MolToxPred: small molecule toxicity prediction using machine learning approach

Different types of chemicals and products may exhibit various health risks when administered into the human body. For toxicity reasons, the number of new drugs entering the market through the conventional drug development process has been reduced over the years. However, with the advent of big data and artificial intelligence, machine learning techniques have emerged as a potential solution for predicting toxicity and ensuring efficient drug development and chemical safety. An ML model for toxicity prediction can reduce experimental costs and time while addressing ethical concerns by drastically reducing the need for animals and clinical trials. Herein, MolToxPred, an ML-based tool, has been developed using a stacked model approach to predict the potential toxicity of small molecules and metabolites. The stacked model consists of random forest, multi-layer perceptron, and LightGBM as base classifiers and Logistic Regression as the meta classifier. For training and validation purposes, a comprehensive set of toxic and non-toxic molecules is curated. Different structural and physicochemical-based features in the form of molecular descriptors and fingerprints were employed. MolToxPred utilizes a comprehensive feature selection process and optimizes its hyperparameters through Bayesian optimization with stratified 5-fold cross-validation. In the evaluation phase, MolToxPred achieved an AUROC of 87.76% on the test set and 88.84% on an external validation set. The McNemar test was used as the post-hoc test to determine if the stacked models' performance was significantly different compared to the base learners. The developed stacked model outperformed its base classifiers and an existing tool in the literature, reaffirming its better performance. The hypothesis is that the incorporation of a diverse set of data, the subsequent feature selection, and a stacked ensemble approach give MolToxPred the edge over other methods. In addition to this, an attempt has been made to identify structural alerts responsible for endpoints of the Tox21 data to determine the association of a molecule with a plausible downstream pathway of action. MolToxPred may be helpful for drug discovery and regulatory pipelines in pharmaceutical and other industries for in silico toxicity prediction of small molecule candidates.

GeoDILI: A Robust and Interpretable Model for Drug-Induced Liver Injury Prediction Using Graph Neural Network-Based Molecular Geometric Representation

Drug-induced liver injury (DILI) is a significant cause of drug failure and withdrawal due to liver damage. Accurate prediction of hepatotoxic compounds is crucial for safe drug development. Several DILI prediction models have been published, but they are built on different data sets, making it difficult to compare model performance. Moreover, most existing models are based on molecular fingerprints or descriptors, neglecting molecular geometric properties and lacking interpretability. To address these limitations, we developed GeoDILI, an interpretable graph neural network that uses a molecular geometric representation. First, we utilized a geometry-based pretrained molecular representation and optimized it on the DILI data set to improve predictive performance. Second, we leveraged gradient information to obtain high-precision atomic-level weights and deduce the dominant substructure. We benchmarked GeoDILI against recently published DILI prediction models, as well as popular GNN models and fingerprint-based machine learning models using the same data set, showing superior predictive performance of our proposed model. We applied the interpretable method in the DILI data set and derived seven precise and mechanistically elucidated structural alerts. Overall, GeoDILI provides a promising approach for accurate and interpretable DILI prediction with potential applications in drug discovery and safety assessment. The data and source code are available at GitHub repository (https://github.com/CSU-QJY/GeoDILI).

cardioToxCSM: A Web Server for Predicting Cardiotoxicity of Small Molecules

The design of novel, safe, and effective drugs to treat human diseases is a challenging venture, with toxicity being one of the main sources of attrition at later stages of development. Failure due to toxicity incurs a significant increase in costs and time to market, with multiple drugs being withdrawn from the market due to their adverse effects. Cardiotoxicity, for instance, was responsible for the failure of drugs such as fenspiride, propoxyphene, and valdecoxib. While significant effort has been dedicated to mitigate this issue by developing computational approaches that aim to identify molecules likely to be toxic, including quantitative structure-activity relationship models and machine learning methods, current approaches present limited performance and interpretability. To overcome these, we propose a new web-based computational method, cardioToxCSM, which can predict six types of cardiac toxicity outcomes, including arrhythmia, cardiac failure, heart block, hERG toxicity, hypertension, and myocardial infarction, efficiently and accurately. cardioToxCSM was developed using the concept of graph-based signatures, molecular descriptors, toxicophore matchings, and molecular fingerprints, leveraging explainable machine learning, and was validated internally via different cross validation schemes and externally via low-redundancy blind sets. The models presented robust performances with areas under ROC curves of up to 0.898 on 5-fold cross-validation, consistent with metrics on blind tests. Additionally, our models provide interpretation of the predictions by identifying whether substructures that are commonly enriched in toxic compounds were present. We believe cardioToxCSM will provide valuable insight into the potential cardiotoxicity of small molecules early on drug screening efforts. The method is made freely available as a web server at https://biosig.lab.uq.edu.au/cardiotoxcsm.

