CYPlebrity: Machine learning models for the prediction of inhibitors of cytochrome P450 enzymes

Harnessing machine learning to predict cytochrome P450 inhibition through molecular properties

Cytochrome P450 enzymes are a superfamily of enzymes responsible for the metabolism of a variety of medicines and xenobiotics. Among the Cytochrome P450 family, five isozymes that include 1A2, 2C9, 2C19, 2D6, and 3A4 are most important for the metabolism of xenobiotics. Inhibition of any of these five CYP isozymes causes drug-drug interactions with high pharmacological and toxicological effects. So, the inhibition or non-inhibition prediction of these isozymes is of great importance. Many techniques based on machine learning and deep learning algorithms are currently being used to predict whether these isozymes will be inhibited or not. In this study, three different molecular or substructural properties that include Morgan, MACCS and Morgan (combined) and RDKit of the various molecules are used to train a distinct SVM model against each isozyme (1A2, 2C9, 2C19, 2D6, and 3A4). On the independent dataset, Morgan fingerprints provided the best results, while MACCS and Morgan (combined) achieved comparable results in terms of balanced accuracy (BA), sensitivity (Sn), and Mathews correlation coefficient (MCC). For the Morgan fingerprints, balanced accuracies (BA), Mathews correlation coefficients (MCC), and sensitivities (Sn) against each CYPs isozyme, 1A2, 2C9, 2C19, 2D6, and 3A4 on an independent dataset ranged between 0.81 and 0.85, 0.61 and 0.70, 0.72 and 0.83, respectively. Similarly, on the independent dataset, MACCS and Morgan (combined) fingerprints achieved competitive results in terms of balanced accuracies (BA), Mathews correlation coefficients (MCC), and sensitivities (Sn) against each CYPs isozyme, 1A2, 2C9, 2C19, 2D6, and 3A4, which ranged between 0.79 and 0.85, 0.59 and 0.69, 0.69 and 0.82, respectively.

Effect of Lipid Bilayer Anchoring on the Conformational Properties of the Cytochrome P450 2D6 Binding Site

Human cytochrome P450 (CYP) proteins metabolize 75% of small-molecule pharmaceuticals, which makes structure-based modeling of CYP metabolism and inhibition, bolstered by the current availability of X-ray crystal structures of CYP globular catalytic domains, an attractive prospect. Accounting for this broad metabolic capacity is a combination of the existence of multiple different CYP proteins and the capacity of a single CYP protein to metabolize multiple different small molecules. It is thought that structural plasticity and flexibility contribute to this latter property; therefore, incorporating diverse conformational states of a particular CYP is likely an important consideration in structure-based CYP metabolism and inhibition modeling. While all-atom explicit-solvent molecular dynamics simulations can be used to generate conformational ensembles under biologically relevant conditions, existing CYP crystal structures are of the globular domain only, whereas human CYPs contain N-terminal transmembrane and linker peptides that anchor the globular catalytic domain to the endoplasmic reticulum. To determine whether this can cause significant differences in the sampled binding site conformations, microsecond scale all-atom explicit-solvent molecular dynamics simulations of the CYP2D6 globular domain in an aqueous environment were compared with those of the full-length protein anchored in a POPC lipid bilayer. While bilayer-anchoring damped some structural fluctuations in the globular domain relative to the aqueous simulations, none of the affected residues included binding site pocket residues. Furthermore, clustering of molecular dynamics snapshots based on either pairwise binding site pocket RMSD or volume differences demonstrated a lack of separation of snapshots from the two simulation conditions into different clusters. These results suggest the substantially simpler and computationally cheaper aqueous simulation approach can be used to generate a relevant conformational ensemble of the CYP2D6 binding site for structure-based metabolism and inhibition modeling.

