Trust, but verify: on the importance of chemical structure curation in cheminformatics and QSAR modeling research

Enhancing hERG Risk Assessment with Interpretable Classificatory and Regression Models

The human Ether-à-go-go-Related Gene (hERG) is a transmembrane protein that regulates cardiac action potential, and its inhibition can induce a potentially deadly cardiac syndrome. In vitro tests help identify hERG blockers at early stages; however, the high cost motivates searching for alternative, cost-effective methods. The primary goal of this study was to enhance the Pred-hERG tool for predicting hERG blockage. To achieve this, we developed new QSAR models that incorporated additional data, updated existing classificatory and multiclassificatory models, and introduced new regression models. Notably, we integrated SHAP (SHapley Additive exPlanations) values to offer a visual interpretation of these models. Utilizing the latest data from ChEMBL v30, encompassing over 14,364 compounds with hERG data, our binary and multiclassification models outperformed both the previous iteration of Pred-hERG and all publicly available models. Notably, the new version of our tool introduces a regression model for predicting hERG activity (pIC50). The optimal model demonstrated an R2 of 0.61 and an RMSE of 0.48, surpassing the only available regression model in the literature. Pred-hERG 5.0 now offers users a swift, reliable, and user-friendly platform for the early assessment of chemically induced cardiotoxicity through hERG blockage. The tool provides versatile outcomes, including (i) classificatory predictions of hERG blockage with prediction reliability, (ii) multiclassificatory predictions of hERG blockage with reliability, (iii) regression predictions with estimated pIC50 values, and (iv) probability maps illustrating the contribution of chemical fragments for each prediction. Furthermore, we implemented explainable AI analysis (XAI) to visualize SHAP values, providing insights into the contribution of each feature to binary classification predictions. A consensus prediction calculated based on the predictions of the three developed models is also present to assist the user's decision-making process. Pred-hERG 5.0 has been designed to be user-friendly, making it accessible to users without computational or programming expertise. The tool is freely available at http://predherg.labmol.com.br.

PubChem synonym filtering process using crowdsourcing

PubChem ( https://pubchem.ncbi.nlm.nih.gov ) is a public chemical information resource containing more than 100 million unique chemical structures. One of the most requested tasks in PubChem and other chemical databases is to search chemicals by name (also commonly called a "chemical synonym"). PubChem performs this task by looking up chemical synonym-structure associations provided by individual depositors to PubChem. In addition, these synonyms are used for many purposes, including creating links between chemicals and PubMed articles (using Medical Subject Headings (MeSH) terms). However, these depositor-provided name-structure associations are subject to substantial discrepancies within and between depositors, making it difficult to unambiguously map a chemical name to a specific chemical structure. The present paper describes PubChem's crowdsourcing-based synonym filtering strategy, which resolves inter- and intra-depositor discrepancies in synonym-structure associations as well as in the chemical-MeSH associations. The PubChem synonym filtering process was developed based on the analysis of four crowd-voting strategies, which differ in the consistency threshold value employed (60% vs 70%) and how to resolve intra-depositor discrepancies (a single vote vs. multiple votes per depositor) prior to inter-depositor crowd-voting. The agreement of voting was determined at six levels of chemical equivalency, which considers varying isotopic composition, stereochemistry, and connectivity of chemical structures and their primary components. While all four strategies showed comparable results, Strategy I (one vote per depositor with a 60% consistency threshold) resulted in the most synonyms assigned to a single chemical structure as well as the most synonym-structure associations disambiguated at the six chemical equivalency contexts. Based on the results of this study, Strategy I was implemented in PubChem's filtering process that cleans up synonym-structure associations as well as chemical-MeSH associations. This consistency-based filtering process is designed to look for a consensus in name-structure associations but cannot attest to their correctness. As a result, it can fail to recognize correct name-structure associations (or incorrect ones), for example, when a synonym is provided by only one depositor or when many contributors are incorrect. However, this filtering process is an important starting point for quality control in name-structure associations in large chemical databases like PubChem.

In silico screening of LRRK2 WDR domain inhibitors using deep docking and free energy simulations

The Critical Assessment of Computational Hit-Finding Experiments (CACHE) Challenge series is focused on identifying small molecule inhibitors of protein targets using computational methods. Each challenge contains two phases, hit-finding and follow-up optimization, each of which is followed by experimental validation of the computational predictions. For the CACHE Challenge #1, the Leucine-Rich Repeat Kinase 2 (LRRK2) WD40 Repeat (WDR) domain was selected as the target for in silico hit-finding and optimization. Mutations in LRRK2 are the most common genetic cause of the familial form of Parkinson's disease. The LRRK2 WDR domain is an understudied drug target with no known molecular inhibitors. Herein we detail the first phase of our winning submission to the CACHE Challenge #1. We developed a framework for the high-throughput structure-based virtual screening of a chemically diverse small molecule space. Hit identification was performed using the large-scale Deep Docking (DD) protocol followed by absolute binding free energy (ABFE) simulations. ABFEs were computed using an automated molecular dynamics (MD)-based thermodynamic integration (TI) approach. 4.1 billion ligands from Enamine REAL were screened with DD followed by ABFEs computed by MD TI for 793 ligands. 76 ligands were prioritized for experimental validation, with 59 compounds successfully synthesized and 5 compounds identified as hits, yielding a 8.5% hit rate. Our results demonstrate the efficacy of the combined DD and ABFE approaches for hit identification for a target with no previously known hits. This approach is widely applicable for the efficient screening of ultra-large chemical libraries as well as rigorous protein-ligand binding affinity estimation leveraging modern computational resources.

Multi-Endpoint Acute Toxicity Assessment of Organic Compounds Using Large-Scale Machine Learning Modeling

In recent years, alternative animal testing methods such as computational and machine learning approaches have become increasingly crucial for toxicity testing. However, the complexity and scarcity of available biomedical data challenge the development of predictive models. Combining nonlinear machine learning together with multicondition descriptors offers a solution for using data from various assays to create a robust model. This work applies multicondition descriptors (MCDs) to develop a QSTR (Quantitative Structure-Toxicity Relationship) model based on a large toxicity data set comprising more than 80,000 compounds and 59 different end points (122,572 data points). The prediction capabilities of developed single-task multi-end point machine learning models as well as a novel data analysis approach with the use of Convolutional Neural Networks (CNN) are discussed. The results show that using MCDs significantly improves the model and using them with CNN-1D yields the best result (R2train = 0.93, R2ext = 0.70). Several structural features showed a high level of contribution to the toxicity, including van der Waals surface area (VSA), number of nitrogen-containing fragments (nN+), presence of S-P fragments, ionization potential, and presence of C-N fragments. The developed models can be very useful tools to predict the toxicity of various compounds under different conditions, enabling quick toxicity assessment of new compounds.

Pharmacophore Virtual Screening Identifies Riboflavin as an Inhibitor of the Schistosome Cathepsin B1 Protease with Antiparasitic Activity

Schistosomiasis is a neglected disease of poverty that affects over 200 million people worldwide and relies on a single drug for therapy. The cathepsin B1 cysteine protease (SmCB1) of Schistosoma mansoni has been investigated as a potential target. Here, a structure-based pharmacophore virtual screening (VS) approach was used on a data set of approved drugs to identify potential antischistosomal agents targeting SmCB1. Pharmacophore (PHP) models underwent validation through receiver operating characteristics curves achieving values >0.8. The data highlighted riboflavin (RBF) as a compound of particular interest. A 1 μs molecular dynamics simulation demonstrated that RBF altered the conformation of SmCB1, causing the protease's binding site to close around RBF while maintaining the protease's overall integrity. RBF inhibited the activity of SmCB1 at low micromolar values and killed the parasite in vitro. Finally, in a murine model of S. mansoni infection, oral administration of 100 mg/kg RBF for 7 days significantly decreased worm burdens by ∼20% and had a major impact on intestinal and fecal egg burdens, which were decreased by ∼80%.