SApredictor: An Expert System for Screening Chemicals Against Structural Alerts

The rapid and accurate evaluation of chemical toxicity is of great significance for estimation of chemical safety. In the past decades, a great number of excellent computational models have been developed for chemical toxicity prediction. But most machine learning models tend to be “black box”, which bring about poor interpretability. In the present study, we focused on the identification and collection of structural alerts (SAs) responsible for a series of important toxicity endpoints. Then, we carried out effective storage of these structural alerts and developed a web-server named SApredictor (www.sapredictor.cn) for screening chemicals against structural alerts. People can quickly estimate the toxicity of chemicals with SApredictor, and the specific key substructures which cause the chemical toxicity will be intuitively displayed to provide valuable information for the structural optimization by medicinal chemists.

Exploration of structural requirements for azole chemicals towards human aromatase CYP19A1 activity: Classification modeling, structure-activity relationships and read-across study

Human aromatase, also called CYP19A1, plays a major role in the conversion of androgens into estrogens. Inhibition of aromatase is an important target for estrogen receptor (ER)-responsive breast cancer therapy. Use of azole compounds as aromatase inhibitors is widespread despite their low selectivity. A toxicological evaluation of commonly used azole-based drugs and agrochemicals with respect to CYP19A1is currently requested by the European Union- Registration, Evaluation, Authorization and Restriction of Chemicals (EU-REACH) regulations due to their potential as endocrine disruptors. In this connection, identification of structural alerts (SAs) is an effective strategy for the toxicological assessment and safe drug design. The present study describes the identification of SAs of azole-based chemicals as guiding experts to predict the aromatase activity. Total 21 SAs associated with aromatase activity were extracted from dataset of 326 azole-based drugs/chemicals obtained from Tox21 library. A cross-validated classification model having high accuracy (error rate 5%) was proposed which can precisely classify azole chemicals into active/inactive toward aromatase. In addition, mechanistic details and toxicological properties (agonism/antagonism) of azoles with respect to aromatase were explored by comparing active and inactive chemicals using structure-activity relationships (SAR). Lastly, few structural alerts were applied to form chemical categories for read-across applications.

In Silico Prediction and Insights Into the Structural Basis of Drug Induced Nephrotoxicity

Drug induced nephrotoxicity is a major clinical challenge, and it is always associated with higher costs for the pharmaceutical industry and due to detection during the late stages of drug development. It is desirable for improving the health outcomes for patients to distinguish nephrotoxic structures at an early stage of drug development. In this study, we focused on in silico prediction and insights into the structural basis of drug induced nephrotoxicity, based on reliable data on human nephrotoxicity. We collected 565 diverse chemical structures, including 287 nephrotoxic drugs on humans in the real world, and 278 non-nephrotoxic approved drugs. Several different machine learning and deep learning algorithms were employed for in silico model building. Then, a consensus model was developed based on three best individual models (RFR_QNPR, XGBOOST_QNPR, and CNF). The consensus model performed much better than individual models on internal validation and it achieved prediction accuracy of 86.24% external validation. The results of analysis of molecular properties differences between nephrotoxic and non-nephrotoxic structures indicated that several key molecular properties differ significantly, including molecular weight (MW), molecular polar surface area (MPSA), AlogP, number of hydrogen bond acceptors (nHBA), molecular solubility (LogS), the number of rotatable bonds (nRotB), and the number of aromatic rings (nAR). These molecular properties may be able to play an important part in the identification of nephrotoxic chemicals. Finally, 87 structural alerts for chemical nephrotoxicity were mined with f-score and positive rate analysis of substructures from Klekota-Roth fingerprint (KRFP). These structural alerts can well identify nephrotoxic drug structures in the data set. The in silico models and the structural alerts could be freely accessed via https://ochem.eu/article/140251 and http://www.sapredictor.cn, respectively. We hope the results should provide useful tools for early nephrotoxicity estimation in drug development.

