Integrated QSAR Models to Predict Acute Oral Systemic Toxicity

A deep learning based multi-model approach for predicting drug-like chemical compound's toxicity

Ensuring the safety and efficacy of chemical compounds is crucial in small-molecule drug development. In the later stages of drug development, toxic compounds pose a significant challenge, losing valuable resources and time. Early and accurate prediction of compound toxicity using deep learning models offers a promising solution to mitigate these risks during drug discovery. In this study, we present the development of several deep-learning models aimed at evaluating different types of compound toxicity, including acute toxicity, carcinogenicity, hERG_cardiotoxicity (the human ether-a-go-go related gene caused cardiotoxicity), hepatotoxicity, and mutagenicity. To address the inherent variations in data size, label type, and distribution across different types of toxicity, we employed diverse training strategies. Our first approach involved utilizing a graph convolutional network (GCN) regression model to predict acute toxicity, which achieved notable performance with Pearson R 0.76, 0.74, and 0.65 for intraperitoneal, intravenous, and oral administration routes, respectively. Furthermore, we trained multiple GCN binary classification models, each tailored to a specific type of toxicity. These models exhibited high area under the curve (AUC) scores, with an impressive AUC of 0.69, 0.77, 0.88, and 0.79 for predicting carcinogenicity, hERG_cardiotoxicity, mutagenicity, and hepatotoxicity, respectively. Additionally, we have used the approved drug dataset to determine the appropriate threshold value for the prediction score in model usage. We integrated these models into a virtual screening pipeline to assess their effectiveness in identifying potential low-toxicity drug candidates. Our findings indicate that this deep learning approach has the potential to significantly reduce the cost and risk associated with drug development by expediting the selection of compounds with low toxicity profiles. Therefore, the models developed in this study hold promise as critical tools for early drug candidate screening and selection.

Machine learning-assisted data filtering and QSAR models for prediction of chemical acute toxicity on rat and mouse

Machine learning (ML) methods provide a new opportunity to build quantitative structure-activity relationship (QSAR) models for predicting chemicals' toxicity based on large toxicity data sets, but they are limited in insufficient model robustness due to poor data set quality for chemicals with certain structures. To address this issue and improve model robustness, we built a large data set on rat oral acute toxicity for thousands of chemicals, then used ML to filter chemicals favorable for regression models (CFRM). In comparison to chemicals not favorable for regression models (CNRM), CFRM accounted for 67% of chemicals in the original data set, and had a higher structural similarity and a smaller toxicity distribution in 2-4 log₁₀ (mg/kg). The performance of established regression models for CFRM was greatly improved, with root-mean-square deviations (RMSE) in the range of 0.45-0.48 log₁₀ (mg/kg). Classification models were built for CNRM using all chemicals in the original data set, and the area under receiver operating characteristic (AUROC) reached 0.75-0.76. The proposed strategy was successfully applied to a mouse oral acute data set, yielding RMSE and AUROC in the range of 0.36-0.38 log₁₀ (mg/kg) and 0.79, respectively.

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

Toxicity prediction is a critical step in the drug discovery process that helps identify and prioritize compounds with the greatest potential for safe and effective use in humans, while also reducing the risk of costly late-stage failures. It is estimated that over 30% of drug candidates are discarded owing to toxicity. Recently, artificial intelligence (AI) has been used to improve drug toxicity prediction as it provides more accurate and efficient methods for identifying the potentially toxic effects of new compounds before they are tested in human clinical trials, thus saving time and money. In this review, we present an overview of recent advances in AI-based drug toxicity prediction, including the use of various machine learning algorithms and deep learning architectures, of six major toxicity properties and Tox21 assay end points. Additionally, we provide a list of public data sources and useful toxicity prediction tools for the research community and highlight the challenges that must be addressed to enhance model performance. Finally, we discuss future perspectives for AI-based drug toxicity prediction. This review can aid researchers in understanding toxicity prediction and pave the way for new methods of drug discovery.

