RS-predictor: a new tool for predicting sites of cytochrome P450-mediated metabolism applied to CYP 3A4

Protein representations: Encoding biological information for machine learning in biocatalysis

Enzymes offer a more environmentally friendly and low-impact solution to conventional chemistry, but they often require additional engineering for their application in industrial settings, an endeavour that is challenging and laborious. To address this issue, the power of machine learning can be harnessed to produce predictive models that enable the in silico study and engineering of improved enzymatic properties. Such machine learning models, however, require the conversion of the complex biological information to a numerical input, also called protein representations. These inputs demand special attention to ensure the training of accurate and precise models, and, in this review, we therefore examine the critical step of encoding protein information to numeric representations for use in machine learning. We selected the most important approaches for encoding the three distinct biological protein representations - primary sequence, 3D structure, and dynamics - to explore their requirements for employment and inductive biases. Combined representations of proteins and substrates are also introduced as emergent tools in biocatalysis. We propose the division of fixed representations, a collection of rule-based encoding strategies, and learned representations extracted from the latent spaces of large neural networks. To select the most suitable protein representation, we propose two main factors to consider. The first one is the model setup, which is influenced by the size of the training dataset and the choice of architecture. The second factor is the model objectives such as consideration about the assayed property, the difference between wild-type models and mutant predictors, and requirements for explainability. This review is aimed at serving as a source of information and guidance for properly representing enzymes in future machine learning models for biocatalysis.

Predictive Modelling in pharmacokinetics: from in-silico simulations to personalized medicine

INTRODUCTION

Pharmacokinetic parameters assessment is a critical aspect of drug discovery and development, yet challenges persist due to limited training data. Despite advancements in machine learning and in-silico predictions, scarcity of data hampers accurate prediction of drug candidates' pharmacokinetic properties.

AREAS COVERED

The study highlights current developments in human pharmacokinetic prediction, talks about attempts to apply synthetic approaches for molecular design, and searches several databases, including Scopus, PubMed, Web of Science, and Google Scholar. The article stresses importance of rigorous analysis of machine learning model performance in assessing progress and explores molecular modeling (MM) techniques, descriptors, and mathematical approaches. Transitioning to clinical drug development, article highlights AI (Artificial Intelligence) based computer models optimizing trial design, patient selection, dosing strategies, and biomarker identification. In-silico models, including molecular interactomes and virtual patients, predict drug performance across diverse profiles, underlining the need to align model results with clinical studies for reliability. Specialized training for human specialists in navigating predictive models is deemed critical. Pharmacogenomics, integral to personalized medicine, utilizes predictive modeling to anticipate patient responses, contributing to more efficient healthcare system. Challenges in realizing potential of predictive modeling, including ethical considerations and data privacy concerns, are acknowledged.

EXPERT OPINION

AI models are crucial in drug development, optimizing trials, patient selection, dosing, and biomarker identification and hold promise for streamlining clinical investigations.

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.

Prediction of Sites of Metabolism of CYP3A4 Substrates Utilizing Docking-Derived Geometric Features

Cytochrome P450 3A4 (CYP3A4) is one of the major drug-metabolizing enzymes in the human body and is responsible for the metabolism of ∼50% of clinically used drugs. Therefore, the identification of the compound's sites of metabolism (SOMs) mediated by CYP3A4 is of utmost importance in the early stage of drug discovery and development. Herein, docking-based approaches incorporating geometric features were used for SOMs prediction of CYP3A4 substrates. The cross-docking poses of a relatively large data set containing 474 substrates were analyzed in depth, and a widely observed geometric pattern called the close proximity of SOMs was derived from the poses. On the basis of the close proximity, several structure-based models have been constructed, which demonstrated better performance than those structure-based models using the criterion of Fe-SOM distance. For further improving the prediction performance, the structure-based models were also combined with the well-known ligand-based model SMARTCyp. One combined model exhibited good performance on the SOMs prediction of an external substrate set containing kinase inhibitors, PROTACs, approved drugs, and some lead compounds.

Debelalactone Prevents Hepatic Cancer via Diminishing the Inflammatory Response and Oxidative Stress on Male Wistar Rats

The current study was conducted to exemplify the effect of debelalactone on tissue protection, chronic hepatic inflammation, hepatic protection and oxidative stress induced by diethyl nitrosamine in Wistar rats. Therefore, DEN (200 mg/kg) was used for the induction the hepatocellular carcinoma (HCC) and the level of serum alpha fetoprotein was used for the estimation and confirmation of HCC. The study illustrated that debelalactone (DL) significantly downregulated the hepatic, non-hepatic parameters such as aspartate aminotransferase, alanine aminotransferase, alpha fetoprotein, NO levels, total protein, albumin, blood urea nitrogen, total bilirubin, and direct bilirubin in dose dependent manner, as well as noticeably improving the body weight, of treated animals. The macroscopically observation of DEN-induced rat liver showed the formation of informalities in liver tissue, which was reduced with treatment of DL at dose dependent manner. However, antioxidant markers and inflammatory mediators such as lipid peroxidation, catalase, superoxide dismutase, glutathione peroxidase and transferase, TNF-α, IL-1β, IL-6, and NF-kB restored up to the normal level by DL. The histopathology studies showed that the treated group of animals returned to a normal status. Collectively, it can be concluded that debelalactone mediated chemoprevention in the DEN-induced rats via an increase in the activities of endogenous enzymes and/or inhibition the precancerous cells.

