In silico prediction of hERG blockers using machine learning and deep learning approaches

CardioGenAI: a machine learning-based framework for re-engineering drugs for reduced hERG liability

The link between in vitro hERG ion channel inhibition and subsequent in vivo QT interval prolongation, a critical risk factor for the development of arrythmias such as Torsade de Pointes, is so well established that in vitro hERG activity alone is often sufficient to end the development of an otherwise promising drug candidate. It is therefore of tremendous interest to develop advanced methods for identifying hERG-active compounds in the early stages of drug development, as well as for proposing redesigned compounds with reduced hERG liability and preserved primary pharmacology. In this work, we present CardioGenAI, a machine learning-based framework for re-engineering both developmental and commercially available drugs for reduced hERG activity while preserving their pharmacological activity. The framework incorporates novel state-of-the-art discriminative models for predicting hERG channel activity, as well as activity against the voltage-gated NaV1.5 and CaV1.2 channels due to their potential implications in modulating the arrhythmogenic potential induced by hERG channel blockade. We applied the complete framework to pimozide, an FDA-approved antipsychotic agent that demonstrates high affinity to the hERG channel, and generated 100 refined candidates. Remarkably, among the candidates is fluspirilene, a compound which is of the same class of drugs as pimozide (diphenylmethanes) and therefore has similar pharmacological activity, yet exhibits over 700-fold weaker binding to hERG. Furthermore, we demonstrated the framework's ability to optimize hERG, NaV1.5 and CaV1.2 profiles of multiple FDA-approved compounds while maintaining the physicochemical nature of the original drugs. We envision that this method can effectively be applied to developmental compounds exhibiting hERG liabilities to provide a means of rescuing drug development programs that have stalled due to hERG-related safety concerns. Additionally, the discriminative models can also serve independently as effective components of virtual screening pipelines. We have made all of our software open-source at https://github.com/gregory-kyro/CardioGenAI to facilitate integration of the CardioGenAI framework for molecular hypothesis generation into drug discovery workflows.Scientific contributionThis work introduces CardioGenAI, an open-source machine learning-based framework designed to re-engineer drugs for reduced hERG liability while preserving their pharmacological activity. The complete CardioGenAI framework can be applied to developmental compounds exhibiting hERG liabilities to provide a means of rescuing drug discovery programs facing hERG-related challenges. In addition, the framework incorporates novel state-of-the-art discriminative models for predicting hERG, NaV1.5 and CaV1.2 channel activity, which can function independently as effective components of virtual screening pipelines.

MONSTROUS: a web-based chemical-transporter interaction profiler

Transporters are membrane proteins that are critical for normal cellular function and mediate the transport of endogenous and exogenous chemicals. Chemical interactions with these transporters have the potential to affect the pharmacokinetic properties of drugs. Inhibition of transporters can cause adverse drug-drug interactions and toxicity, whereas if a drug is a substrate of a transporter, it could lead to reduced therapeutic effects. The importance of transporters in drug efficacy and toxicity has led regulatory agencies, such as the U.S. Food and Drug Administration and the European Medicines Agency, to recommend screening of new molecular entities for potential transporter interactions. To aid in the rapid screening and prioritization of drug candidates without transporter liability, we developed a publicly available, web-based transporter profiler, MOlecular traNSporT inhibitoR and substrate predictOr Utility Server (MONSTROUS), that predicts the potential of a chemical to interact with transporters recommended for testing by regulatory agencies. We utilized publicly available data and developed machine learning or similarity-based classification models to predict inhibitors and substrates for 12 transporters. We used graph convolutional neural networks (GCNNs) to develop predictive models for transporters with sufficient bioactivity data, and we implemented two-dimensional similarity-based approach for those without sufficient data. The GCNN inhibitor models have an average five-fold cross-validated receiver operating characteristic area under the curve (ROC-AUC) of 0.85 ± 0.07, and the GCNN substrate models have an average ROC-AUC of 0.79 ± 0.08. We implemented the models along with applicability domain calculations in an easy-to-use web interface and made it publicly available at https://monstrous.bhsai.org/.

hERGAT: predicting hERG blockers using graph attention mechanism through atom- and molecule-level interaction analyses