Multitarget anti-parasitic activities of isoquinoline alkaloids isolated from Hippeastrum aulicum (Amaryllidaceae)

BACKGROUND

Chagas disease and leishmaniasis affect a significant portion of the Latin American population and still lack efficient treatments. In this context, natural products emerge as promising compounds for developing more effective therapies, aiming to mitigate side effects and drug resistance. Notably, species from the Amaryllidaceae family emerge as potential reservoirs of antiparasitic agents due to the presence of diverse biologically active alkaloids.

PURPOSE

To assess the anti-Trypanosoma cruzi and anti-Leishmania infantum activity of five isolated alkaloids from Hippeastrum aulicum Herb. (Amaryllidaceae) against different life stages of the parasites using in silico and in vitro assays. Furthermore, molecular docking was employed to evaluate the interaction of the most active alkaloids.

METHODS

Five natural isoquinoline alkaloids isolated in suitable quantities for in vitro testing underwent preliminary in silico analysis to predict their potential efficacy against Trypanosoma cruzi (amastigote and trypomastigote forms) and Leishmania infantum (amastigote and promastigote forms). The in vitro antiparasitic activity and mammalian cytotoxicity were investigated with a subsequent comparison of both analysis (in silico and in vitro) findings. Additionally, this study employed the molecular docking technique, utilizing cruzain (T. cruzi) and sterol 14α-demethylase (CYP51, L. infantum) as crucial biological targets for parasite survival, specifically focusing on compounds that exhibited promising activities against both parasites.

RESULTS

Through computational techniques, it was identified that the alkaloids haemanthamine (1) and lycorine (8) were the most active against T. cruzi (amastigote and trypomastigote) and L. infantum (amastigote and promastigote), while also revealing unprecedented activity of alkaloid 7‑methoxy-O-methyllycorenine (6). The in vitro analysis confirmed the in silico tests, in which compound 1 presented the best activities against the promastigote and amastigote forms of L. infantum with half-maximal inhibitory concentration (IC50) 0.6 µM and 1.78 µM, respectively. Compound 8 exhibited significant activity against the amastigote form of T. cruzi (IC50 7.70 µM), and compound 6 demonstrated activity against the trypomastigote forms of T. cruzi and amastigote of L. infantum, with IC50 values of 89.55 and 86.12 µM, respectively. Molecular docking analyses indicated that alkaloids 1 and 8 exhibited superior interaction energies compared to the inhibitors.

CONCLUSION

The hitherto unreported potential of compound 6 against T. cruzi trypomastigotes and L. infantum amastigotes is now brought to the forefront. Furthermore, the acquired dataset signifies that the isolated alkaloids 1 and 8 from H. aulicum might serve as prototypes for subsequent structural refinements aimed at the exploration of novel leads against both T. cruzi and L. infantum parasites.

Modeling interactions between Heparan sulfate and proteins based on the Heparan sulfate microarray analysis

Heparan sulfate (HS), a sulfated polysaccharide abundant in the extracellular matrix, plays pivotal roles in various physiological and pathological processes by interacting with proteins. Investigating the binding selectivity of HS oligosaccharides to target proteins is essential, but the exhaustive inclusion of all possible oligosaccharides in microarray experiments is impractical. To address this challenge, we present a hybrid pipeline that integrates microarray and in silico techniques to design oligosaccharides with desired protein affinity. Using fibroblast growth factor 2 (FGF2) as a model protein, we assembled an in-house dataset of HS oligosaccharides on microarrays and developed two structural representations: a standard representation with all atoms explicit and a simplified representation with disaccharide units as "quasi-atoms." Predictive Quantitative Structure-Activity Relationship (QSAR) models for FGF2 affinity were developed using the Random Forest (RF) algorithm. The resulting models, considering the applicability domain, demonstrated high predictivity, with a correct classification rate of 0.81-0.80 and improved positive predictive values (PPV) up to 0.95. Virtual screening of 40 new oligosaccharides using the simplified model identified 15 computational hits, 11 of which were experimentally validated for high FGF2 affinity. This hybrid approach marks a significant step toward the targeted design of oligosaccharides with desired protein interactions, providing a foundation for broader applications in glycobiology.

Toward the Reconciliation of Inconsistent Molecular Structures from Biochemical Databases

Information on the structure of molecules, retrieved via biochemical databases, plays a pivotal role in various disciplines, including metabolomics, systems biology, and drug discovery. No such database can be complete and it is often necessary to incorporate data from several sources. However, the molecular structure for a given compound is not necessarily consistent between databases. This article presents StructRecon, a novel tool for resolving unique molecular structures from database identifiers. Currently, identifiers from BiGG, ChEBI, Escherichia coli Metabolome Database (ECMDB), MetaNetX, and PubChem are supported. StructRecon traverses the cross-links between entries in different databases to construct what we call identifier graphs. The goal of these graphs is to offer a more complete view of the total information available on a given compound across all the supported databases. To reconcile discrepancies met during the traversal of the databases, we develop an extensible model for molecular structure supporting multiple independent levels of detail, which allows standardization of the structure to be applied iteratively. In some cases, our standardization approach results in multiple candidate structures for a given compound, in which case a random walk-based algorithm is used to select the most likely structure among incompatible alternatives. As a case study, we applied StructRecon to the EColiCore2 model. We found at least one structure for 98.66% of its compounds, which is more than twice as many as possible when using the databases in more standard ways not considering the complex network of cross-database references captured by our identifier graphs. StructRecon is open-source and modular, which enables support for more databases in the future.

Best practices for machine learning in antibody discovery and development

In the past 40 years, therapeutic antibody discovery and development have advanced considerably, with machine learning (ML) offering a promising way to speed up the process by reducing costs and the number of experiments required. Recent progress in ML-guided antibody design and development (D&D) has been hindered by the diversity of data sets and evaluation methods, which makes it difficult to conduct comparisons and assess utility. Establishing standards and guidelines will be crucial for the wider adoption of ML and the advancement of the field. This perspective critically reviews current practices, highlights common pitfalls and proposes method development and evaluation guidelines for various ML-based techniques in therapeutic antibody D&D. Addressing challenges across the ML process, best practices are recommended for each stage to enhance reproducibility and progress.

A Chemical Structure and Machine Learning Approach to Assess the Potential Bioactivity of Endogenous Metabolites and Their Association with Early Childhood Systemic Inflammation

Metabolomics has gained much attention due to its potential to reveal molecular disease mechanisms and present viable biomarkers. This work uses a panel of untargeted serum metabolomes from 602 children from the COPSAC2010 mother-child cohort. The annotated part of the metabolome consists of 517 chemical compounds curated using automated procedures. We created a filtering method for the quantified metabolites using predicted quantitative structure-bioactivity relationships for the Tox21 database on nuclear receptors and stress response in cell lines. The metabolites measured in the children's serums are predicted to affect specific targeted models, known for their significance in inflammation, immune function, and health outcomes. The targets from Tox21 have been used as targets with quantitative structure-activity relationships (QSARs). They were trained for ~7000 structures, saved as models, and then applied to the annotated metabolites to predict their potential bioactivities. The models were selected based on strict accuracy criteria surpassing random effects. After application, 52 metabolites showed potential bioactivity based on structural similarity with known active compounds from the Tox21 set. The filtered compounds were subsequently used and weighted by their bioactive potential to show an association with early childhood hs-CRP levels at six months in a linear model supporting a physiological adverse effect on systemic low-grade inflammation.