Evaluation of machine learning models for cytochrome P450 3A4, 2D6, and 2C9 inhibition

Cytochrome P450 (CYP) enzymes are involved in the metabolism of approximately 75% of marketed drugs. Inhibition of the major drug-metabolizing P450s could alter drug metabolism and lead to undesirable drug-drug interactions. Therefore, it is of great significance to explore the inhibition of P450s in drug discovery. Currently, machine learning including deep learning algorithms has been widely used for constructing in silico models for the prediction of P450 inhibition. These models exhibited varying predictive performance depending on the use of machine learning algorithms and molecular representations. This leads to the difficulty in the selection of appropriate models for practical use. In this study, we systematically evaluated the conventional machine learning and deep learning models for three major P450 enzymes, CYP3A4, CYP2D6, and CYP2C9 from several perspectives, such as algorithms, molecular representation, and data partitioning strategies. Our results showed that the XGBoost and CatBoost algorithms coupled with the combined fingerprint/physicochemical descriptor features exhibited the best performance with Area Under Curve (AUC)  of 0.92, while the deep learning models were generally inferior to the conventional machine learning models (average AUC reached 0.89) on the same test sets. We also found that data volume and sampling strategy had a minor effect on model performance. We anticipate that these results are helpful for the selection of molecular representations and machine learning/deep learning algorithms in the P450 model construction and the future model development of P450 inhibition.

Predicting ADMET Properties from Molecule SMILE: A Bottom-Up Approach Using Attention-Based Graph Neural Networks

Understanding the pharmacokinetics, safety and efficacy of candidate drugs is crucial for their success. One key aspect is the characterization of absorption, distribution, metabolism, excretion and toxicity (ADMET) properties, which require early assessment in the drug discovery and development process. This study aims to present an innovative approach for predicting ADMET properties using attention-based graph neural networks (GNNs). The model utilizes a graph-based representation of molecules directly derived from Simplified Molecular Input Line Entry System (SMILE) notation. Information is processed sequentially, from substructures to the whole molecule, employing a bottom-up approach. The developed GNN is tested and compared with existing approaches using six benchmark datasets and by encompassing regression (lipophilicity and aqueous solubility) and classification (CYP2C9, CYP2C19, CYP2D6 and CYP3A4 inhibition) tasks. Results show the effectiveness of our model, which bypasses the computationally expensive retrieval and selection of molecular descriptors. This approach provides a valuable tool for high-throughput screening, facilitating early assessment of ADMET properties and enhancing the likelihood of drug success in the development pipeline.

ADME/PK Insights of Crocetin: A Molecule Having an Unusual Chemical Structure with Druglike Features

Crocetin is a promising phyto-based molecule to treat Alzheimer's disease (AD). The chemical structure of crocetin is incongruent with various standard structural features of CNS drugs. As poor pharmacokinetic behavior is the major hurdle for any candidate to become a drug, we elucidated its druggable characteristics by implementing in silico, in vitro, and in vivo approaches, as limited ADME/PK information is available. Results demonstrate several attributes of crocetin based on rules of drug-likeness, lipophilicity, pKa, P-gp inhibitory activity, plasma stability, RBC partitioning, metabolic stability, CYP inhibitory action, blood-brain barrier (BBB) permeability, oral bioavailability, and pharmacokinetic interaction with marketed anti-Alzheimer's drugs (memantine, donepezil, galantamine, and rivastigmine). However, aqueous solubility, chemical stability, plasma protein binding, and P-gp induction are some concerns associated with this molecule that should be taken into consideration during its further development. Overall results indicate favorable ADME/PK behavior and potential druggable candidature of crocetin.

Deep Learning Models Compared to Experimental Variability for the Prediction of CYP3A4 Time-Dependent Inhibition