Validation strategies for target prediction methods

Computational methods for target prediction, based on molecular similarity and network-based approaches, machine learning, docking and others, have evolved as valuable and powerful tools to aid the challenging task of mode of action identification for bioactive small molecules such as drugs and drug-like compounds. Critical to discerning the scope and limitations of a target prediction method is understanding how its performance was evaluated and reported. Ideally, large-scale prospective experiments are conducted to validate the performance of a model; however, this expensive and time-consuming endeavor is often not feasible. Therefore, to estimate the predictive power of a method, statistical validation based on retrospective knowledge is commonly used. There are multiple statistical validation techniques that vary in rigor. In this review we discuss the validation strategies employed, highlighting the usefulness and constraints of the validation schemes and metrics that are employed to measure and describe performance. We address the limitations of measuring only generalized performance, given that the underlying bioactivity and structural data are biased towards certain small-molecule scaffolds and target families, and suggest additional aspects of performance to consider in order to produce more detailed and realistic estimates of predictive power. Finally, we describe the validation strategies that were employed by some of the most thoroughly validated and accessible target prediction methods.

Metabolism-Coupled Cell-Independent Acetylcholinesterase Activity Assay for Evaluation of the Effects of Chlorination on Diazinon Toxicity

Drinking water quality guideline values for toxic compounds are determined based on their acceptable daily intake. The toxicological end point for determining the acceptable daily intake of most organophosphorus insecticides is inhibition of acetylcholinesterase (AChE). Although insecticides ingested with drinking water are partly metabolized by the liver before transport to the rest of the body, no current cell-independent AChE activity assay takes the effects of metabolism into account. Here, we incorporated metabolism into a cell-independent AChE activity assay and then evaluated the change in anti-AChE activity during chlorination of a solution containing the organophosphorus insecticide diazinon. The anti-AChE activities of solutions of diazinon or diazinon-oxon, the major transformation product of diazinon during chlorination, were dramatically changed by metabolism: the activity of diazinon solution was markedly increased, whereas that of diazinon-oxon solution was slightly decreased, clearly indicating the importance of incorporating metabolism into assays examining toxicity after oral ingestion. Upon chlorination, diazinon was completely transformed, in part to diazinon-oxon. Although diazinon solution without metabolism did not show anti-AChE activity before chlorination, it did after chlorination. In contrast, with metabolism, diazinon solution did show anti-AChE activity before chlorination, but chlorination gradually decreased this activity over time. The observed anti-AChE activities were attributable solely to diazinon and diazinon-oxon having been contained in the samples before metabolism, clearly suggesting that the presence not only of diazinon but also of diazinon-oxon should be monitored in drinking water. Further examination using a combination of tandem mass spectrometry and in silico site-of-metabolism analyses revealed the structure of a single metabolite that was responsible for the observed anti-AChE activity after metabolism. However, because this compound is produced via metabolism in the human body after oral ingestion of diazinon, its presence in drinking water need not be monitored and regulated.

CyProduct: A Software Tool for Accurately Predicting the Byproducts of Human Cytochrome P450 Metabolism

In silico metabolism prediction is a cheminformatic task of autonomously predicting the set of metabolic byproducts produced from a specified molecule and a set of enzymes or reactions. Here, we describe a novel machine learned in silico cytochrome P450 (CYP450) metabolism prediction suite, called CyProduct, that accurately predicts metabolic byproducts for a specified molecule and a human CYP450 isoform. It includes three modules: (1) CypReact, a tool that predicts if the query compound reacts with a given CYP450 enzyme, (2) CypBoM, a tool that accurately predicts the "bond site" of the reaction (i.e., which specific bonds within the query molecule react with the CYP isoform), and (3) MetaboGen, a tool that generates the metabolic byproducts based on CypBoM's bond-site prediction. CyProduct predicts metabolic biotransformation products for each of the nine most important human CYP450 enzymes. CypBoM uses an important new concept called "bond of metabolism" (BoM), which extends the traditional "site of metabolism" (SoM) by specifying the information about the set of chemical bonds that is modified or formed in a metabolic reaction (rather than the specific atom). We created a BoM database for 1845 CYP450-mediated Phase I reactions, then used this to train the CypBoM Predictor to predict the reactive bond locations on substrate molecules. CypBoM Predictor's cross-validated Jaccard score for reactive bond prediction ranged from 0.380 to 0.452 over the nine CYP450 enzymes. Over variants of a test set of 68 known CYP450 substrates and 30 nonreactants, CyProduct outperformed the other packages, including ADMET Predictor, BioTransformer, and GLORY, by an average of 200% (with respect to Jaccard score) in terms of predicting metabolites. The CyProduct suite and the data sets are freely available at https://bitbucket.org/wishartlab/cyproduct/src/master/.