In-silico studies to recognize repurposing therapeutics toward arginase-I inhibitors as a potential onco-immunomodulators

Rudolf Virchow was the first person to point out the important link between immune function and cancer. He did this by noticing that leukocytes were often found in tumors. Overexpression of arginase 1 (ARG1) and inducible nitric oxide synthase (iNOS) in myeloid-derived suppressor cells (MDSCs) and tumour-associated macrophages (TAMs) depletes both intracellular and extracellular arginine. TCR signalling is slowed as a result, and the same types of cells produce reactive oxygen and nitrogen species (ROS and RNS), which aggravates the situation. Human arginase I is a double-stranded manganese metalloenzyme that helps L-arginine break down into L-ornithine and urea. Thus, a quantitative structure-activity relationship (QSAR) analysis was performed to unearth the unrecognised structural aspects crucial for arginase-I inhibition. In this work, a balanced QSAR model with good prediction performance and clear mechanistic interpretation was developed using a dataset of 149 molecules encompassing a broad range of structural scaffolds and compositions. The model was made to meet OECD standards, and all of its validation parameters have values that are higher than the minimum requirements (R² _tr = 0.89, Q² _LMO = 0.86, and R² _ex = 0.85). The present QSAR study linked structural factors to arginase-I inhibitory action, including the proximity of lipophilic atoms to the molecule's centre of mass (within 3A), the position of the donor to the ring nitrogen (exactly 3 bonds away), and the surface area ratio. As OAT-1746 and two others are the only arginase-I inhibitors in development at the time, we have performed a QSAR-based virtual screening with 1650 FDA compounds taken from the zinc database. In this screening, 112 potential hit compounds were found to have a PIC50 value of less than 10 nm against the arginase-I receptor. The created QSAR model's application domain was evaluated in relation to the most active hit molecules identified using QSAR-based virtual screening, utilising a training set of 149 compounds and a prediction set of 112 hit molecules. As shown in the Williams plot, the top hit molecule, ZINC000252286875, has a low leverage value of HAT i/i h* = 0.140, placing it towards the boundary of the usable range. Furthermore, one of 112 hit molecules with a docking score of -10.891 kcal/mol (_PIC₅₀ = 10.023 M) was isolated from a study of arginase-I using molecular docking. Protonated ZINC000252286875-linked arginase-1 showed 2.9 RMSD, whereas non-protonated had 1.8. RMSD plots illustrate protein stability in protonated and non-protonated ZINC000252286875-bound states. Protonated-ZINC000252286875-bound proteins contain 25 Rg. The non-protonated protein-ligand combination exhibits a 25.2-Rg, indicating compactness. Protonated and non-protonated ZINC000252286875 stabilised protein targets in binding cavities posthumously. Significant root mean square fluctuations (RMSF) were seen in the arginase-1 protein at a small number of residues for a time function of 500 ns in both the protonated and unprotonated states. Protonated and non-protonated ligands interacted with proteins throughout the simulation. ZINC000252286875 bound Lys64, Asp124, Ala171, Arg222, Asp232, and Gly250. Aspartic acid residue 232 exhibited 200% ionic contact. 500-ns simulations-maintained ions. Salt bridges for ZINC000252286875 aided docking. ZINC000252286875 created six ionic bonds with Lys68, Asp117, His126, Ala171, Lys224, and Asp232 residues. Asp117, His126, and Lys224 showed 200% ionic interactions. In protonated and deprotonated states, GbindvdW, GbindLipo, and GbindCoulomb energies played crucial role. Moreover, ZINC000252286875 meets all of the ADMET standards to serve as a drug. As a result, the current analyses were successful in locating a novel and potent hit molecule that inhibits arginase-I effectively at nanomolar concentrations. The results of this investigation can be used to develop brand-new arginase I inhibitors as an alternative immune-modulating cancer therapy.

Principles and Procedures for Assessment of Acute Toxicity Incorporating In Silico Methods

Acute toxicity in silico models are being used to support an increasing number of application areas including (1) product research and development, (2) product approval and registration as well as (3) the transport, storage and handling of chemicals. The adoption of such models is being hindered, in part, because of a lack of guidance describing how to perform and document an in silico analysis. To address this issue, a framework for an acute toxicity hazard assessment is proposed. This framework combines results from different sources including in silico methods and in vitro or in vivo experiments. In silico methods that can assist the prediction of in vivo outcomes (i.e., LD50) are analyzed concluding that predictions obtained using in silico approaches are now well-suited for reliably supporting assessment of LD50-based acute toxicity for the purpose of GHS classification. A general overview is provided of the endpoints from in vitro studies commonly evaluated for predicting acute toxicity (e.g., cytotoxicity/cytolethality as well as assays targeting specific mechanisms). The increased understanding of pathways and key triggering mechanisms underlying toxicity and the increased availability of in vitro data allow for a shift away from assessments solely based on endpoints such as LD50, to mechanism-based endpoints that can be accurately assessed in vitro or by using in silico prediction models. This paper also highlights the importance of an expert review of all available information using weight-of-evidence considerations and illustrates, using a series of diverse practical use cases, how in silico approaches support the assessment of acute toxicity.