Most drugs are mainly metabolized by cytochrome P450 (CYP450), which can lead to drug-drug interactions (DDI). Specifically, time-dependent inhibition (TDI) of CYP3A4 isoenzyme has been associated with clinically relevant DDI. To overcome potential DDI issues, high-throughput in vitro assays were established to assess the TDI of CYP3A4 during the discovery and lead optimization phases. However, in silico machine learning models would enable an earlier and larger-scale assessment of TDI potential liabilities. For CYP inhibition, most modeling efforts have focused on highly imbalanced and small data sets. Moreover, assay variability is rarely considered, which is key to understand the model's quality and suitability for decision-making. In this work, machine learning models were built for the prediction of TDI of CYP3A4, evaluated prospectively, and compared to the variability of the experimental assay. Different modeling strategies were investigated to assess their influence on the model's performance. Through multitask learning, additional data sets were leveraged for model building, coming from public databases, in-house CYP-related assays, or other pharmaceutical companies (federated learning). Apart from the numerical prediction of inactivation rates of CYP3A4 TDI, three-class predictions were carried out, giving a negative (inactivation rate kobs < 0.01 min-1), weak positive (0.01 ≤ kobs ≤ 0.025 min-1), or positive (kobs > 0.025 min-1) output. The final multitask graph neural network model achieved misclassification rates of 8 and 7% for positive and negative TDI, respectively. Importantly, the presented deep learning-based predictions had a similar precision to the reproducibility of in vitro experiments and thus offered great opportunities for drug design, early derisk of DDI potential, and selection of experiments. To facilitate CYP inhibition modeling efforts in the public domain, the developed model was used to annotate ∼16 000 publicly available structures, and a surrogate data set is shared as Supporting Information.

A Review of CYP-Mediated Drug Interactions: Mechanisms and In Vitro Drug-Drug Interaction Assessment

Drug metabolism is a major determinant of drug concentrations in the body. Drug-drug interactions (DDIs) caused by the co-administration of multiple drugs can lead to alteration in the exposure of the victim drug, raising safety or effectiveness concerns. Assessment of the DDI potential starts with in vitro experiments to determine kinetic parameters and identify risks associated with the use of comedication that can inform future clinical studies. The diverse range of experimental models and techniques has significantly contributed to the examination of potential DDIs. Cytochrome P450 (CYP) enzymes are responsible for the biotransformation of many drugs on the market, making them frequently implicated in drug metabolism and DDIs. Consequently, there has been a growing focus on the assessment of DDI risk for CYPs. This review article provides mechanistic insights underlying CYP inhibition/induction and an overview of the in vitro assessment of CYP-mediated DDIs.

Computational and artificial intelligence-based approaches for drug metabolism and transport prediction

Drug metabolism and transport, orchestrated by drug-metabolizing enzymes (DMEs) and drug transporters (DTs), are implicated in drug-drug interactions (DDIs) and adverse drug reactions (ADRs). Reliable and precise predictions of DDIs and ADRs are critical in the early stages of drug development to reduce the rate of drug candidate failure. A variety of experimental and computational technologies have been developed to predict DDIs and ADRs. Recent artificial intelligence (AI) approaches offer new opportunities for better predicting and understanding the complex processes related to drug metabolism and transport. We summarize the role of major DMEs and DTs, and provide an overview of current progress in computational approaches for the prediction of drug metabolism, transport, and DDIs, with an emphasis on AI including machine learning (ML) and deep learning (DL) modeling.

An Uncertainty-Guided Deep Learning Method Facilitates Rapid Screening of CYP3A4 Inhibitors

Cytochrome P450 3A4 (CYP3A4), a prominent member of the P450 enzyme superfamily, plays a crucial role in metabolizing various xenobiotics, including over 50% of clinically significant drugs. Evaluating CYP3A4 inhibition before drug approval is essential to avoiding potentially harmful pharmacokinetic drug-drug interactions (DDIs) and adverse drug reactions (ADRs). Despite the development of several CYP inhibitor prediction models, the primary approach for screening CYP inhibitors still relies on experimental methods. This might stem from the limitations of existing models, which only provide deterministic classification outcomes instead of precise inhibition intensity (e.g., IC50) and often suffer from inadequate prediction reliability. To address this challenge, we propose an uncertainty-guided regression model to accurately predict the IC50 values of anti-CYP3A4 activities. First, a comprehensive data set of CYP3A4 inhibitors was compiled, consisting of 27,045 compounds with classification labels, including 4395 compounds with explicit IC50 values. Second, by integrating the predictions of the classification model trained on a larger data set and introducing an evidential uncertainty method to rank prediction confidence, we obtained a high-precision and reliable regression model. Finally, we use the evidential uncertainty values as a trustworthy indicator to perform a virtual screening of an in-house compound set. The in vitro experiment results revealed that this new indicator significantly improved the hit ratio and reduced false positives among the top-ranked compounds. Specifically, among the top 20 compounds ranked with uncertainty, 15 compounds were identified as novel CYP3A4 inhibitors, and three of them exhibited activities less than 1 μM. In summary, our findings highlight the effectiveness of incorporating uncertainty in compound screening, providing a promising strategy for drug discovery and development.