Design, synthesis, antibacterial evaluation, and computational studies of hybrid oxothiazolidin-1,2,4-triazole scaffolds

Bacterial infections are a serious threat to human health due to the development of resistance against the presently used antibiotics. The problem of growing and widespread antibiotic resistance is only getting worse with the shortage of new classes of antibiotics, creating a substantial unmet medical need in the treatment of serious bacterial infections. Therefore, in the present work, we report 18 novel hybrid thiazolidine-1,2,4-triazole derivatives as DNA gyrase inhibitors. The derivatives were synthesized by multistep organic synthesis and characterized by spectroscopic methods (¹ H and ¹³ C nuclear magnetic resonance and mass spectroscopy). The derivatives were tested for DNA gyrase inhibition, and the result emphasized that the synthesized derivatives have a tendency to inhibit the function of DNA gyrase. Furthermore, the compounds were also tested for antibacterial activity against three Gram-positive (Bacillus subtilis [NCIM 2063], Bacillus cereus [NCIM 2156], Staphylococcus aureus [NCIM 2079]) and two Gram-negative (Escherichia coli [NCIM 2065], Proteus vulgaris [NCIM 2027]) bacteria. The derivatives showed a significant-to-moderate antibacterial activity with noticeable antibiofilm efficacy. Quantitative structure-activity relationship (QSAR), ADME (absorption, distribution, metabolism, elimination) calculation, molecular docking, radial distribution function, and 2D fingerprinting were also performed to elucidate fundamental structural fragments essential for their bioactivity. These studies suggest that the derivatives 10b and 10n have lead antibacterial properties with significant DNA gyrase inhibitory efficacy, and they can serve as a starting scaffold for the further development of new broad-spectrum antibacterial agents.

Molecular modeling approaches to address drug-metabolizing enzymes (DMEs) mediated chemoresistance: a review

Resistance against clinically approved anticancer drugs is the main roadblock in cancer treatment. Drug metabolizing enzymes (DMEs) that are capable of metabolizing a variety of xenobiotic get overexpressed in malignant cells, therefore, catalyzing drug inactivation. As evident from the literature reports, the levels of DMEs increase in cancer cells that ultimately lead to drug inactivation followed by drug resistance. To puzzle out this issue, several strategies inclusive of analog designing, prodrug designing, and inhibitor designing have been forged. On that front, the implementation of computational tools can be considered a fascinating approach to address the problem of chemoresistance. Various research groups have adopted different molecular modeling tools for the investigation of DMEs mediated toxicity problems. However, the utilization of these in-silico tools in maneuvering the DME mediated chemoresistance is least considered and yet to be explored. These tools can be employed in the designing of such chemotherapeutic agents that are devoid of the resistance problem. The current review canvasses various molecular modeling approaches that can be implemented to address this issue. Special focus was laid on the development of specific inhibitors of DMEs. Additionally, the strategies to bypass the DMEs mediated drug metabolism were also contemplated in this report that includes analogs and pro-drugs designing. Different strategies discussed in the review will be beneficial in designing novel chemotherapeutic agents that depreciate the resistance problem.

Metabolic Forest: Predicting the Diverse Structures of Drug Metabolites

Adverse drug metabolism often severely impacts patient morbidity and mortality. Unfortunately, drug metabolism experimental assays are costly, inefficient, and slow. Instead, computational modeling could rapidly flag potentially toxic molecules across thousands of candidates in the early stages of drug development. Most metabolism models focus on predicting sites of metabolism (SOMs): the specific substrate atoms targeted by metabolic enzymes. However, SOMs are merely a proxy for metabolic structures: knowledge of an SOM does not explicitly provide the actual metabolite structure. Without an explicit metabolite structure, computational systems cannot evaluate the new molecule's properties. For example, the metabolite's reactivity cannot be automatically predicted, a crucial limitation because reactive drug metabolites are a key driver of adverse drug reactions (ADRs). Additionally, further metabolic events cannot be forecast, even though the metabolic path of the majority of substrates includes two or more sequential steps. To overcome the myopia of the SOM paradigm, this study constructs a well-defined system-termed the metabolic forest-for generating exact metabolite structures. We validate the metabolic forest with the substrate and product structures from a large, chemically diverse, literature-derived dataset of 20 736 records. The metabolic forest finds a pathway linking each substrate and product for 79.42% of these records. By performing a breadth-first search of depth two or three, we improve performance to 88.43 and 88.77%, respectively. The metabolic forest includes a specialized algorithm for producing accurate quinone structures, the most common type of reactive metabolite. To our knowledge, this quinone structure algorithm is the first of its kind, as the diverse mechanisms of quinone formation are difficult to systematically reproduce. We validate the metabolic forest on a previously published dataset of 576 quinone reactions, predicting their structures with a depth three performance of 91.84%. The metabolic forest accurately enumerates metabolite structures, enabling promising new directions such as joint metabolism and reactivity modeling.

Metabolic stability studies of lead compounds supported by separation techniques and chemometrics analysis

With metabolism being one of the main routes of drug elimination from the body (accounting for removal of around 75% of known drugs), it is crucial to understand and study metabolic stability of drug candidates. Metabolically unstable compounds are uncomfortable to administer (requiring repetitive dosage during therapy), while overly stable drugs increase risk of adverse drug reactions. Additionally, biotransformation reactions can lead to formation of toxic or pharmacologically active metabolites (either less-active than parent drug, or even with different action). There were numerous approaches in estimating metabolic stability, including in vitro, in vivo, in silico, and high-throughput screening to name a few. This review aims at describing separation techniques used in in vitro metabolic stability estimation, as well as chemometric techniques allowing for creation of predictive models which enable high-throughput screening approach for estimation of metabolic stability. With a very low rate of drug approval, it is important to understand in silico methods that aim at supporting classical in vitro approach. Predictive models that allow assessment of certain biological properties of drug candidates allow for cutting not only cost, but also time required to synthesize compounds predicted to be unstable or inactive by in silico models.