The human ether-a-go-go-related gene (hERG) channel plays a critical role in the electrical activity of the heart, and its blockers can cause serious cardiotoxic effects. Thus, screening for hERG channel blockers is a crucial step in the drug development process. Many in silico models have been developed to predict hERG blockers, which can efficiently save time and resources. However, previous methods have found it hard to achieve high performance and to interpret the predictive results. To overcome these challenges, we have proposed hERGAT, a graph neural network model with an attention mechanism, to consider compound interactions on atomic and molecular levels. In the atom-level interaction analysis, we applied a graph attention mechanism (GAT) that integrates information from neighboring nodes and their extended connections. The hERGAT employs a gated recurrent unit (GRU) with the GAT to learn information between more distant atoms. To confirm this, we performed clustering analysis and visualized a correlation heatmap, verifying the interactions between distant atoms were considered during the training process. In the molecule-level interaction analysis, the attention mechanism enables the target node to focus on the most relevant information, highlighting the molecular substructures that play crucial roles in predicting hERG blockers. Through a literature review, we confirmed that highlighted substructures have a significant role in determining the chemical and biological characteristics related to hERG activity. Furthermore, we integrated physicochemical properties into our hERGAT model to improve the performance. Our model achieved an area under the receiver operating characteristic of 0.907 and an area under the precision-recall of 0.904, demonstrating its effectiveness in modeling hERG activity and offering a reliable framework for optimizing drug safety in early development stages.Scientific contribution:hERGAT is a deep learning model for predicting hERG blockers by combining GAT and GRU, enabling it to capture complex interactions at atomic and molecular levels. We improve the model's interpretability by analyzing the highlighted molecular substructures, providing valuable insights into their roles in determining hERG activity. The model achieves high predictive performance, confirming its potential as a preliminary tool for early cardiotoxicity assessment and enhancing the reliability of the results.

hERGBoost: A gradient boosting model for quantitative IC50 prediction of hERG channel blockers

The human ether-a-go-go-related gene (hERG) potassium channel is pivotal in drug discovery due to its susceptibility to blockage by drug candidate molecules, which can cause severe cardiotoxic effects. Consequently, identifying and excluding potential hERG channel blockers at the earliest stages of drug development is crucial. Most traditional machine learning models predict a molecule's cardiotoxicity or non-cardiotoxicity typically at 10 μM, which doesn't account for compounds with low IC50 values that are non-toxic at therapeutic levels due to their high effectiveness at lower concentrations. To address the need for more precise, quantitative predictions, we developed hERGBoost, a cutting-edge machine learning model employing a gradient-boosting algorithm. This model demonstrates superior accuracy in predicting the IC50 of drug candidates. Trained on a specially curated dataset for this study, hERGBoost not only exhibited excellent performance in external validation, achieving an R2 score of 0.394 and a low root mean square error of 0.616 but also significantly outstripped previous models in both qualitative and quantitative assessments. Representing a notable leap forward in the prediction of hERG channel blockers, the hERGBoost model and its datasets are freely available to the drug discovery community on our web server at. http://ssbio.cau.ac.kr/software/hergboost This resource promises to be invaluable in advancing safer pharmaceutical development.

QSAR Classification Modeling Using Machine Learning with a Consensus-Based Approach for Multivariate Chemical Hazard End Points

This study introduces an innovative computational approach using hybrid machine learning models to predict toxicity across eight critical end points: cardiac toxicity, inhalation toxicity, dermal toxicity, oral toxicity, skin irritation, skin sensitization, eye irritation, and respiratory irritation. Leveraging advanced cheminformatics tools, we extracted relevant features from curated data sets, incorporating a range of descriptors such as Morgan circular fingerprints, MACCS keys, Mordred calculation descriptors, and physicochemical properties. The consensus model was developed by selecting the best-performing classifier-Random Forest (RF), eXtreme Gradient Boosting (XGBoost), or Support Vector Machines (SVM)-for each descriptor, optimizing predictive accuracy and robustness across the end points. The model obtained strong predictive performance, with area under the curve (AUC) scores ranging from 0.78 to 0.90. This framework offers a reliable, ethical, and effective in silico approach to chemical safety assessment, underscoring the potential of advanced computational methods to support both regulatory and research applications in toxicity prediction.