Exploring different classification-dependent QSAR modelling strategies for HDAC3 inhibitors in search of meaningful structural contributors

Histone deacetylase 3 (HDAC3), a Zn2+-dependent class I HDACs, contributes to numerous disorders such as neurodegenerative disorders, diabetes, cardiovascular disease, kidney disease and several types of cancers. Therefore, the development of novel and selective HDAC3 inhibitors might be promising to combat such diseases. Here, different classification-based molecular modelling studies such as Bayesian classification, recursive partitioning (RP), SARpy and linear discriminant analysis (LDA) were conducted on a set of HDAC3 inhibitors to pinpoint essential structural requirements contributing to HDAC3 inhibition followed by molecular docking study and molecular dynamics (MD) simulation analyses. The current study revealed the importance of hydroxamate function for Zn2+ chelation as well as hydrogen bonding interaction with Tyr298 residue. The importance of hydroxamate function for higher HDAC3 inhibition was noticed in the case of Bayesian classification, recursive partitioning and SARpy models. Also, the importance of substituted thiazole ring was revealed, whereas the presence of linear alkyl groups with carboxylic acid function, any type of ester function, benzodiazepine moiety and methoxy group in the molecular structure can be detrimental to HDAC3 inhibition. Therefore, this study can aid in the design and discovery of effective novel HDAC3 inhibitors in the future.

Pharmacokinetics Profiler (PhaKinPro): Model Development, Validation, and Implementation as a Web Tool for Triaging Compounds with Undesired Pharmacokinetics Profiles

Computational models that predict pharmacokinetic properties are critical to deprioritize drug candidates that emerge as hits in high-throughput screening campaigns. We collected, curated, and integrated a database of compounds tested in 12 major end points comprising over 10,000 unique molecules. We then employed these data to build and validate binary quantitative structure-activity relationship (QSAR) models. All trained models achieved a correct classification rate above 0.60 and a positive predictive value above 0.50. To illustrate their utility in drug discovery, we used these models to predict the pharmacokinetic properties for drugs in the NCATS Inxight Drugs database. In addition, we employed the developed models to predict the pharmacokinetic properties of all compounds in the DrugBank. All models described in this paper have been integrated and made publicly available via the PhaKinPro Web-portal that can be accessed at https://phakinpro.mml.unc.edu/.

Merging Counter-Propagation and Back-Propagation Algorithms: Overcoming the Limitations of Counter-Propagation Neural Network Models

Artificial neural networks (ANNs) are nowadays applied as the most efficient methods in the majority of machine learning approaches, including data-driven modeling for assessment of the toxicity of chemicals. We developed a combined neural network methodology that can be used in the scope of new approach methodologies (NAMs) assessing chemical or drug toxicity. Here, we present QSAR models for predicting the physical and biochemical properties of molecules of three different datasets: aqueous solubility, acute fish toxicity toward fat head minnow, and bio-concentration factors. A novel neural network modeling method is developed by combining two neural network algorithms, namely, the counter-propagation modeling strategy (CP-ANN) with the back-propagation-of-errors algorithm (BPE-ANN). The advantage is a short training time, robustness, and good interpretability through the initial CP-ANN part, while the extension with BPE-ANN improves the precision of predictions in the range between minimal and maximal property values of the training data, regardless of the number of neurons in both neural networks, either CP-ANN or BPE-ANN.

Estrogen Receptor Alpha Binders for Hormone-Dependent Forms of Breast Cancer: e-QSAR and Molecular Docking Supported by X-ray Resolved Structures

Cancer, a life-disturbing and lethal disease with a high global impact, causes significant economic, social, and health challenges. Breast cancer refers to the abnormal growth of cells originating from breast tissues. Hormone-dependent forms of breast cancer, such as those influenced by estrogen, prompt the exploration of estrogen receptors as targets for potential therapeutic interventions. In this study, we conducted e-QSAR molecular docking and molecular dynamics analyses on a diverse set of inhibitors targeting estrogen receptor alpha (ER-α). The e-QSAR model is based on a genetic algorithm combined with multilinear regression analysis. The newly developed model possesses a balance between predictive accuracy and mechanistic insights adhering to the OECD guidelines. The e-QSAR model pointed out that sp2-hybridized carbon and nitrogen atoms are important atoms governing binding profiles. In addition, a specific combination of H-bond donors and acceptors with carbon, nitrogen, and ring sulfur atoms also plays a crucial role. The results are supported by molecular docking, MD simulations, and X-ray-resolved structures. The novel results could be useful for future drug development for ER-α.

Andrographolide: A Diterpenoid from Cymbopogon schoenanthus Identified as a New Hit Compound against Trypanosoma cruzi Using Machine Learning and Experimental Approaches

American Trypanosomiasis, also known as Chagas disease, is caused by the protozoan Trypanosoma cruzi and exhibits limited options for treatment. Natural products offer various structurally complex metabolites with biological activities, including those with anti-T. cruzi potential. The discovery and development of prototypes based on natural products frequently display multiple phases that could be facilitated by machine learning techniques to provide a fast and efficient method for selecting new hit candidates. Using Random Forest and k-Nearest Neighbors, two models were constructed to predict the biological activity of natural products from plants against intracellular amastigotes of T. cruzi. The diterpenoid andrographolide was identified from a virtual screening as a promising hit compound. Hereafter, it was isolated from Cymbopogon schoenanthus and chemically characterized by spectral data analysis. Andrographolide was evaluated against trypomastigote and amastigote forms of T. cruzi, showing IC50 values of 29.4 and 2.9 μM, respectively, while the standard drug benznidazole displayed IC50 values of 17.7 and 5.0 μM, respectively. Additionally, the isolated compound exhibited a reduced cytotoxicity (CC50 = 92.8 μM) against mammalian cells and afforded a selectivity index (SI) of 32, similar to that of benznidazole (SI = 39). From the in silico analyses, we can conclude that andrographolide fulfills many requirements implemented by DNDi to be a hit compound. Therefore, this work successfully obtained machine learning models capable of predicting the activity of compounds against intracellular forms of T. cruzi.

Generation of a drug-induced renal injury list to facilitate the development of new approach methodologies for nephrotoxicity

Drug-induced renal injury (DIRI) causes >1.5 million adverse events annually in the USA alone. Although standard biomarkers exist for DIRI, they lack the sensitivity or specificity to detect nephrotoxicity before the significant loss of renal function. In this study, we describe the creation of DIRIL - a list of drugs associated with DIRI and nephrotoxicity - from two literature datasets with DIRI annotation, confirmed using FDA drug labeling. DIRIL comprises 317 orally administered drugs covering all 14 anatomical, therapeutic and chemical (ATC) classification categories. Of the 317 drugs, 171 were DIRI-positive and 146 were DIRI-negative. DIRIL will be a relevant and invaluable resource for discovery of new approach methods (NAMs) to predict the occurrence and possible severity of DIRI earlier in drug development.

Predicting Cardiotoxicity of Molecules Using Attention-Based Graph Neural Networks

In drug discovery, the search for new and effective medications is often hindered by concerns about toxicity. Numerous promising molecules fail to pass the later phases of drug development due to strict toxicity assessments. This challenge significantly increases the cost, time, and human effort needed to discover new therapeutic molecules. Additionally, a considerable number of drugs already on the market have been withdrawn or re-evaluated because of their unwanted side effects. Among the various types of toxicity, drug-induced heart damage is a severe adverse effect commonly associated with several medications, especially those used in cancer treatments. Although a number of computational approaches have been proposed to identify the cardiotoxicity of molecules, the performance and interpretability of the existing approaches are limited. In our study, we proposed a more effective computational framework to predict the cardiotoxicity of molecules using an attention-based graph neural network. Experimental results indicated that the proposed framework outperformed the other methods. The stability of the model was also confirmed by our experiments. To assist researchers in evaluating the cardiotoxicity of molecules, we have developed an easy-to-use online web server that incorporates our model.