PreMetabo: An in silico phase I and II drug metabolism prediction platform

This study aimed to develop a drug metabolism prediction platform using knowledge-based prediction models. Site of Metabolism (SOM) prediction models for four cytochrome P450 (CYP) subtypes were developed along with uridine 5'-diphosphoglucuronosyltransferase (UGT) and sulfotransferase (SULT) substrate classification models. The SOM substrate for a certain CYP was determined using the sum of the activation energy required for the reaction at the reaction site of the substrate and the binding energy of the substrate to the CYP enzyme. Activation energy was calculated using the EaMEAD model and binding energy was calculated by docking simulation. Phase II prediction models were developed to predict whether a molecule is the substrate of a certain phase II conjugate protein, i.e., UGT or SULT. Using SOM prediction models, the predictability of the major metabolite in the top-3 was obtained as 72.5-84.5% for four CYPs, respectively. For internal validation, the accuracy of the UGT and SULT substrate classification model was obtained as 93.94% and 80.68%, respectively. Additionally, for external validation, the accuracy of the UGT substrate classification model was obtained as 81% in the case of 11 FDA-approved drugs. PreMetabo is implemented in a web environment and is available at https://premetabo.bmdrc.kr/.

Metatox - Web application for generation of metabolic pathways and toxicity estimation

Xenobiotics biotransformation in humans is a process of the chemical modifications, which may lead to the formation of toxic metabolites. The prediction of such metabolites is very important for drug development and ecotoxicology studies. We created the web-application MetaTox ( http://way2drug.com/mg ) for the generation of xenobiotics metabolic pathways in the human organism. For each generated metabolite, the estimations of the acute toxicity (based on GUSAR software prediction), organ-specific carcinogenicity and adverse effects (based on PASS software prediction) are performed. Generation of metabolites by MetaTox is based on the fragments datasets, which describe transformations of substrates structures to a metabolites structure. We added three new classes of biotransformation reactions: Dehydrogenation, Glutathionation, and Hydrolysis, and now metabolite generation for 15 most frequent classes of xenobiotic's biotransformation reactions are available. MetaTox calculates the probability of formation of generated metabolite - it is the integrated assessment of the biotransformation reactions probabilities and their sites using the algorithm of PASS ( http://way2drug.com/passonline ). The prediction accuracy estimated by the leave-one-out cross-validation (LOO-CV) procedure calculated separately for the probabilities of biotransformation reactions and their sites is about 0.9 on the average for all reactions.

Concepts of Artificial Intelligence for Computer-Assisted Drug Discovery

Artificial intelligence (AI), and, in particular, deep learning as a subcategory of AI, provides opportunities for the discovery and development of innovative drugs. Various machine learning approaches have recently (re)emerged, some of which may be considered instances of domain-specific AI which have been successfully employed for drug discovery and design. This review provides a comprehensive portrayal of these machine learning techniques and of their applications in medicinal chemistry. After introducing the basic principles, alongside some application notes, of the various machine learning algorithms, the current state-of-the art of AI-assisted pharmaceutical discovery is discussed, including applications in structure- and ligand-based virtual screening, de novo drug design, physicochemical and pharmacokinetic property prediction, drug repurposing, and related aspects. Finally, several challenges and limitations of the current methods are summarized, with a view to potential future directions for AI-assisted drug discovery and design.

Prediction of UGT-mediated Metabolism Using the Manually Curated MetaQSAR Database

Even though glucuronidations are the most frequent metabolic reactions of conjugation, both in quantitative and qualitative terms, they have rather seldom been investigated using computational approaches. To fill this gap, we have used the manually collected MetaQSAR metabolic reaction database to generate two models for the prediction of UGT-mediated metabolism, both based on molecular descriptors and implementing the Random Forest algorithm. The first model predicts the occurrence of the reaction and was internally validated with a Matthew correlation coefficient (MCC) of 0.76 and an area under the ROC curve (AUC) of 0.94, and further externally validated using a test set composed of 120 additional xenobiotics (MCC of 0.70 and AUC of 0.90). The second model distinguishes between O- and N-glucuronidations and was optimized by the random undersampling procedure to improve the predictive accuracy during the internal validation, with the recall measure of the minority class increasing from 0.55 to 0.78.

Predicting mammalian metabolism and toxicity of pesticides in silico

Pesticides must be effective to be commercially viable but they must also be reasonably safe for those who manufacture them, apply them, or consume the food they are used to produce. Animal testing is key to ensuring safety, but it comes late in the agrochemical development process, is expensive, and requires relatively large amounts of material. Surrogate assays used as in vitro models require less material and shift identification of potential mammalian toxicity back to earlier stages in development. Modern in silico methods are cost-effective complements to such in vitro models that make it possible to predict mammalian metabolism, toxicity and exposure for a pesticide, crop residue or other metabolite before it has been synthesized. Their broader use could substantially reduce the amount of time and effort wasted in pesticide development. This contribution reviews the kind of in silico models that are currently available for vetting ideas about what to synthesize and how to focus development efforts; the limitations of those models; and the practical considerations that have slowed development in the area. Detailed discussions are provided of how bacterial mutagenicity, human cytochrome P450 (CYP) metabolism, and bioavailability in humans and rats can be predicted. © 2018 The Authors. Pest Management Science published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

