected to thorough validation. Our results show that the models yield competitive performance and are favorable for the detection of CYP substrates. In particular, a new consensus model reached high performance, with Matthews correlation coefficients (MCCs) between 0.45 (CYP2C8) and 0.85 (CYP3A4), although at the cost of coverage. The best models presented in this work are accessible free of charge via the "CYPstrate" module of the New E-Resource for Drug Discovery (NERDD).

Evidence of Pyrimethamine and Cycloguanil Analogues as Dual Inhibitors of Trypanosoma brucei Pteridine Reductase and Dihydrofolate Reductase

Trypanosoma and Leishmania parasites are the etiological agents of various threatening neglected tropical diseases (NTDs), including human African trypanosomiasis (HAT), Chagas disease, and various types of leishmaniasis. Recently, meaningful progresses in the treatment of HAT, due to Trypanosoma brucei (Tb), have been achieved by the introduction of fexinidazole and the combination therapy eflornithine-nifurtimox. Nevertheless, due to drug resistance issues and the exitance of animal reservoirs, the development of new NTD treatments is still required. For this purpose, we explored the combined targeting of two key folate enzymes, dihydrofolate reductase (DHFR) and pteridine reductase 1 (PTR1). We formerly showed that the TbDHFR inhibitor cycloguanil (CYC) also targets TbPTR1, although with reduced affinity. Here, we explored a small library of CYC analogues to understand how their substitution pattern affects the inhibition of both TbPTR1 and TbDHFR. Some novel structural features responsible for an improved, but preferential, ability of CYC analogues to target TbPTR1 were disclosed. Furthermore, we showed that the known drug pyrimethamine (PYR) effectively targets both enzymes, also unveiling its binding mode to TbPTR1. The structural comparison between PYR and CYC binding modes to TbPTR1 and TbDHFR provided key insights for the future design of dual inhibitors for HAT therapy.

Open, High-Resolution EI+ Spectral Library of Anthropogenic Compounds

To address the lack of high-resolution electron ionisation mass spectral libraries (HR-[EI+]-MS) for environmental chemicals, a retention-indexed HR-[EI+]-MS library has been constructed following analysis of authentic compounds via GC-Orbitrap MS. The library is freely provided alongside a compound database of predicted physicochemical properties. Currently, the library contains over 350 compounds from 56 compound classes and includes a range of legacy and emerging contaminants. The RECETOX Exposome HR-[EI+]-MS library expands the number of freely available resources for use in full-scan chemical exposure studies and is available at: https://doi.org/10.5281/zenodo.4471217.

Skin Doctor: Machine Learning Models for Skin Sensitization Prediction that Provide Estimates and Indicators of Prediction Reliability

The ability to predict the skin sensitization potential of small organic molecules is of high importance to the development and safe application of cosmetics, drugs and pesticides. One of the most widely accepted methods for predicting this hazard is the local lymph node assay (LLNA). The goal of this work was to develop in silico models for the prediction of the skin sensitization potential of small molecules that go beyond the state of the art, with larger LLNA data sets and, most importantly, a robust and intuitive definition of the applicability domain, paired with additional indicators of the reliability of predictions. We explored a large variety of molecular descriptors and fingerprints in combination with random forest and support vector machine classifiers. The most suitable models were tested on holdout data, on which they yielded competitive performance (Matthews correlation coefficients up to 0.52; accuracies up to 0.76; areas under the receiver operating characteristic curves up to 0.83). The most favorable models are available via a public web service that, in addition to predictions, provides assessments of the applicability domain and indicators of the reliability of the individual predictions.