Comparison and summary of in silico prediction tools for CYP450-mediated drug metabolism

The cytochrome P450 (CYP450) enzyme system is responsible for the metabolism of more than two-thirds of xenobiotics. This review summarizes reports of a series of in silico tools for CYP450 enzyme-drug interaction predictions, including the prediction of sites of metabolism (SOM) of a drug and the identification of inhibitor/substrates for CYP subtypes. We also evaluated four prediction tools to identify CYP inhibitors utilizing 52 of the most frequently prescribed drugs. ADMET Predictor and CYPlebrity demonstrated the best performance. We hope that this review provides guidance for choosing appropriate enzyme prediction tools from a variety of in silico platforms to meet individual needs. Such predictions are useful for medicinal chemists to prioritize their designed compounds for further drug discovery.

Machine Learning Models to Predict Cytochrome P450 2B6 Inhibitors and Substrates

Cytochrome P450 2B6 (CYP2B6) is responsible for the metabolism of ∼7% of marketed drugs. The in vitro drug interaction studies guidance for industry issued by the FDA stipulates that drug sponsors need to evaluate whether the investigated drugs interact with the major drug-metabolizing P450s including CYP2B6. Therefore, there has been greater attention to the development of predictive models for CYP2B6 inhibitors and substrates. In this study, conventional machine learning and deep learning models were developed to predict CYP2B6 inhibitors and substrates. Our results showed that the best CYP2B6 inhibitor model yielded the AUC values of 0.95 and 0.75 with the 10-fold cross-validation and the test set, respectively, and the best CYP2B6 substrate model produced the AUC values of 0.93 and 0.90 with the 10-fold cross-validation and the test set, respectively. The generalization ability of the CYP2B6 inhibitor and substrate models was assessed by using the external validation sets. Several significant substructural fragments relevant to CYP2B6 inhibitors and substrates were detected via frequency substructure analysis and information gain. In addition, the applicability domain of the models was defined by employing a nonparametric method based on the probability density distribution. We anticipate that our results would be useful for the prediction of potential CYP2B6 inhibitors and substrates in the early stage of drug discovery.

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

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

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

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

A unified GCNN model for predicting CYP450 inhibitors by using graph convolutional neural networks with attention mechanism

Undesirable drug-drug interactions (DDIs) may lead to serious adverse side effects when more than two drugs are administered to a patient simultaneously. One of the most common DDIs is caused by unexpected inhibition of a specific human cytochrome P450 (CYP450), which plays a dominant role in the metabolism of the co-administered drugs. Therefore, a unified and reliable method for predicting the potential inhibitors of CYP450 family is extremely important in drug development. In this work, graph convolutional neural network (GCN) with attention mechanism and 1-D convolutional neural network (CNN) were used to extract the features of CYP ligands and the binding sites of CYP450 respectively, which were then combined to establish a unified GCN-CNN (GCNN) model for predicting the inhibitors of 5 dominant CYP isoforms, i.e., 1A2, 2C9, 2C19, 2D6, and 3A4. Overall, the established GCNN model showed good performances on the test samples and achieved better performances than the recently proposed iCYP-MFE model by using the same datasets. Based on the heat-map analysis of the resulting molecular graphs, the key structural determinants of the CYP inhibitors were further explored.