Using chemical bond-based method to predict site of metabolism for five biotransformations mediated by CYP 3A4, 2D6, and 2C9

Although it has been proposed for decades to predict site of metabolism (SOM) by in silico methods, identifying SOM correctly remains an unsolved fundamental problem and is an active area of research. In our prior works, we proposed a chemical bond-based approach to construction of SOM prediction models by integrating chemical bond descriptors and drug-metabolizing enzymes data. Although it has been evaluated with both 10-fold cross-validation and independent validation, we believe comparisons between this method and prior methods using publicly accessible external datasets are indispensable and more desirable. In the current study, based on chemical bond-based method, metabolism data released by Sheridan et al. and Zaretzki et al. was utilized to establish metabolite prediction models for CYP450 3A4, 2D6, and 2C9. Five major reaction types were involved, including Aliphatic C-hydroxylation, Aromatic C-hydroxylation, N-dealkylation, O-dealkylation, and S-Oxidation. Consequently, all our five models showed impressive performance on predicting SOMs, with accuracy and area under curve exceeded 0.940 and 0.953, respectively. Compared to prior works, our models were better than SOMP both in "SOM-scale" and "molecule-scale". In conclusion, comparisons between chemical-bond based method and prior works were conducted for the first time, which demonstrated that chemical-bond based method is better than or at least comparable to prior works.

QNA-Based Prediction of Sites of Metabolism

Metabolism of xenobiotics (Greek xenos: exogenous substances) plays an essential role in the prediction of biological activity and testing for the subsequent research and development of new drug candidates. Integration of various methods and techniques using different computational and experimental approaches is one of the keys to a successful metabolism prediction. While multiple structure-based and ligand-based approaches to metabolism prediction exist, the most important problem arises at the first stage of metabolism prediction: detection of the sites of metabolism (SOMs). In this paper, we describe the application of Quantitative Neighborhoods of Atoms (QNA) descriptors for prediction of the SOMs using potential function method, as well as several different machine learning techniques: naïve Bayes, random forest classifier, multilayer perceptron with back propagation and convolutional neural networks, and deep neural networks.

Prediction of metabolites of epoxidation reaction in MetaTox

Biotransformation is a process of the chemical modifications which may lead to the reactive metabolites, in particular the epoxides. Epoxide reactive metabolites may cause the toxic effects. The prediction of such metabolites is important for drug development and ecotoxicology studies. Epoxides are formed by some oxidation reactions, usually catalysed by cytochromes P450, and represent a large class of three-membered cyclic ethers. Identification of molecules, which may be epoxidized, and indication of the specific location of epoxide functional group (which is called SOE - site of epoxidation) are important for prediction of epoxide metabolites. Datasets from 355 molecules and 615 reactions were created for training and validation. The prediction of SOE is based on a combination of LMNA (Labelled Multilevel Neighbourhood of Atom) descriptors and Bayesian-like algorithm implemented in PASS software and MetaTox web-service. The average invariant accuracy of prediction (AUC) calculated in leave-one-out and 20-fold cross-validation procedures is 0.9. Prediction of epoxide formation based on the created SAR model is included as the component of MetaTox web-service ( http://www.way2drug.com/mg ).

RD-Metabolizer: an integrated and reaction types extensive approach to predict metabolic sites and metabolites of drug-like molecules

Background

Experimental approaches for determining the metabolic properties of the drug candidates are usually expensive, time-consuming and labor intensive. There is a great deal of interest in developing computational methods to accurately and efficiently predict the metabolic decomposition of drug-like molecules, which can provide decisive support and guidance for experimentalists.

Results

Here, we developed an integrated, low false positive and reaction types extensive metabolism prediction approach called RD-Metabolizer (Reaction Database-based Metabolizer). RD-Metabolizer firstly employed the detailed reaction SMARTS patterns to encode different metabolism reaction types with the aim of covering larger chemical reaction space. 2D fingerprint similarity calculation model was built to calculate the metabolic probability of each site in a molecule. RDKit was utilized to act on pre-written reaction SMARTS patterns to correct the metabolic ranking of each site in a molecule generated by the 2D fingerprint similarity calculation model as well as generate corresponding structures of metabolites, thus helping to reduce the false positive metabolites. Two test sets were adopted to evaluate the performance of RD-Metabolizer in predicting SOMs and structures of metabolites. The results indicated that RD-Metabolizer was better than or at least as good as several widely used SOMs prediction methods. Besides, the number of false positive metabolites was obviously reduced compared with MetaPrint2D-React.