Unveiling the Antiviral Capabilities of Targeting Human Dihydroorotate Dehydrogenase against SARS-CoV-2

The urgent need for effective treatments against emerging viral diseases, driven by drug-resistant strains and new viral variants, remains critical. We focus on inhibiting the human dihydroorotate dehydrogenase (HsDHODH), one of the main enzymes responsible for pyrimidine nucleotide synthesis. This strategy could impede viral replication without provoking resistance. We evaluated naphthoquinone fragments, discovering potent HsDHODH inhibition with IC50 ranging from 48 to 684 nM, and promising in vitro anti-SARS-CoV-2 activity with EC50 ranging from 1.2 to 2.3 μM. These compounds exhibited low toxicity, indicating potential for further development. Additionally, we employed computational tools such as molecular docking and quantitative structure-activity relationship (QSAR) models to analyze protein-ligand interactions, revealing that these naphthoquinones exhibit a protein binding pattern similar to brequinar, a potent HsDHODH inhibitor. These findings represent a significant step forward in the search for effective antiviral treatments and have great potential to impact the development of new broad-spectrum antiviral drugs.

Development of a standardized methodology for transfer learning with QSAR models: a purely data-driven approach for source task selection

Transfer learning is a machine learning technique that works well with chemical endpoints, with several papers confirming its efficiency. Although effective, because the choice of source/assistant tasks is non-trivial, the application of this technique is severely limited by the domain knowledge of the modeller. Considering this limitation, we developed a purely data-driven approach for source task selection that abstracts the need for domain knowledge. To achieve this, we created a supervised learning setting in which transfer outcome (positive/negative) is the variable to be predicted, and a set of six transferability metrics, calculated based on information from target and source datasets, are the features for prediction. We used the ChEMBL database to generate 100,000 transfers using random pairing, and with these transfers, we trained and evaluated our transferability prediction model (TP-Model). Our TP-Model achieved a 135-fold increase in precision while achieving a sensitivity of 92%, demonstrating a clear superiority against random search. In addition, we observed that transfer learning could provide considerable performance increases when applicable, with an average Matthews Correlation Coefficient (MCC) increase of 0.19 when using a single source and an average MCC increase of 0.44 when using multiple sources.

ToxMPNN: A deep learning model for small molecule toxicity prediction

Machine learning (ML) has shown a great promise in predicting toxicity of small molecules. However, the availability of data for such predictions is often limited. Because of the unsatisfactory performance of models trained on a single toxicity endpoint, we collected toxic small molecules with multiple toxicity endpoints from previous study. The dataset comprises 27 toxic endpoints categorized into seven toxicity classes, namely, carcinogenicity and mutagenicity, acute oral toxicity, respiratory toxicity, irritation and corrosion, cardiotoxicity, CYP450, and endocrine disruption. In addition, a binary classification Common-Toxicity task was added based on the aforementioned dataset. To improve the performance of the models, we added marketed drugs as negative samples. This study presents a toxicity predictive model, ToxMPNN, based on the message passing neural network (MPNN) architecture, aiming to predict the toxicity of small molecules. The results demonstrate that ToxMPNN outperforms other models in capturing toxic features within the molecular structure, resulting in more precise predictions with the ROC_AUC testing score of 0.886 for the Toxicity_drug dataset. Furthermore, it was observed that adding marketed drugs as negative samples not only improves the predictive performance of the binary classification Common-Toxicity task but also enhances the stability of the model prediction. It shows that the graph-based deep learning (DL) algorithms in this study can be used as a trustworthy and effective tool to assess small molecule toxicity in the development of new drugs.

Mechanistic QSAR modeling derived virtual screening, drug repurposing, ADMET and in-vitro evaluation to identify anticancer lead as lysine-specific demethylase 5a inhibitor

A lysine-specific demethylase is an enzyme that selectively eliminates methyl groups from lysine residues. KDM5A, also known as JARID1A or RBP2, belongs to the KDM5 Jumonji histone demethylase subfamily. To identify novel molecules that interact with the LSD5A receptor, we created a quantitative structure-activity relationship (QSAR) model. A group of 435 compounds was used in a study of the quantitative relationship between structure and activity to guess the IC50 values for blocking LASD5A. We used a genetic algorithm-multilinear regression-based quantitative structure-activity connection model to forecast the bioactivity (PIC50) of 1615 food and drug administration pharmaceuticals from the zinc database with the goal of repurposing clinically used medications. We used molecular docking, molecular dynamic simulation modelling, and molecular mechanics generalised surface area analysis to investigate the molecule's binding mechanism. A genetic algorithm and multi-linear regression method were used to make six variable-based quantitative structure-activity relationship models that worked well (R2 = 0.8521, Q2LOO = 0.8438, and Q2LMO = 0.8414). ZINC000000538621 was found to be a new hit against LSD5A after a quantitative structure-activity relationship-based virtual screening of 1615 zinc food and drug administration compounds. The docking analysis revealed that the hit molecule 11 in the KDM5A binding pocket adopted a conformation similar to the pdb-6bh1 ligand (docking score: -8.61 kcal/mol). The results from molecular docking and the quantitative structure-activity relationship were complementary and consistent. The most active lead molecule 11, which has shown encouraging results, has good absorption, distribution, metabolism, and excretion (ADME) properties, and its toxicity has been shown to be minimal. In addition, the MTT assay of ZINC000000538621 with MCF-7 cell lines backs up the in silico studies. We used molecular mechanics generalise borne surface area analysis and a 200-ns molecular dynamics simulation to find structural motifs for KDM5A enzyme interactions. Thus, our strategy will likely expand food and drug administration molecule repurposing research to find better anticancer drugs and therapies.Communicated by Ramaswamy H. Sarma.

Free and open-source QSAR-ready workflow for automated standardization of chemical structures in support of QSAR modeling

The rapid increase of publicly available chemical structures and associated experimental data presents a valuable opportunity to build robust QSAR models for applications in different fields. However, the common concern is the quality of both the chemical structure information and associated experimental data. This is especially true when those data are collected from multiple sources as chemical substance mappings can contain many duplicate structures and molecular inconsistencies. Such issues can impact the resulting molecular descriptors and their mappings to experimental data and, subsequently, the quality of the derived models in terms of accuracy, repeatability, and reliability. Herein we describe the development of an automated workflow to standardize chemical structures according to a set of standard rules and generate two and/or three-dimensional "QSAR-ready" forms prior to the calculation of molecular descriptors. The workflow was designed in the KNIME workflow environment and consists of three high-level steps. First, a structure encoding is read, and then the resulting in-memory representation is cross-referenced with any existing identifiers for consistency. Finally, the structure is standardized using a series of operations including desalting, stripping of stereochemistry (for two-dimensional structures), standardization of tautomers and nitro groups, valence correction, neutralization when possible, and then removal of duplicates. This workflow was initially developed to support collaborative modeling QSAR projects to ensure consistency of the results from the different participants. It was then updated and generalized for other modeling applications. This included modification of the "QSAR-ready" workflow to generate "MS-ready structures" to support the generation of substance mappings and searches for software applications related to non-targeted analysis mass spectrometry. Both QSAR and MS-ready workflows are freely available in KNIME, via standalone versions on GitHub, and as docker container resources for the scientific community. Scientific contribution: This work pioneers an automated workflow in KNIME, systematically standardizing chemical structures to ensure their readiness for QSAR modeling and broader scientific applications. By addressing data quality concerns through desalting, stereochemistry stripping, and normalization, it optimizes molecular descriptors' accuracy and reliability. The freely available resources in KNIME, GitHub, and docker containers democratize access, benefiting collaborative research and advancing diverse modeling endeavors in chemistry and mass spectrometry.