Hybrid consensus and k-nearest neighbours (kNN) strategies to classify dual BRD4/PLK1 inhibitors

A novel decision-making procedure is proposed here for the first time to identify active/inactive and selective/non-selective dual inhibitors using consensus approaches and pools of k-nearest neighbours (kNN) classifications instead of individual models. Dual BRD4/PLK1 inhibition with adequate selectivity is a potential therapeutic strategy for targeting tumour cells in high-risk patients. We report the unique way to identify both active and selective dual BRD4/PLK1 inhibitors using consensus and kNN strategies together with two sources of receptor-based and ligand-based information which are the ranked binding energies of residues and important molecular features, respectively. The results of consensus approaches were compared with the results of individual kNN models. The chemical space similarity was measured using three different distance functions to increase the reliability. All activity and selectivity classification models were validated using cross-validation and y-randomization tests. The outcomes show that consensus approaches can increase the reliability and accuracy of active/inactive or selective/non-selective detections up to 90%. Consensus approaches also reached more balanced values of sensitivity and specificity compared to the individual kNN models because of the compensation in the integration of diverse sources of information.

Integrating concept of pharmacophore with graph neural networks for chemical property prediction and interpretation

Recently, graph neural networks (GNNs) have revolutionized the field of chemical property prediction and achieved state-of-the-art results on benchmark data sets. Compared with the traditional descriptor- and fingerprint-based QSAR models, GNNs can learn task related representations, which completely gets rid of the rules defined by experts. However, due to the lack of useful prior knowledge, the prediction performance and interpretability of the GNNs may be affected. In this study, we introduced a new GNN model called RG-MPNN for chemical property prediction that integrated pharmacophore information hierarchically into message-passing neural network (MPNN) architecture, specifically, in the way of pharmacophore-based reduced-graph (RG) pooling. RG-MPNN absorbed not only the information of atoms and bonds from the atom-level message-passing phase, but also the information of pharmacophores from the RG-level message-passing phase. Our experimental results on eleven benchmark and ten kinase data sets showed that our model consistently matched or outperformed other existing GNN models. Furthermore, we demonstrated that applying pharmacophore-based RG pooling to MPNN architecture can generally help GNN models improve the predictive power. The cluster analysis of RG-MPNN representations and the importance analysis of pharmacophore nodes will help chemists gain insights for hit discovery and lead optimization.

Mapping Mechanistic Pathways of Acute Oral Systemic Toxicity Using Chemical Structure and Bioactivity Measurements

Regulatory agencies around the world have committed to reducing or eliminating animal testing for establishing chemical safety. Adverse outcome pathways can facilitate replacement by providing a mechanistic framework for identifying the appropriate non-animal methods and connecting them to apical adverse outcomes. This study separated 11,992 chemicals with curated rat oral acute toxicity information into clusters of structurally similar compounds. Each cluster was then assigned one or more ToxCast/Tox21 assays by looking for the minimum number of assays required to record at least one positive hit call below cytotoxicity for all acutely toxic chemicals in the cluster. When structural information is used to select assays for testing, none of the chemicals required more than four assays and 98% required two assays or less. Both the structure-based clusters and activity from the associated assays were significantly associated with the GHS toxicity classification of the chemicals, which suggests that a combination of bioactivity and structural information could be as reproducible as traditional in vivo studies. Predictivity is improved when the in vitro assay directly corresponds to the mechanism of toxicity, but many indirect assays showed promise as well. Given the lower cost of in vitro testing, a small assay battery including both general cytotoxicity assays and two or more orthogonal assays targeting the toxicological mechanism could be used to improve performance further. This approach illustrates the promise of combining existing in silico approaches, such as the Collaborative Acute Toxicity Modeling Suite (CATMoS), with structure-based bioactivity information as part of an efficient tiered testing strategy that can reduce or eliminate animal testing for acute oral toxicity.

In Silico Models for Predicting Acute Systemic Toxicity

In this chapter, we give a brief overview of the regulatory requirements for acute systemic toxicity information in the European Union, and we review structure-based computational models that are available and potentially useful in the assessment of acute systemic toxicity. Emphasis is placed on quantitative structure-activity relationship (QSAR) models implemented by means of a range of software tools. The most recently published literature models for acute systemic toxicity are also discussed, and perspectives for future developments in this field are offered.