Conclusions

The accuracy and efficiency of RD-Metabolizer was further illustrated by a metabolism prediction case of AZD9291, which is a mutant-selective EGFR inhibitor. RD-Metabolizer will serve as a useful toolkit for the early metabolic properties assessment of drug-like molecules at the preclinical stage of drug discovery.Graphical abstract

Deep Learning to Predict the Formation of Quinone Species in Drug Metabolism

Many adverse drug reactions are thought to be caused by electrophilically reactive drug metabolites that conjugate to nucleophilic sites within DNA and proteins, causing cancer or toxic immune responses. Quinone species, including quinone-imines, quinone-methides, and imine-methides, are electrophilic Michael acceptors that are often highly reactive and comprise over 40% of all known reactive metabolites. Quinone metabolites are created by cytochromes P450 and peroxidases. For example, cytochromes P450 oxidize acetaminophen to N-acetyl-p-benzoquinone imine, which is electrophilically reactive and covalently binds to nucleophilic sites within proteins. This reactive quinone metabolite elicits a toxic immune response when acetaminophen exceeds a safe dose. Using a deep learning approach, this study reports the first published method for predicting quinone formation: the formation of a quinone species by metabolic oxidation. We model both one- and two-step quinone formation, enabling accurate quinone formation predictions in nonobvious cases. We predict atom pairs that form quinones with an AUC accuracy of 97.6%, and we identify molecules that form quinones with 88.2% AUC. By modeling the formation of quinones, one of the most common types of reactive metabolites, our method provides a rapid screening tool for a key drug toxicity risk. The XenoSite quinone formation model is available at http://swami.wustl.edu/xenosite/p/quinone .

Predicting the Metabolic Sites by Flavin-Containing Monooxygenase on Drug Molecules Using SVM Classification on Computed Quantum Mechanics and Circular Fingerprints Molecular Descriptors

As an important enzyme in Phase I drug metabolism, the flavin-containing monooxygenase (FMO) also metabolizes some xenobiotics with soft nucleophiles. The site of metabolism (SOM) on a molecule is the site where the metabolic reaction is exerted by an enzyme. Accurate prediction of SOMs on drug molecules will assist the search for drug leads during the optimization process. Here, some quantum mechanics features such as the condensed Fukui function and attributes from circular fingerprints (called Molprint2D) are computed and classified using the support vector machine (SVM) for predicting some potential SOMs on a series of drugs that can be metabolized by FMO enzymes. The condensed Fukui function fA- representing the nucleophilicity of central atom A and the attributes from circular fingerprints accounting the influence of neighbors on the central atom. The total number of FMO substrates and non-substrates collected in the study is 85 and they are equally divided into the training and test sets with each carrying roughly the same number of potential SOMs. However, only N-oxidation and S-oxidation features were considered in the prediction since the available C-oxidation data was scarce. In the training process, the LibSVM package of WEKA package and the option of 10-fold cross validation are employed. The prediction performance on the test set evaluated by accuracy, Matthews correlation coefficient and area under ROC curve computed are 0.829, 0.659, and 0.877 respectively. This work reveals that the SVM model built can accurately predict the potential SOMs for drug molecules that are metabolizable by the FMO enzymes.

Predicting Regioselectivity and Lability of Cytochrome P450 Metabolism Using Quantum Mechanical Simulations

We describe methods for predicting cytochrome P450 (CYP) metabolism incorporating both pathway-specific reactivity and isoform-specific accessibility considerations. Semiempirical quantum mechanical (QM) simulations, parametrized using experimental data and ab initio calculations, estimate the reactivity of each potential site of metabolism (SOM) in the context of the whole molecule. Ligand-based models, trained using high-quality regioselectivity data, correct for orientation and steric effects of the different CYP isoform binding pockets. The resulting models identify a SOM in the top 2 predictions for between 82% and 91% of compounds in independent test sets across seven CYP isoforms. In addition to predicting the relative proportion of metabolite formation at each site, these methods estimate the activation energy at each site, from which additional information can be derived regarding their lability in absolute terms. We illustrate how this can guide the design of compounds to overcome issues with rapid CYP metabolism.

Predicting the binding modes and sites of metabolism of xenobiotics

Metabolism studies are an essential integral part of ADMET profiling of drug candidates to evaluate their safety and efficacy. Cytochrome P-450 (CYP) metabolizes a wide variety of xenobiotics/drugs. The binding modes of these compounds with CYP and their intrinsic reactivities decide the metabolic products. We report here a novel computational protocol, which comprises docking of ligands to heme-containing CYPs and prediction of binding energies through a newly developed scoring function, followed by analyses of the docked structures and molecular orbitals of the ligand molecules, for predicting the sites of metabolism (SOM) of ligands. The calculated binding free energies of 121 heme-containing protein-ligand docked complexes yielded a correlation coefficient of 0.84 against experiment. Molecular orbital analyses of the resultant top three unique poses of the docked complexes provided a success rate of 87% in identifying the experimentally known sites of metabolism of the xenobiotics. The SOM prediction methodology is freely accessible at .

Effects of protein flexibility and active site water molecules on the prediction of sites of metabolism for cytochrome P450 2C19 substrates

Structure-based prediction of sites of metabolism (SOMs) mediated by cytochrome P450s (CYPs) is of great interest in drug discovery and development.

A Mechanism-Based Model for the Prediction of the Metabolic Sites of Steroids Mediated by Cytochrome P450 3A4

Early prediction of xenobiotic metabolism is essential for drug discovery and development. As the most important human drug-metabolizing enzyme, cytochrome P450 3A4 has a large active cavity and metabolizes a broad spectrum of substrates. The poor substrate specificity of CYP3A4 makes it a huge challenge to predict the metabolic site(s) on its substrates. This study aimed to develop a mechanism-based prediction model based on two key parameters, including the binding conformation and the reaction activity of ligands, which could reveal the process of real metabolic reaction(s) and the site(s) of modification. The newly established model was applied to predict the metabolic site(s) of steroids; a class of CYP3A4-preferred substrates. 38 steroids and 12 non-steroids were randomly divided into training and test sets. Two major metabolic reactions, including aliphatic hydroxylation and N-dealkylation, were involved in this study. At least one of the top three predicted metabolic sites was validated by the experimental data. The overall accuracy for the training and test were 82.14% and 86.36%, respectively. In summary, a mechanism-based prediction model was established for the first time, which could be used to predict the metabolic site(s) of CYP3A4 on steroids with high predictive accuracy.