Selective Bias Virtual Screening for Discovery of Promising Antimalarial Candidates targeting Plasmodium N-Myristoyltransferase

Malaria remains a significant public health challenge, with Plasmodium vivax being the species responsible for the most prevalent form of the disease. Given the limited therapeutic options available, the search for new antimalarials against P. vivax is urgent. This study aims to identify new inhibitors for P. vivax N-myristoyltransferase (PvNMT), an essential drug target against malaria. Through a validated virtual screening campaign, we prioritized 23 candidates for further testing. In the yeast NMT system, seven compounds exhibit a potential inhibitor phenotype. In vitro antimalarial phenotypic assays confirmed the activity of four candidates while demonstrating an absence of cytotoxicity. Enzymatic assays reveal LabMol-394 as the most promising inhibitor, displaying selectivity against the parasite and a strong correlation within the yeast system. Furthermore, molecular dynamics simulations shed some light into its binding mode. This study constitutes a substantial contribution to the exploration of a selective quinoline scaffold and provides valuable insights into the development of new antimalarial candidates.

AttentiveSkin: To Predict Skin Corrosion/Irritation Potentials of Chemicals via Explainable Machine Learning Methods

Skin Corrosion/Irritation (Corr./Irrit.) has long been a health hazard in the Globally Harmonized System (GHS). Several in silico models have been built to predict Skin Corr./Irrit. as an alternative to the increasingly restricted animal testing. However, current studies are limited by data amount/quality and model availability. To address these issues, we compiled a traceable consensus GHS data set comprising 731 Corr., 1283 Irrit., and 1205 negative (Neg.) samples from 6 governmental databases and 2 external data sets. Then, a series of binary classifiers were developed with five machine learning (ML) algorithms and six molecular representations. For 10-fold cross-validation, the best Corr. vs Neg. classifier achieved an Area Under the Receiver Operating Characteristic Curve (AUC) of 97.1%, while the best Irrit. vs Neg. classifier achieved an AUC of 84.7%. Compared with existing in silico tools on external validation, our Attentive FP classifiers showed the highest metrics on Corr. vs Neg. and the second highest accuracy on Irrit. vs Neg. The SHapley Additive exPlanation approach was further applied to figure out important molecular features, and the attention weights were visualized to perform interpretable prediction. Structural alerts associated with Skin Corr./Irrit. were also identified. The interpretable Attentive FP classifiers were integrated into the software AttentiveSkin at https://github.com/BeeBeeWong/AttentiveSkin. The conventional ML classifiers are also provided on our platform admetSAR at http://lmmd.ecust.edu.cn/admetsar2/. Considering the data deficiency and the limited model availability of Skin Corr./Irrit., we believe that our data set and models could facilitate chemical safety assessment and relevant studies.

Integrating QSAR modelling and deep learning in drug discovery: the emergence of deep QSAR

Quantitative structure-activity relationship (QSAR) modelling, an approach that was introduced 60 years ago, is widely used in computer-aided drug design. In recent years, progress in artificial intelligence techniques, such as deep learning, the rapid growth of databases of molecules for virtual screening and dramatic improvements in computational power have supported the emergence of a new field of QSAR applications that we term 'deep QSAR'. Marking a decade from the pioneering applications of deep QSAR to tasks involved in small-molecule drug discovery, we herein describe key advances in the field, including deep generative and reinforcement learning approaches in molecular design, deep learning models for synthetic planning and the application of deep QSAR models in structure-based virtual screening. We also reflect on the emergence of quantum computing, which promises to further accelerate deep QSAR applications and the need for open-source and democratized resources to support computer-aided drug design.

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.

Advancing Computational Toxicology by Interpretable Machine Learning

Chemical toxicity evaluations for drugs, consumer products, and environmental chemicals have a critical impact on human health. Traditional animal models to evaluate chemical toxicity are expensive, time-consuming, and often fail to detect toxicants in humans. Computational toxicology is a promising alternative approach that utilizes machine learning (ML) and deep learning (DL) techniques to predict the toxicity potentials of chemicals. Although the applications of ML- and DL-based computational models in chemical toxicity predictions are attractive, many toxicity models are "black boxes" in nature and difficult to interpret by toxicologists, which hampers the chemical risk assessments using these models. The recent progress of interpretable ML (IML) in the computer science field meets this urgent need to unveil the underlying toxicity mechanisms and elucidate the domain knowledge of toxicity models. In this review, we focused on the applications of IML in computational toxicology, including toxicity feature data, model interpretation methods, use of knowledge base frameworks in IML development, and recent applications. The challenges and future directions of IML modeling in toxicology are also discussed. We hope this review can encourage efforts in developing interpretable models with new IML algorithms that can assist new chemical assessments by illustrating toxicity mechanisms in humans.

A chemical structure and machine learning approach to assess the potential bioactivity of endogenous metabolites and their association with early-childhood hs-CRP levels

Metabolomics has gained much attraction due to its potential to reveal molecular disease mechanisms and present viable biomarkers. In this work we used a panel of untargeted serum metabolomes in 602 childhood patients of the COPSAC2010 mother-child cohort. The annotated part of the metabolome consists of 493 chemical compounds curated using automated procedures. Using predicted quantitative-structure-bioactivity relationships for the Tox21 database on nuclear receptors and stress response in cell lines, we created a filtering method for the vast number of quantified metabolites. The metabolites measured in children's serums used here have predicted potential against the chosen target modelled targets. The targets from Tox21 have been used with quantitative structure-activity relationships (QSARs) and were trained for ~7000 structures, saved as models, and then applied to 493 metabolites to predict their potential bioactivities. The models were selected based on strict accuracy criteria surpassing random effects. After application, 52 metabolites showed potential bioactivity based on structural similarity with known active compounds from the Tox21 set. The filtered compounds were subsequently used and weighted by their bioactive potential to show an association with early childhood hs-CRP levels at six months in a linear model supporting a physiological adverse effect on systemic low-grade inflammation. The significant metabolites were reported.

Integrated Molecular Modeling and Machine Learning for Drug Design

Modern therapeutic development often involves several stages that are interconnected, and multiple iterations are usually required to bring a new drug to the market. Computational approaches have increasingly become an indispensable part of helping reduce the time and cost of the research and development of new drugs. In this Perspective, we summarize our recent efforts on integrating molecular modeling and machine learning to develop computational tools for modulator design, including a pocket-guided rational design approach based on AlphaSpace to target protein-protein interactions, delta machine learning scoring functions for protein-ligand docking as well as virtual screening, and state-of-the-art deep learning models to predict calculated and experimental molecular properties based on molecular mechanics optimized geometries. Meanwhile, we discuss remaining challenges and promising directions for further development and use a retrospective example of FDA approved kinase inhibitor Erlotinib to demonstrate the use of these newly developed computational tools.

Artificial intelligence-guided approach for efficient virtual screening of hits against Schistosoma mansoni

Background: The impact of schistosomiasis, which affects over 230 million people, emphasizes the urgency of developing new antischistosomal drugs. Artificial intelligence is vital in accelerating the drug discovery process. Methodology & results: We developed classification and regression machine learning models to predict the schistosomicidal activity of compounds not experimentally tested. The prioritized compounds were tested on schistosomula and adult stages of Schistosoma mansoni. Four compounds demonstrated significant activity against schistosomula, with 50% effective concentration values ranging from 9.8 to 32.5 μM, while exhibiting no toxicity in animal and human cell lines. Conclusion: These findings represent a significant step forward in the discovery of antischistosomal drugs. Further optimization of these active compounds can pave the way for their progression into preclinical studies.