Design and investigation of novel iridoid-based peptide conjugates for targeting EGFR and its mutants L858R and T790M/L858R/C797S: an in silico study

In this work, we designed novel peptide conjugates with plant-based iridoid and lichen-derived depside derivatives to target the wild-type EGFR (WT) and its mutants, L858R and T790M/L858R/C797S triple mutant. These mutations are often expressed in multiple cancers, particularly lung cancer. Specifically, the iridoids included 7-deoxyloganetic acid (7-DGA) and loganic acid (LG), while the depside derivative was sekikaic acid (SK). These compounds are known for their innate anticancer properties and were conjugated with two separate peptide sequences KLPGWSG (K) and YSIPKSS (Y). These sequences have been shown to target EGFR in previous phage display library screening, although the mechanism is unknown. Thus, we created the di-conjugates for dual targeting and investigated their interactions of the di-conjugates and that of the neat peptides with the kinase domain of EGFR (WT) and the two mutants using molecular docking, molecular dynamics (MD) simulations, and MM-GBSA analysis. Docking studies revealed that the (7-DGA)2-K showed the highest binding affinity at - 9.3 kcal/mol with the L858R mutant, while (LG)2-Y displayed the highest binding affinity at - 9.0 kcal/mol for the triple mutant receptor. Our results indicated that several of the conjugates interacted with crucial residues of the kinase domain, including ASP855 and THR854 (activation loop), MET793 and PRO794 (hinge region), ARG841 (catalytic loop), and LYS728 and LEU718 of the glycine-rich P-loop. Interestingly, strong hydrophobic interactions were also observed with the C-terminal tail residues, such as PHE997 and ALA1000 as well as with ARG999 for the YSIPKSS peptide and most of the conjugates. The hydroxyl group of the cyclopentane ring and the oxygen of the pyran ring of the (7-DGA)2-peptide conjugates contributed to binding particularly in the hinge region, while the peptide components formed an extended structure that bound well into the C-lobe. The (SK)2-Y di-conjugate and KLPGWSG peptide formed hydrogen bonds with the SER797 residue of the triple mutant. Overall, our results show that the (7-DGA)2-K, di-conjugate, the (7-DGA)2-Y di-conjugate, and the neat YSIPKSS demonstrated strong and stable binding with the L858R mutant and the highly resistant triple mutant EGFR, respectively. The novel designed conjugates demonstrate potential for further optimization for laboratory studies aimed at developing new therapeutics for targeting specific EGFR mutant expressing cells.

Unlocking the potential of AI: Machine learning and deep learning models for predicting carcinogenicity of chemicals

The escalating apprehension surrounding the carcinogenic potential of chemicals emphasizes the imperative need for efficient methods of assessing carcinogenicity. Conventional experimental approaches such as in vitro and in vivo assays, albeit effective, suffer from being costly and time-consuming. In response to this challenge, new alternative methodologies, notably machine learning and deep learning techniques, have attracted attention for their potential in developing carcinogenicity prediction models. This article reviews the progress in predicting carcinogenicity using various machine learning and deep learning algorithms. A comparative analysis on these developed models reveals that support vector machine, random forest, and ensemble learning are commonly preferred for their robustness and effectiveness in predicting chemical carcinogenicity. Conversely, models based on deep learning algorithms, such as feedforward neural network, convolutional neural network, graph convolutional neural network, capsule neural network, and hybrid neural networks, exhibit promising capabilities but are limited by the size of available carcinogenicity datasets. This review provides a comprehensive analysis of current machine learning and deep learning models for carcinogenicity prediction, underscoring the importance of high-quality and large datasets. These observations are anticipated to catalyze future advancements in developing effective and generalizable machine learning and deep learning models for predicting chemical carcinogenicity.