Prediction of cytochrome P450 mediated metabolism

Cytochrome P450 enzymes (CYPs) form one of the most important enzyme families involved in the metabolism of xenobiotics. CYPs comprise many isoforms, which catalyze a wide variety of reactions, and potentially, a large number of different metabolites can be formed. However, it is often hard to rationalize what metabolites these enzymes generate. In recent years, many different in silico approaches have been developed to predict binding or regioselective product formation for the different CYP isoforms. These comprise ligand-based methods that are trained on experimental CYP data and structure-based methods that consider how the substrate is oriented in the active site or/and how reactive the part of the substrate that is accessible to the heme group is. We will review key aspects for various approaches that are available to predict binding and site of metabolism (SOM), what outcome can be expected from the predictions, and how they could potentially be improved.

Modeling of interactions between xenobiotics and cytochrome P450 (CYP) enzymes

The adverse effects to humans and environment of only few chemicals are well known. Absorption, distribution, metabolism, and excretion (ADME) are the steps of pharmaco/toxicokinetics that determine the internal dose of chemicals to which the organism is exposed. Of all the xenobiotic-metabolizing enzymes, the cytochrome P450 (CYP) enzymes are the most important due to their abundance and versatility. Reactions catalyzed by CYPs usually turn xenobiotics to harmless and excretable metabolites, but sometimes an innocuous xenobiotic is transformed into a toxic metabolite. Data on ADME and toxicity properties of compounds are increasingly generated using in vitro and modeling (in silico) tools. Both physics-based and empirical modeling approaches are used. Numerous ligand-based and target-based as well as combined modeling methods have been employed to evaluate determinants of CYP ligand binding as well as predicting sites of metabolism and inhibition characteristics of test molecules. In silico prediction of CYP–ligand interactions have made crucial contributions in understanding (1) determinants of CYP ligand binding recognition and affinity; (2) prediction of likely metabolites from substrates; (3) prediction of inhibitors and their inhibition potency. Truly predictive models of toxic outcomes cannot be created without incorporating metabolic characteristics; in silico methods help producing such information and filling gaps in experimentally derived data. Currently modeling methods are not mature enough to replace standard in vitro and in vivo approaches, but they are already used as an important component in risk assessment of drugs and other chemicals.

Predicting drug metabolism: experiment and/or computation?

Drug metabolism can produce metabolites with physicochemical and pharmacological properties that differ substantially from those of the parent drug, and consequently has important implications for both drug safety and efficacy. To reduce the risk of costly clinical-stage attrition due to the metabolic characteristics of drug candidates, there is a need for efficient and reliable ways to predict drug metabolism in vitro, in silico and in vivo. In this Perspective, we provide an overview of the state of the art of experimental and computational approaches for investigating drug metabolism. We highlight the scope and limitations of these methods, and indicate strategies to harvest the synergies that result from combining measurement and prediction of drug metabolism.

Prediction of overall in vitro microsomal stability of drug candidates based on molecular modeling and support vector machines. Case study of novel arylpiperazines derivatives

Other than efficacy of interaction with the molecular target, metabolic stability is the primary factor responsible for the failure or success of a compound in the drug development pipeline. The ideal drug candidate should be stable enough to reach its therapeutic site of action. Despite many recent excellent achievements in the field of computational methods supporting drug metabolism studies, a well-recognized procedure to model and predict metabolic stability quantitatively is still lacking. This study proposes a workflow for developing quantitative metabolic stability-structure relationships, taking a set of 30 arylpiperazine derivatives as an example. The metabolic stability of the compounds was assessed in in vitro incubations in the presence of human liver microsomes and NADPH and subsequently quantified by liquid chromatography-mass spectrometry (LC-MS). Density functional theory (DFT) calculations were used to obtain 30 models of the molecules, and Dragon software served as a source of structure-based molecular descriptors. For modeling structure-metabolic stability relationships, Support Vector Machines (SVM), a non-linear machine learning technique, were found to be more effective than a regression technique, based on the validation parameters obtained. Moreover, for the first time, general sites of metabolism for arylpiperazines bearing the 4-aryl-2H-pyrido[1,2-c]pyrimidine-1,3-dione system were defined by analysis of Q-TOF-MS/MS spectra. The results indicated that the application of one of the most advanced chemometric techniques combined with a simple and quick in vitro procedure and LC-MS analysis provides a novel and valuable tool for predicting metabolic half-life values. Given the reduced time and simplicity of analysis, together with the accuracy of the predictions obtained, this is a valid approach for predicting metabolic stability using structural data. The approach presented provides a novel, comprehensive and reliable tool for investigating metabolic stability, factors that affect it, and the proposed structures of metabolites at the same time. The performance of the DFT-SVM-based approach provides an opportunity to implement it in a standard drug development pipeline.