Collaborative analysis for drug discovery by federated learning on non-IID data

With the increasing availability of large-scale QSAR (Quantitative Structure-Activity Relationship) datasets, collaborative analysis has become a promising approach for drug discovery. Traditional centralized analysis which typically concentrates data on a central server for training faces challenges such as data privacy and security. Distributed analysis such as federated learning offers a solution by enabling collaborative model training without sharing raw data. However, it may fail when the training data in the local devices are non-independent and identically distributed (non-IID). In this paper, we propose a novel framework for collaborative drug discovery using federated learning on non-IID datasets. We address the difficulty of training on non-IID data by globally sharing a small subset of data among all institutions. Our framework allows multiple institutions to jointly train a robust predictive model while preserving the privacy of their individual data. We leverage the federated learning paradigm to distribute the model training process across local devices, eliminating the need for data exchange. The experimental results on 15 benchmark datasets demonstrate that the proposed method achieves competitive predictive accuracy to centralized analysis while respecting data privacy. Moreover, our framework offers benefits such as reduced data transmission and enhanced scalability, making it suitable for large-scale collaborative drug discovery efforts.

DeepAmes: A deep learning-powered Ames test predictive model with potential for regulatory application

The Ames assay is required by the regulatory agencies worldwide to assess the mutagenic potential risk of consumer products. As well as this in vitro assay, in silico approaches have been widely used to predict Ames test results as outlined in the International Council for Harmonization (ICH) guidelines. Building on this in silico approach, here we describe DeepAmes, a high performance and robust model developed with a novel deep learning (DL) approach for potential utility in regulatory science. DeepAmes was developed with a large and consistent Ames dataset (>10,000 compounds) and was compared with other five standard Machine Learning (ML) methods. Using a test set of 1,543 compounds, DeepAmes was the best performer in predicting the outcome of Ames assay. In addition, DeepAmes yielded the best and most stable performance up to when compounds were >30% outside of the applicability domain (AD). Regarding the potential for regulatory application, a revised version of DeepAmes with a much-improved sensitivity of 0.87 from 0.47. In conclusion, DeepAmes provides a DL-powered Ames test predictive model for predicting the results of Ames tests; with its defined AD and clear context of use, DeepAmes has potential for utility in regulatory application.

Lies and Liabilities: Computational Assessment of High-Throughput Screening Hits to Identify Artifact Compounds

Hits from high-throughput screening (HTS) of chemical libraries are often false positives due to their interference with assay detection technology. In response, we generated the largest publicly available library of chemical liabilities and developed "Liability Predictor," a free web tool to predict HTS artifacts. More specifically, we generated, curated, and integrated HTS data sets for thiol reactivity, redox activity, and luciferase (firefly and nano) activity and developed and validated quantitative structure-interference relationship (QSIR) models to predict these nuisance behaviors. The resulting models showed 58-78% external balanced accuracy for 256 external compounds per assay. QSIR models developed and validated herein identify nuisance compounds among experimental hits more reliably than do popular PAINS filters. Both the models and the curated data sets were implemented in "Liability Predictor," publicly available at https://liability.mml.unc.edu/. "Liability Predictor" may be used as part of chemical library design or for triaging HTS hits.

Current Perspectives in Antiviral Research

Studies on virus-host interactions are of high significance for a number of reasons [...].

Making the Case for Quantum Mechanics in Predictive Toxicology─Nearly 100 Years Too Late?

The use of quantum mechanics (QM) has long been the norm to study covalent-binding phenomena in chemistry and biochemistry. The pharmaceutical industry leverages QM models explicitly in covalent drug discovery and implicitly to characterize short-range interactions in noncovalent binding. Predictive toxicology has resisted widespread adoption of QM, including in the pharmaceutical industry, despite its obvious relevance to the metabolic processes in the upstream of adverse outcome pathways and advances in both QM methods and computational resources, which support fit-for-purpose applications in reasonable timeframes. Here, we make the case for embracing QM as an indispensable part of a toxicologist's toolkit. We argue that QM provides the necessary orthogonality to alert-based expert systems and traditional QSARs, consistent with calls for animal-free integrated testing strategies for safety assessments of commercial chemicals. We outline existing roadblocks to this transition, including the need to train model developers in QM and the shift toward service-based toxicity models that utilize high-performance computing clusters. Lastly, we describe recent examples of successful implementations of QM in hazard assessments and propose how in silico toxicology can be further advanced by integrating QM with artificial intelligence.

Enhancing developmental and reproductive toxicity knowledge: A new AOP stemming from glutathione depletion

Integrated approaches to testing and assessments (IATAs) have been proposed as a method to organise new approach methodologies in order to replace traditional animal testing for chemical safety assessments. To capture the mechanistic aspects of toxicity assessments, IATAs can be framed around the adverse outcome pathway (AOP) concept. To utilise AOPs fully in this context, a sufficient number of pathways need to be present to develop fit for purpose IATAs. In silico approaches can support IATA through the provision of predictive models and also through data integration to derive conclusions using a weight-of-evidence approach. To examine the maturity of a developmental and reproductive toxicity (DART) AOP network derived from the literature, an assessment of its coverage was performed against a novel toxicity dataset. A dataset of diverse compounds, with data from studies performed according to OECD test guidelines TG-421 and TG-422, was curated to test the performance of an in silico model based on the AOP network - allowing for the identification of knowledge gaps within the network. One such gap in the knowledge was filled through the development of an AOP stemming from the molecular initiating event 'glutathione reaction with an electrophile' leading to male fertility toxicity. The creation of the AOP provided the mechanistic rationale for the curation of pre-existing structural alerts to relevant key events. Integrating this new knowledge and associated alerts into the DART AOP network will improve its coverage of DART-relevant chemical space. In addition, broadening the coverage of AOPs for a particular regulatory endpoint may facilitate the development of, and confidence in, robust IATAs.

DIY Virtual Chemical Libraries - Novel Starting Points for Drug Discovery

The advancement of in silico technologies such as library enumeration and synthetic feasibility prediction has made drug discovery pipelines rely more and more on virtual libraries, which provide a significantly larger pool of compounds than in-stock supplier catalogs. Virtual libraries from external sources, however, may be associated with long delivery time and high cost. In this study, we present a Do-It-Yourself (DIY) combinatorial chemistry library containing over 14 million almost completely novel products built from 1000 low-cost building blocks based on robust reactions frequently applied at medicinal chemistry laboratories. The applicability of the DIY library for various drug discovery approaches is demonstrated by extensive physicochemical property, structural diversity profiling, and the generation of focused libraries. We found that internally built DIY chemical libraries present a viable alternative of external virtual catalogs by providing access to a large number of low-cost and quickly accessible potential chemical starting points for drug discovery.

Small molecule antiviral compound collection (SMACC): A comprehensive, highly curated database to support the discovery of broad-spectrum antiviral drug molecules

Diseases caused by new viruses cost thousands if not millions of human lives and trillions of dollars. We have identified, collected, curated, and integrated all chemogenomics data from ChEMBL for 13 emerging viruses that hold the greatest potential threat to global human health. By identifying and solving several challenges related to data annotation accuracy, we developed a highly curated and thoroughly annotated database of compounds tested in both phenotypic and target-based assays for these viruses that we dubbed SMACC (Small Molecule Antiviral Compound Collection). The pilot version of the SMACC database contains over 32,500 entries for 13 viruses. By analyzing data in SMACC, we have identified ∼50 compounds with polyviral inhibition profile, mostly covering flavi- and coronaviruses. The SMACC database may serve as a reference for virologists and medicinal chemists working on the development of novel BSA agents in preparation for future viral outbreaks. SMACC is publicly available at https://smacc.mml.unc.edu.