Machine learning models for classification tasks related to drug safety

In this review, we outline the current trends in the field of machine learning-driven classification studies related to ADME (absorption, distribution, metabolism and excretion) and toxicity endpoints from the past six years (2015-2021). The study focuses only on classification models with large datasets (i.e. more than a thousand compounds). A comprehensive literature search and meta-analysis was carried out for nine different targets: hERG-mediated cardiotoxicity, blood-brain barrier penetration, permeability glycoprotein (P-gp) substrate/inhibitor, cytochrome P450 enzyme family, acute oral toxicity, mutagenicity, carcinogenicity, respiratory toxicity and irritation/corrosion. The comparison of the best classification models was targeted to reveal the differences between machine learning algorithms and modeling types, endpoint-specific performances, dataset sizes and the different validation protocols. Based on the evaluation of the data, we can say that tree-based algorithms are (still) dominating the field, with consensus modeling being an increasing trend in drug safety predictions. Although one can already find classification models with great performances to hERG-mediated cardiotoxicity and the isoenzymes of the cytochrome P450 enzyme family, these targets are still central to ADMET-related research efforts.

On the use of electronegativity and electron affinity based pseudo-molecular field descriptors in developing correlations for quantitative structure-activity relationship modeling of drug activities

For quantitative structure-activity relationship (QSAR) modeling in ligand-based drug discovery programs, pseudo-molecular field (PMF) descriptors using intrinsic atomic properties, namely, electronegativity and electron affinity are studied. In combination with partial least squares analysis and Procrustes transformation, these PMF descriptors were employed successfully to develop correlations that predict the activities of target protein inhibitors involved in various diseases (cancer, neurodegenerative disorders, HIV, and malaria). The results show that the present QSAR approach is competitive to existing QSAR models. In order to demonstrate the use of this algorithm, we present results of screening naturally occurring molecules with unknown bioactivities. The pIC₅₀ predictions can screen molecules that have desirable activity before assessment by docking studies.

CATMoS: Collaborative Acute Toxicity Modeling Suite

BACKGROUND

Humans are exposed to tens of thousands of chemical substances that need to be assessed for their potential toxicity. Acute systemic toxicity testing serves as the basis for regulatory hazard classification, labeling, and risk management. However, it is cost- and time-prohibitive to evaluate all new and existing chemicals using traditional rodent acute toxicity tests. In silico models built using existing data facilitate rapid acute toxicity predictions without using animals.

OBJECTIVES

The U.S. Interagency Coordinating Committee on the Validation of Alternative Methods (ICCVAM) Acute Toxicity Workgroup organized an international collaboration to develop in silico models for predicting acute oral toxicity based on five different end points: Lethal Dose 50 (LD50 value, U.S. Environmental Protection Agency hazard (four) categories, Globally Harmonized System for Classification and Labeling hazard (five) categories, very toxic chemicals [LD50 (LD50≤50mg/kg)], and nontoxic chemicals (LD50>2,000mg/kg).

METHODS

An acute oral toxicity data inventory for 11,992 chemicals was compiled, split into training and evaluation sets, and made available to 35 participating international research groups that submitted a total of 139 predictive models. Predictions that fell within the applicability domains of the submitted models were evaluated using external validation sets. These were then combined into consensus models to leverage strengths of individual approaches.

RESULTS

The resulting consensus predictions, which leverage the collective strengths of each individual model, form the Collaborative Acute Toxicity Modeling Suite (CATMoS). CATMoS demonstrated high performance in terms of accuracy and robustness when compared with in vivo results.

DISCUSSION

CATMoS is being evaluated by regulatory agencies for its utility and applicability as a potential replacement for in vivo rat acute oral toxicity studies. CATMoS predictions for more than 800,000 chemicals have been made available via the National Toxicology Program's Integrated Chemical Environment tools and data sets (ice.ntp.niehs.nih.gov). The models are also implemented in a free, standalone, open-source tool, OPERA, which allows predictions of new and untested chemicals to be made. https://doi.org/10.1289/EHP8495.

Predicting the reproductive toxicity of chemicals using ensemble learning methods and molecular fingerprints

Reproductive toxicity endpoints are a significant safety concern in the assessment of the adverse effects of chemicals in drug discovery. Computational models that can accurately predict a chemical's toxic potential are increasingly pursued to replace traditional animal experiments. Thus, ensemble learning models were built to predict the reproductive toxicity of compounds. Our ensemble models were developed using support vector machine, random forest, and extreme gradient boosting methods and 9 molecular fingerprints calculated for a dataset containing 1823 chemicals. The best prediction performance was achieved by the Ensemble-Top12 model, with an accuracy (ACC) of 86.33 %, a sensitivity (SEN) of 82.02 %, a specificity (SPE) of 90.19 %, and an area under the receiver operating characteristic curve (AUC) of 0.937 in 5-fold cross-validation and ACC, SEN, SPE, and AUC values of 84.38 %, 86.90 %, 90.67 %, and 0.920, respectively, in external validation. We also defined the applicability domain (AD) of the ensemble model by calculating the Tanimoto distance of the training set. Compared with models in existing literature, our ensemble model achieves relatively high ACC, SPE and AUC values. We also identified several fingerprint features related to chemical reproductive toxicity. Considering the performance of model, we recommend using the Ensemble-Top12 model to predict reproductive toxicity in early drug development.