Combining structure- and ligand-based approaches to improve site of metabolism prediction in CYP2C9 substrates

PURPOSE

Predicting atoms in a potential drug compound that are susceptible to oxidation by cytochrome P450 (CYP) enzymes is of great interest to the pharmaceutical community. We aimed to develop a computational approach combining ligand- and structure-based design principles to accurately predict sites of metabolism (SoMs) in a series of CYP2C9 substrates.

METHODS

We employed the reactivity model, SMARTCyp, ensemble docking, and pseudo-receptor modeling based on quantitative structure-activity relationships (QSAR) to account for influences of both the inherent reactivity of each atom and the physical structure of the CYP2C9 binding site.

RESULTS

We tested ligand-based prediction alone (i.e. SMARTCyp), structure-based prediction alone (i.e. AutoDock Vina docking), the linear combination of the SMARTCYP and docking scores, and finally a pseudo-receptor QSAR model based on the docked compounds in combination with SMARTCyp. We found that by using the latter combined approach we were able to accurately predict 88% and 96% of the true SoMs, within the top-1 and top-2 predictions, respectively.

CONCLUSIONS

We have outlined a novel combination approach for accurately predicting SoMs in CYP2C9 ligands. We believe that this method may be applied to other CYP2C9 ligands as well as to other CYP systems.

Cytochrome P450 site of metabolism prediction from 2D topological fingerprints using GPU accelerated probabilistic classifiers

Background

The prediction of sites and products of metabolism in xenobiotic compounds is key to the development of new chemical entities, where screening potential metabolites for toxicity or unwanted side-effects is of crucial importance. In this work 2D topological fingerprints are used to encode atomic sites and three probabilistic machine learning methods are applied: Parzen-Rosenblatt Window (PRW), Naive Bayesian (NB) and a novel approach called RASCAL (Random Attribute Subsampling Classification ALgorithm). These are implemented by randomly subsampling descriptor space to alleviate the problem often suffered by data mining methods of having to exactly match fingerprints, and in the case of PRW by measuring a distance between feature vectors rather than exact matching. The classifiers have been implemented in CUDA/C++ to exploit the parallel architecture of graphical processing units (GPUs) and is freely available in a public repository.

Results

It is shown that for PRW a SoM (Site of Metabolism) is identified in the top two predictions for 85%, 91% and 88% of the CYP 3A4, 2D6 and 2C9 data sets respectively, with RASCAL giving similar performance of 83%, 91% and 88%, respectively. These results put PRW and RASCAL performance ahead of NB which gave a much lower classification performance of 51%, 73% and 74%, respectively.

Conclusions

2D topological fingerprints calculated to a bond depth of 4-6 contain sufficient information to allow the identification of SoMs using classifiers based on relatively small data sets. Thus, the machine learning methods outlined in this paper are conceptually simpler and more efficient than other methods tested and the use of simple topological descriptors derived from 2D structure give results competitive with other approaches using more expensive quantum chemical descriptors. The descriptor space subsampling approach and ensemble methodology allow the methods to be applied to molecules more distant from the training data where data mining would be more likely to fail due to the lack of common fingerprints. The RASCAL algorithm is shown to give equivalent classification performance to PRW but at lower computational expense allowing it to be applied more efficiently in the ensemble scheme.

Exposition and reactivity optimization to predict sites of metabolism in chemicals

Chemical modifications of drugs induced by phase I biotransformations significantly affect their pharmacokinetic properties. Because the metabolites produced can themselves have a pharmacological effect and an intrinsic toxicity, medicinal chemists need to accurately predict the sites of metabolism (SoM) of drugs as early as possible. However, site of metabolism prediction is rarely accompanied by a prediction of the relative abundance of the various metabolites. Such a prediction would be a great help in the study of drug– drug interactions and in the process of reducing the toxicity of potential drug candidates. The aim of this paper is to present recent developments in the prediction of xenobiotic metabolism and to use concrete examples to explain the computational mechanism employed.

Modelling cytochromes P450 binding modes to predict P450 inhibition, metabolic stability and isoform selectivity

The cytochromes P450 (P450) superfamily is a diverse group of enzymes involved in the metabolism of xenobiotics, whose orientations within the catalytic site can lead to different binding modes, namely productive, nonproductive, and inhibitory. This article collects the most recent approaches that individually study P450- ligand interactions, including a novel in silico technology, developed in the framework of the Human Cytochrome P450 Consortium initiative, that provides reliable in silico predictions of P450 inhibition, metabolic stability and isoform selectivity.

Metabolism site prediction based on xenobiotic structural formulas and PASS prediction algorithm

A new ligand-based method for the prediction of sites of metabolism (SOMs) for xenobiotics has been developed on the basis of the LMNA (labeled multilevel neighborhoods of atom) descriptors and the PASS (prediction of activity spectra for substances) algorithm and applied to predict the SOMs of the 1A2, 2C9, 2C19, 2D6, and 3A4 isoforms of cytochrome P450. An average IAP (invariant accuracy of prediction) of SOMs calculated by the leave-one-out cross-validation procedure was 0.89 for the developed method. The external validation was made with evaluation sets containing data on biotransformations for 57 cardiovascular drugs. An average IAP of regioselectivity for evaluation sets was 0.83. It was shown that the proposed method exceeds accuracy of SOM prediction by RS-Predictor for CYP 1A2, 2D6, 2C9, 2C19, and 3A4 and is comparable to or better than SMARTCyp for CYP 2C9 and 2D6.