Mechanistic Task Groupings Enhance Multitask Deep Learning of Strain-Specific Ames Mutagenicity

The Ames test is a gold standard mutagenicity assay that utilizes various Salmonella typhimurium strains with and without S9 fraction to provide insights into the mechanisms by which a chemical can mutate DNA. Multitask deep learning is an ideal framework for developing QSAR models with multiple end points, such as the Ames test, as the joint training of multiple predictive tasks may synergistically improve the prediction accuracy of each task. This work investigated how toxicology domain knowledge can be used to handcraft task groupings that better guide the training of multitask neural networks compared to a naïve ungrouped multitask neural network developed on a complete set of tasks. Sixteen S. typhimurium ± S9 strain tasks were used to generate groupings based on mutagenic and metabolic mechanisms that were reflected in correlation data analyses. Both grouped and ungrouped multitask neural networks predicted the 16 strain tasks with a higher balanced accuracy compared with single task controls, with grouped multitask neural networks consistently featuring incremental increases in predictivity over the ungrouped approach. We conclude that the main variable driving these performance improvements is the general multitask effect with mechanistic task groupings acting as an enhancement step to further concentrate synergistic training signals united by a common biological mechanism. This approach enables incorporation of toxicology domain knowledge into multitask QSAR model development allowing for more transparent and accurate Ames mutagenicity prediction.

A Review on the Recent Applications of Deep Learning in Predictive Drug Toxicological Studies

Drug toxicity prediction is an important step in ensuring patient safety during drug design studies. While traditional preclinical studies have historically relied on animal models to evaluate toxicity, recent advances in deep-learning approaches have shown great promise in advancing drug safety science and reducing animal use in preclinical studies. However, deep-learning-based approaches also face challenges in handling large biological data sets, model interpretability, and regulatory acceptance. In this review, we provide an overview of recent developments in deep-learning-based approaches for predicting drug toxicity, highlighting their potential advantages over traditional methods and the need to address their limitations. Deep-learning models have demonstrated excellent performance in predicting toxicity outcomes from various data sources such as chemical structures, genomic data, and high-throughput screening assays. The potential of deep learning for automated feature engineering is also discussed. This review emphasizes the need to address ethical concerns related to the use of deep learning in drug toxicity studies, including the reduction of animal use and ensuring regulatory acceptance. Furthermore, emerging applications of deep learning in drug toxicity prediction, such as predicting drug-drug interactions and toxicity in rare subpopulations, are highlighted. The integration of deep-learning-based approaches with traditional methods is discussed as a way to develop more reliable and efficient predictive models for drug safety assessment, paving the way for safer and more effective drug discovery and development. Overall, this review highlights the critical role of deep learning in predictive toxicology and drug safety evaluation, emphasizing the need for continued research and development in this rapidly evolving field. By addressing the limitations of traditional methods, leveraging the potential of deep learning for automated feature engineering, and addressing ethical concerns, deep-learning-based approaches have the potential to revolutionize drug toxicity prediction and improve patient safety in drug discovery and development.

Multitask learning-driven identification of novel antitrypanosomal compounds

Background: Chagas disease and human African trypanosomiasis cause substantial death and morbidity, particularly in low- and middle-income countries, making the need for novel drugs urgent. Methodology & results: Therefore, an explainable multitask pipeline to profile the activity of compounds against three trypanosomes (Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense and Trypanosoma cruzi) were created. These models successfully discovered four new experimental hits (LC-3, LC-4, LC-6 and LC-15). Among them, LC-6 showed promising results, with IC50 values ranging 0.01-0.072 μM and selectivity indices >10,000. Conclusion: These results demonstrate that the multitask protocol offers predictivity and interpretability in the virtual screening of new antitrypanosomal compounds and has the potential to improve hit rates in Chagas and human African trypanosomiasis projects.

ProfhEX: AI-based platform for small molecules liability profiling

Off-target drug interactions are a major reason for candidate failure in the drug discovery process. Anticipating potential drug's adverse effects in the early stages is necessary to minimize health risks to patients, animal testing, and economical costs. With the constantly increasing size of virtual screening libraries, AI-driven methods can be exploited as first-tier screening tools to provide liability estimation for drug candidates. In this work we present ProfhEX, an AI-driven suite of 46 OECD-compliant machine learning models that can profile small molecules on 7 relevant liability groups: cardiovascular, central nervous system, gastrointestinal, endocrine, renal, pulmonary and immune system toxicities. Experimental affinity data was collected from public and commercial data sources. The entire chemical space comprised 289'202 activity data for a total of 210'116 unique compounds, spanning over 46 targets with dataset sizes ranging from 819 to 18896. Gradient boosting and random forest algorithms were initially employed and ensembled for the selection of a champion model. Models were validated according to the OECD principles, including robust internal (cross validation, bootstrap, y-scrambling) and external validation. Champion models achieved an average Pearson correlation coefficient of 0.84 (SD of 0.05), an R² determination coefficient of 0.68 (SD = 0.1) and a root mean squared error of 0.69 (SD of 0.08). All liability groups showed good hit-detection power with an average enrichment factor at 5% of 13.1 (SD of 4.5) and AUC of 0.92 (SD of 0.05). Benchmarking against already existing tools demonstrated the predictive power of ProfhEX models for large-scale liability profiling. This platform will be further expanded with the inclusion of new targets and through complementary modelling approaches, such as structure and pharmacophore-based models. ProfhEX is freely accessible at the following address: https://profhex.exscalate.eu/ .

Activity Models of Key GPCR Families in the Central Nervous System: A Tool for Many Purposes

G protein-coupled receptors (GPCRs) are targets of many drugs, of which ∼25% are indicated for central nervous system (CNS) disorders. Drug promiscuity affects their efficacy and safety profiles. Predicting the polypharmacology profile of compounds against GPCRs can thus provide a basis for producing more precise therapeutics by considering the targets and the anti-targets in that family of closely related proteins. We provide a tool for predicting the polypharmacology of compounds within prominent GPCR families in the CNS: serotonin, dopamine, histamine, muscarinic, opioid, and cannabinoid receptors. Our in-house algorithm, "iterative stochastic elimination" (ISE), produces high-quality ligand-based models for agonism and antagonism at 31 GPCRs. The ISE models correctly predict 68% of CNS drug-GPCR interactions, while the "similarity ensemble approach" predicts only 33%. The activity models correctly predict 56% of reported activities of DrugBank molecules for these CNS receptors. We conclude that the combination of interactions and activity profiles generated by screening through our models form the basis for subsequent designing and discovering novel therapeutics, either single, multitargeting, or repurposed.

Toxic External Exposure Leading to Ocular Surface Injury

The surface of the eye is directly exposed to the external environment, protected only by a thin tear film, and may therefore be damaged by contact with ambient particulate matter, liquids, aerosols, or vapors. In the workplace or home, the eye is subject to accidental or incidental exposure to cleaning products and pesticides. Organic matter may enter the eye and cause infection. Ocular surface damage can trigger a range of symptoms such as itch, discharge, hyperemia, photophobia, blurred vision, and foreign body sensation. Toxin exposure can be assessed clinically in multiple ways, including via measurement of tear production, slit-lamp examination, corneal staining, and conjunctival staining. At the cellular level, environmental toxins can cause oxidative damage, apoptosis of corneal and conjunctival cells, cell senescence, and impaired motility. Outcomes range from transient and reversible with complete healing to severe and sight-compromising structural changes. Classically, evaluation of tolerance and safety was carried out using live animal testing; however, new in vitro and computer-based, in silico modes are superseding the gold standard Draize test. This review examines how environmental features such as pollutants, temperature, and seasonality affect the ocular surface. Chemical burns to the eye are considered, and approaches to protect the ocular surface are detailed.