Machine learning in preclinical drug discovery

Drug-discovery and drug-development endeavors are laborious, costly and time consuming. These programs can take upward of 12 years and cost US $2.5 billion, with a failure rate of more than 90%. Machine learning (ML) presents an opportunity to improve the drug-discovery process. Indeed, with the growing abundance of public and private large-scale biological and chemical datasets, ML techniques are becoming well positioned as useful tools that can augment the traditional drug-development process. In this Perspective, we discuss the integration of algorithmic methods throughout the preclinical phases of drug discovery. Specifically, we highlight an array of ML-based efforts, across diverse disease areas, to accelerate initial hit discovery, mechanism-of-action (MOA) elucidation and chemical property optimization. With advances in the application of ML across diverse therapeutic areas, we posit that fully ML-integrated drug-discovery pipelines will define the future of drug-development programs.

In Silico Prediction of Chemical Acute Dermal Toxicity Using Explainable Machine Learning Methods

The research on acute dermal toxicity has consistently been a crucial component in assessing the potential risks of human exposure to active ingredients in pesticides and related plant protection products. However, it is difficult to directly identify the acute dermal toxicity of potential compounds through animal experiments alone. In our study, we separately integrated 1735 experimental data based on rabbits and 1679 experimental data based on rats to construct acute dermal toxicity prediction models using machine learning and deep learning algorithms. The best models for the two animal species achieved AUC values of 78.0 and 82.0%, respectively, on 10-fold cross-validation. Additionally, we employed SARpy to extract structural alerts, and in conjunction with Shapley additive explanation and attentive FP heatmap, we identified important features and structural fragments associated with acute dermal toxicity. This approach offers valuable insights for the detection of positive compounds. Moreover, a standalone software tool was developed to make acute dermal toxicity prediction easier. In summary, our research would provide an effective tool for acute dermal toxicity evaluation of pesticides, cosmetics, and drug safety assessment.

Exploring the artificial intelligence and machine learning models in the context of drug design difficulties and future potential for the pharmaceutical sectors

Artificial intelligence (AI), particularly deep learning as a subcategory of AI, provides opportunities to accelerate and improve the process of discovering and developing new drugs. The use of AI in drug discovery is still in its early stages, but it has the potential to revolutionize the way new drugs are discovered and developed. As AI technology continues to evolve, it is likely that AI will play an even greater role in the future of drug discovery. AI is used to identify new drug targets, design new molecules, and predict the efficacy and safety of potential drugs. The inclusion of AI in drug discovery can screen millions of compounds in a matter of hours, identifying potential drug candidates that would have taken years to find using traditional methods. AI is highly utilized in the pharmaceutical industry by optimizing processes, reducing waste, and ensuring quality control. This review covers much-needed topics, including the different types of machine-learning techniques, their applications in drug discovery, and the challenges and limitations of using machine learning in this field. The state-of-the-art of AI-assisted pharmaceutical discovery is described, covering applications in structure and ligand-based virtual screening, de novo drug creation, prediction of physicochemical and pharmacokinetic properties, drug repurposing, and related topics. Finally, many obstacles and limits of present approaches are outlined, with an eye on potential future avenues for AI-assisted drug discovery and design.

Review of machine learning and deep learning models for toxicity prediction

The ever-increasing number of chemicals has raised public concerns due to their adverse effects on human health and the environment. To protect public health and the environment, it is critical to assess the toxicity of these chemicals. Traditional in vitro and in vivo toxicity assays are complicated, costly, and time-consuming and may face ethical issues. These constraints raise the need for alternative methods for assessing the toxicity of chemicals. Recently, due to the advancement of machine learning algorithms and the increase in computational power, many toxicity prediction models have been developed using various machine learning and deep learning algorithms such as support vector machine, random forest, k-nearest neighbors, ensemble learning, and deep neural network. This review summarizes the machine learning- and deep learning-based toxicity prediction models developed in recent years. Support vector machine and random forest are the most popular machine learning algorithms, and hepatotoxicity, cardiotoxicity, and carcinogenicity are the frequently modeled toxicity endpoints in predictive toxicology. It is known that datasets impact model performance. The quality of datasets used in the development of toxicity prediction models using machine learning and deep learning is vital to the performance of the developed models. The different toxicity assignments for the same chemicals among different datasets of the same type of toxicity have been observed, indicating benchmarking datasets is needed for developing reliable toxicity prediction models using machine learning and deep learning algorithms. This review provides insights into current machine learning models in predictive toxicology, which are expected to promote the development and application of toxicity prediction models in the future.