Computational Prediction of Inhibitors and Inducers of the Major Isoforms of Cytochrome P450

Human cytochrome P450 enzymes (CYPs) are heme-containing monooxygenases. This superfamily of drug-metabolizing enzymes is responsible for the metabolism of most drugs and other xenobiotics. The inhibition of CYPs may lead to drug–drug interactions and impair the biotransformation of drugs. CYP inducers may decrease the bioavailability and increase the clearance of drugs. Based on the freely available databases ChEMBL and PubChem, we have collected over 70,000 records containing the structures of inhibitors and inducers together with the IC50 values for the inhibitors of the five major human CYPs: 1A2, 3A4, 2D6, 2C9, and 2C19. Based on the collected data, we developed (Q)SAR models for predicting inhibitors and inducers of these CYPs using GUSAR and PASS software. The developed (Q)SAR models could be applied for assessment of the interaction of novel drug-like substances with the major human CYPs. The created (Q)SAR models demonstrated reasonable accuracy of prediction. They have been implemented in the web application P450-Analyzer that is freely available via the Internet.

Machine Learning in Drug Metabolism Study

Metabolic reactions in the body transform the administered drug into metabolites. These metabolites exhibit diverse biological activities. Drug metabolism is the major underlying cause of drug overdose-related toxicity, adversative drug effects and the drug's reduced efficacy. Though metabolic reactions deactivate a drug, drug metabolites are often considered pivotal agents for off-target effects or toxicity. On the other side, in combination drug therapy, one drug may influence another drug's metabolism and clearance and is thus considered one of the primary causes of drug-drug interactions. Today with the advancement of machine learning, the metabolic fate of a drug candidate can be comprehensively studied throughout the drug development procedure. Naïve Bayes, Logistic Regression, k-Nearest Neighbours, Decision Trees, different Boosting and Ensemble methods, Support Vector Machines and Artificial Neural Network boosted Deep Learning are some machine learning algorithms which are being extensively used in such studies. Such tools are covering several attributes of drug metabolism, with an emphasis on the prediction of drug-drug interactions, drug-target-interactions, clinical drug responses, metabolite predictions, sites of metabolism, etc. These reports are crucial for evaluating metabolic stability and predicting prospective drug-drug interactions, and can help pharmaceutical companies accelerate the drug development process in a less resourcedemanding manner than what in vitro studies offer. It could also help medical practitioners to use combinatorial drug therapy in a more resourceful manner. Also, with the help of the enormous growth of deep learning, traditional fields of computational drug development like molecular interaction fields, molecular docking, quantitative structure-toactivity relationship (QSAR) studies and quantum mechanical simulations are producing results which were unimaginable couple of years back. This review provides a glimpse of a few contextually relevant machine learning algorithms and then focuses on their outcomes in different studies.

Machine learning-driven identification of drugs inhibiting cytochrome P450 2C9

Cytochrome P450 2C9 (CYP2C9) is a major drug-metabolizing enzyme that represents 20% of the hepatic CYPs and is responsible for the metabolism of 15% of drugs. A general concern in drug discovery is to avoid the inhibition of CYP leading to toxic drug accumulation and adverse drug-drug interactions. However, the prediction of CYP inhibition remains challenging due to its complexity. We developed an original machine learning approach for the prediction of drug-like molecules inhibiting CYP2C9. We created new predictive models by integrating CYP2C9 protein structure and dynamics knowledge, an original selection of physicochemical properties of CYP2C9 inhibitors, and machine learning modeling. We tested the machine learning models on publicly available data and demonstrated that our models successfully predicted CYP2C9 inhibitors with an accuracy, sensitivity and specificity of approximately 80%. We experimentally validated the developed approach and provided the first identification of the drugs vatalanib, piriqualone, ticagrelor and cloperidone as strong inhibitors of CYP2C9 with IC values <18 μM and sertindole, asapiprant, duvelisib and dasatinib as moderate inhibitors with IC50 values between 40 and 85 μM. Vatalanib was identified as the strongest inhibitor with an IC50 value of 0.067 μM. Metabolism assays allowed the characterization of specific metabolites of abemaciclib, cloperidone, vatalanib and tarafenacin produced by CYP2C9. The obtained results demonstrate that such a strategy could improve the prediction of drug-drug interactions in clinical practice and could be utilized to prioritize drug candidates in drug discovery pipelines.