Reactive metabolites of the anticonvulsant drugs and approaches to minimize the adverse drug reaction

Several generations of antiepileptic drugs (AEDs) are available in the market for the treatment of seizures, but these are amalgamated with acute to chronic side effects. The most common side effects of AEDs are dose-related, but some are idiosyncratic adverse drug reactions (ADRs) that transpire due to the formation of reactive metabolite (RM) after the bioactivation process. Because of the adverse reactions patients usually discontinue the medication in between the treatment. The AEDs such as valproic acid, lamotrigine, phenytoin etc., can be categorized under such types because they form the RM which may prevail with life-threatening adverse effects or immune-mediated reactions. Hepatotoxicity, teratogenicity, cutaneous hypersensitivity, dizziness, addiction, serum sickness reaction, renal calculi, metabolic acidosis are associated with the metabolites of drugs such as arene oxide, N-desmethyldiazepam, 2-(1-hydroxyethyl)-2-methylsuccinimide, 2-(sulphamoy1acetyl)-phenol, E-2-en-VPA and 4-en-VPA and carbamazepine-10,11-epoxide, etc. The major toxicities are associated with the moieties that are either capable of forming RM or the functional groups may itself be too reactive prior to the metabolism. These functional groups or fragment structures are typically known as structural alerts or toxicophores. Therefore, minimizing the bioactivation potential of lead structures in the early phases of drug discovery by a modification to low-risk drug molecules is a priority for the pharmaceutical companies. Additionally, excellent potency and pharmacokinetic (PK) behaviour help in ensuring that appropriate (low dose) candidate drugs progress into the development phase. The current review discusses about RMs in the anticonvulsant drugs along with their mechanism vis-a-vis research efforts that have been taken to minimize the toxic effects of AEDs therapy.

Discriminant models on mitochondrial toxicity improved by consensus modeling and resolving imbalance in training

Humans and animals may be exposed to tens of thousands of natural and synthetic chemicals during their lifespan. It is difficult to assess risk for all the chemicals with experimental toxicity tests. An alternative approach is to use computational toxicology methods such as quantitative structure-activity relationship (QSAR) modeling. Mitochondrial toxicity is involved in many diseases such as cancer, neurodegeneration, type 2 diabetes, cardiovascular diseases and autoimmune diseases. Thus, it is important to rapidly and efficiently identify chemicals with mitochondrial toxicity. In this study, five machine learning algorithms and twelve types of molecular fingerprints were employed to generate QSAR discriminant models for mitochondrial toxicity. A threshold moving method was adopted to resolve the imbalance issue in the training data. Consensus of the models by an averaging probability strategy improved prediction performance. The best model has correct classification rates of 81.8% and 88.3% in ten-fold cross validation and external validation, respectively. Substructures such as phenol, carboxylic acid, nitro and arylchloride were found informative through analysis of information gain and frequency of substructures. The results demonstrate that resolving imbalance in training and building consensus models can improve classification rates for mitochondrial toxicity prediction.

Application of the hard and soft, acids and bases (HSAB) theory as a method to predict cumulative neurotoxicity

Xenobiotic electrophiles can form covalent adducts that may impair protein function, damage DNA, and may lead a range of adverse effects. Cumulative neurotoxicity is one adverse effect that has been linked to covalent protein binding as a Molecular Initiating Event (MIE). This paper describes a mechanistic in silico chemical screening approach for neurotoxicity based on Hard and Soft Acids and Bases (HSAB) theory. We evaluated the applicability of HSAB-based electrophilicity screening protocol for neurotoxicity on 19 positive and 19 negative reference chemicals. These reference chemicals were identified from the literature, using available information on mechanisms of neurotoxicity whenever possible. In silico screening was based on structural alerts for protein binding motifs and electrophilicity index in the range of known neurotoxicants. The approach demonstrated both a high positive prediction rate (82-90 %) and specificity (90 %). The overall sensitivity was relatively lower (47 %). However, when predicting the toxicity of chemicals known or suspected of acting via non-specific adduct formation mechanism, the HSAB approach identified 7/8 (sensitivity 88 %) of positive control chemicals correctly. Consequently, the HSAB-based screening is a promising approach of identifying possible neurotoxins with adduct formation molecular initiating events. While the approach must be expanded over time to capture a wider range of MIEs involved in neurotoxicity, the mechanistic nature of the screen allows users to flag chemicals for possible adduct formation MIEs. Thus, the HSAB based toxicity screening is a promising strategy for toxicity assessment and chemical prioritization in neurotoxicology and other health endpoints that involve adduct formation.