An Analysis of QSAR Research Based on Machine Learning Concepts

Quantitative Structure-Activity Relationship (QSAR) is a popular approach developed to correlate chemical molecules with their biological activities based on their chemical structures. Machine learning techniques have proved to be promising solutions to QSAR modeling. Due to the significant role of machine learning strategies in QSAR modeling, this area of research has attracted much attention from researchers. A considerable amount of literature has been published on machine learning based QSAR modeling methodologies whilst this domain still suffers from lack of a recent and comprehensive analysis of these algorithms. This study systematically reviews the application of machine learning algorithms in QSAR, aiming to provide an analytical framework. For this purpose, we present a framework called 'ML-QSAR'. This framework has been designed for future research to: a) facilitate the selection of proper strategies among existing algorithms according to the application area requirements, b) help to develop and ameliorate current methods and c) providing a platform to study existing methodologies comparatively. In ML-QSAR, first a structured categorization is depicted which studied the QSAR modeling research based on machine models. Then several criteria are introduced in order to assess the models. Finally, inspired by aforementioned criteria the qualitative analysis is carried out.

Consensus versus Individual QSARs in Classification: Comparison on a Large-Scale Case Study

Consensus strategies have been widely applied in many different scientific fields, based on the assumption that the fusion of several sources of information increases the outcome reliability. Despite the widespread application of consensus approaches, their advantages in quantitative structure–activity relationship (QSAR) modeling have not been thoroughly evaluated, mainly due to the lack of appropriate large-scale data sets. In this study, we evaluated the advantages and drawbacks of consensus approaches compared to single classification QSAR models. To this end, we used a data set of three properties (androgen receptor binding, agonism, and antagonism) for approximately 4000 molecules with predictions performed by more than 20 QSAR models, made available in a large-scale collaborative project. The individual QSAR models were compared with two consensus approaches, majority voting and the Bayes consensus with discrete probability distributions, in both protective and nonprotective forms. Consensus strategies proved to be more accurate and to better cover the analyzed chemical space than individual QSARs on average, thus motivating their widespread application for property prediction. Scripts and data to reproduce the results of this study are available for download.

SAR and QSAR modeling of a large collection of LD50 rat acute oral toxicity data

The median lethal dose for rodent oral acute toxicity (LD50) is a standard piece of information required to categorize chemicals in terms of the potential hazard posed to human health after acute exposure. The exclusive use of in vivo testing is limited by the time and costs required for performing experiments and by the need to sacrifice a number of animals. (Quantitative) structure-activity relationships [(Q)SAR] proved a valid alternative to reduce and assist in vivo assays for assessing acute toxicological hazard. In the framework of a new international collaborative project, the NTP Interagency Center for the Evaluation of Alternative Toxicological Methods and the U.S. Environmental Protection Agency's National Center for Computational Toxicology compiled a large database of rat acute oral LD50 data, with the aim of supporting the development of new computational models for predicting five regulatory relevant acute toxicity endpoints. In this article, a series of regression and classification computational models were developed by employing different statistical and knowledge-based methodologies. External validation was performed to demonstrate the real-life predictability of models. Integrated modeling was then applied to improve performance of single models. Statistical results confirmed the relevance of developed models in regulatory frameworks, and confirmed the effectiveness of integrated modeling. The best integrated strategies reached RMSEs lower than 0.50 and the best classification models reached balanced accuracies over 0.70 for multi-class and over 0.80 for binary endpoints. Computed predictions will be hosted on the EPA's Chemistry Dashboard and made freely available to the scientific community.

QSAR investigation of acute toxicity of organic compounds during oral administration to mice

The effect of the structure of organic compounds on the acute toxicity upon oral injection in mice was studied using 2D simplex representation of the molecular structure and Random forest (RF) methods. Satisfactory quantitative structure-activity relationship (QSAR) models were constructed (R2 test = 0,61–0,62). The interpretation of the obtained QSAR models was carried out. The contributions of known toxicophores with established mechanisms of action were calculated in order to confirm the ability of the interpretation approach to correctly rank them relative to other structural fragments. The influence of the molecular surroundings of some toxicophores was analyzed. We analyzed the contributions of other highly ranked fragments from the list of common functional groups and ring systems in order to find new potential toxicophores. The on-line version of the expert system “OCHEM” (https://ochem.eu) and Arithmetic Mean Toxicity (AMT) approach were used for a comparative QSAR study.