Transparency in Modeling through Careful Application of OECD's QSAR/QSPR Principles via a Curated Water Solubility Data Set

The need for careful assembly, training, and validation of quantitative structure-activity/property models (QSAR/QSPR) is more significant than ever as data sets become larger and sophisticated machine learning tools become increasingly ubiquitous and accessible to the scientific community. Regulatory agencies such as the United States Environmental Protection Agency must carefully scrutinize each aspect of a resulting QSAR/QSPR model to determine its potential use in environmental exposure and hazard assessment. Herein, we revisit the goals of the Organisation for Economic Cooperation and Development (OECD) in our application and discuss the validation principles for structure-activity models. We apply these principles to a model for predicting water solubility of organic compounds derived using random forest regression, a common machine learning approach in the QSA/PR literature. Using public sources, we carefully assembled and curated a data set consisting of 10,200 unique chemical structures with associated water solubility measurements. This data set was then used as a focal narrative to methodically consider the OECD's QSA/PR principles and how they can be applied to random forests. Despite some expert, mechanistically informed supervision of descriptor selection to enhance model interpretability, we achieved a model of water solubility with comparable performance to previously published models (5-fold cross validated performance 0.81 R2 and 0.98 RMSE). We hope this work will catalyze a necessary conversation around the importance of cautiously modernizing and explicitly leveraging OECD principles while pursuing state-of-the-art machine learning approaches to derive QSA/PR models suitable for regulatory consideration.

Comprehensive exploration of graphically defined reaction spaces

Existing reaction transition state (TS) databases are comparatively small and lack chemical diversity. Here, this data gap has been addressed using the concept of a graphically-defined model reaction to comprehensively characterize a reaction space associated with C, H, O, and N containing molecules with up to 10 heavy (non-hydrogen) atoms. The resulting dataset is composed of 176,992 organic reactions possessing at least one validated TS, activation energy, heat of reaction, reactant and product geometries, frequencies, and atom-mapping. For 33,032 reactions, more than one TS was discovered by conformational sampling, allowing conformational errors in TS prediction to be assessed. Data is supplied at the GFN2-xTB and B3LYP-D3/TZVP levels of theory. A subset of reactions were recalculated at the CCSD(T)-F12/cc-pVDZ-F12 and ωB97X-D2/def2-TZVP levels to establish relative errors. The resulting collection of reactions and properties are called the Reaction Graph Depth 1 (RGD1) dataset. RGD1 represents the largest and most chemically diverse TS dataset published to date and should find immediate use in developing novel machine learning models for predicting reaction properties.

Computer-Aided drug design of new 2-amino-thiophene derivatives as anti-leishmanial agents

The leishmaniasis is a neglected disease caused by a group of protozoan parasites from the genus Leishmania whose treatment is limited, obsolete, toxic, and ineffective in certain cases. These characteristics motivate researchers worldwide to plan new therapeutic alternatives for the treatment of leishmaniasis, where the use of cheminformatics tools applied to computer-assisted drug design has allowed research to make great advances in the search for new drugs candidates. In this study, a series of 2-amino-thiophene (2-AT) derivatives was screened virtually using QSAR tools, ADMET filters and prediction models, allowing direct the synthesis of compounds, which were evaluated in vitro against promastigotes and axenic amastigotes of Leishmania amazonensis. The combination of different descriptors and machine learning methods led to obtaining robust and predictive QSAR models, which was obtained from a dataset composed of 1862 compounds extracted from the ChEMBL database, with correct classification rates ranging from 0.53 (for amastigotes) to 0.91 (for promastigotes), allowing to select eleven 2-AT derivatives, which do not violate Lipinski's rules, exhibit good druglikeness, and with probability ≤70% of potential activity against the two evolutionary forms of the parasite. All compounds were properly synthesized and 8 of them were shown to be active at least against one of the evolutionary forms of the parasite with IC₅₀ values lower than 10 μM, being more active than the reference drug meglumine antimoniate, and showing low or no citotoxicity against macrophage J774.A1 for the most part. Compounds 8CN and DCN-83, respectively, are the most active against promastigote and amastigote forms, with IC₅₀ values of 1.20 and 0.71 μM, and selectivity indexes (SI) of 36.58 and 119.33. Structure Activity Relationship (SAR) study was carried out and allowed to identify some favorable and/or essential substitution patterns for the leishmanial activity of 2-AT derivatives. Taken together, these findings demonstrate that the use of ligand-based virtual screening proved to be quite effective and saved time, effort, and money in the selection of potential anti-leishmanial agents, and confirm, once again that 2-AT derivatives are promising hit compounds for the development of new anti-leishmanial agents.

Identification of a Family of Glycoside Derivatives Biologically Active against Acinetobacter baumannii and Other MDR Bacteria Using a QSPR Model

As the rate of discovery of new antibacterial compounds for multidrug-resistant bacteria is declining, there is an urge for the search for molecules that could revert this tendency. Acinetobacter baumannii has emerged as a highly virulent Gram-negative bacterium that has acquired multiple resistance mechanisms against antibiotics and is considered of critical priority. In this work, we developed a quantitative structure-property relationship (QSPR) model with 592 compounds for the identification of structural parameters related to their property as antibacterial agents against A. baumannii. QSPR mathematical validation (R2 = 70.27, RN = −0.008, a(R2) = 0.014, and δK = 0.021) and its prediction ability (Q2LMO = 67.89, Q2EXT = 67.75, a(Q2) = −0.068, δQ = 0.0, rm2¯ = 0.229, and Δrm2 = 0.522) were obtained with different statistical parameters; additional validation was done using three sets of external molecules (R2 = 72.89, 71.64 and 71.56). We used the QSPR model to perform a virtual screening on the BIOFACQUIM natural product database. From this screening, our model showed that molecules 32 to 35 and 54 to 68, isolated from different extracts of plants of the Ipomoea sp., are potential antibacterials against A. baumannii. Furthermore, biological assays showed that molecules 56 and 60 to 64 have a wide antibacterial activity against clinically isolated strains of A. baumannii, as well as other multidrug-resistant bacteria, including Staphylococcus aureus, Escherichia coli, Klebsiella pneumonia, and Pseudomonas aeruginosa. Finally, we propose 60 as a potential lead compound due to its broad-spectrum activity and its structural simplicity. Therefore, our QSPR model can be used as a tool for the investigation and search for new antibacterial compounds against A. baumannii.

Insight into potent TLR2 inhibitors for the treatment of disease caused by Mycoplasma pneumoniae based on machine learning approaches

Mycoplasma pneumoniae (MP) is one of the most common pathogens that causes acute respiratory tract infections. Children experiencing MP infection often suffer severe complications, lung injury, and even death. Previous studies have demonstrated that Toll-like receptor 2 (TLR2) is a potential therapeutic target for treating the MP-induced inflammatory response. However, the screening of natural compounds has received more attention for the treatment of bacterial infections to reduce the likelihood of bacterial resistance. Herein, we screened compounds by combining molecular docking and machine learning approaches to find potential lead compounds for treating MP infection. First, all compounds were docked with the TLR2 receptor protein to screen for potential candidates. To predict drug bioactivity, a machine learning model (random forest) was trained for TLR2 inhibitors to obtain the predictive model. The model achieved significant squared correlation coefficient (R²) values for the training set (0.85) and validation set (0.84) of compounds. The developed machine learning model was then used to predict the pIC50 values of the top 50 candidates from the Traditional Chinese compounds and Discovery Diversity sets of compounds. As a result, these compounds are capable of inhibiting the inflammatory response induced by MP. However, prior to bringing these compounds to market, it is necessary to verify these results with additional biological testing, including preclinical and clinical studies. Moreover, the present study provides a theoretical basis for the use of natural compounds as potential candidates to treat pneumonia caused by MP.