Computational prediction of human drug metabolism

An investigation of binding interactions of tumor-targeted peptide conjugated polyphenols with the kinase domain of ephrin B4 and B2 receptors

Recent studies have shown that Ephrin receptors may be upregulated in several types of cancers including breast, ovarian and endometrial cancers, making them a target for drug design. In this work, we have utilized a target-hopping approach to design new natural product-peptide conjugates and examined their interactions with the kinase-binding domain of EphB4 and EphB2 receptors. The peptide sequences were generated through point mutations of the known EphB4 antagonist peptide TNYLFSPNGPIA. Their anticancer properties and secondary structures were analyzed computationally. Conjugates of most optimum of peptides were then designed by binding the N-terminal of the peptides with the free carboxyl group of the polyphenols sinapate, gallate and coumarate, which are known for their inherent anticancer properties. To investigate if these conjugates have a potential to bind to the kinase domain, we carried out docking studies and MMGBSA free energy calculations of the trajectories based on the molecular dynamics simulations, with both the apo and the ATP bound kinase domains of both receptors. In most cases binding interactions occurred within the catalytic loop region, while in some cases the conjugates were found to spread out across the N-lobe and the DFG motif region. The conjugates were further tested for prediction of pharmacokinetic properties using ADME studies. Our results indicated that the conjugates were lipophilic and MDCK permeable with no CYP interactions. These findings provide an insight into the molecular interactions of these peptides and conjugates with the kinase domain of the EphB4 and EphB2 receptor. As a proof of concept, we synthesized and carried out SPR analysis with two of the conjugates (gallate-TNYLFSPNGPIA and sinapate-TNYLFSPNGPIA). Results indicated that the conjugates showed higher binding with the EphB4 receptor and minimal binding to EphB2 receptor. Sinapate-TNYLFSPNGPIA showed inhibitory activity against EphB4. These studies reveal that some of the conjugates may be developed for further investigation into in vitro and in vivo studies and potential development as therapeutics.

Improving ADMET Prediction Accuracy for Candidate Drugs: Factors to Consider in QSPR Modeling Approaches

Quantitative Structure-Property Relationship (QSPR) employs mathematical and statistical methods to reveal quantitative correlations between the pharmacokinetics of compounds and their molecular structures, as well as their physical and chemical properties. QSPR models have been widely applied in the prediction of drug absorption, distribution, metabolism, excretion, and toxicity (ADMET). However, the accuracy of QSPR models for predicting drug ADMET properties still needs improvement. Therefore, this paper comprehensively reviews the tools employed in various stages of QSPR predictions for drug ADMET. It summarizes commonly used approaches to building QSPR models, systematically analyzing the advantages and limitations of each modeling method to ensure their judicious application. We provide an overview of recent advancements in the application of QSPR models for predicting drug ADMET properties. Furthermore, this review explores the inherent challenges in QSPR modeling while also proposing a range of considerations aimed at enhancing model prediction accuracy. The objective is to enhance the predictive capabilities of QSPR models in the field of drug development and provide valuable reference and guidance for researchers in this domain.

Machine Learning Models Identify New Inhibitors for Human OATP1B1

The uptake transporter OATP1B1 (SLC01B1) is largely localized to the sinusoidal membrane of hepatocytes and is a known victim of unwanted drug-drug interactions. Computational models are useful for identifying potential substrates and/or inhibitors of clinically relevant transporters. Our goal was to generate OATP1B1 in vitro inhibition data for [3H] estrone-3-sulfate (E3S) transport in CHO cells and use it to build machine learning models to facilitate a comparison of seven different classification models (Deep learning, Adaboosted decision trees, Bernoulli naïve bayes, k-nearest neighbors (knn), random forest, support vector classifier (SVC), logistic regression (lreg), and XGBoost (xgb)] using ECFP6 fingerprints to perform 5-fold, nested cross validation. In addition, we compared models using 3D pharmacophores, simple chemical descriptors alone or plus ECFP6, as well as ECFP4 and ECFP8 fingerprints. Several machine learning algorithms (SVC, lreg, xgb, and knn) had excellent nested cross validation statistics, particularly for accuracy, AUC, and specificity. An external test set containing 207 unique compounds not in the training set demonstrated that at every threshold SVC outperformed the other algorithms based on a rank normalized score. A prospective validation test set was chosen using prediction scores from the SVC models with ECFP fingerprints and were tested in vitro with 15 of 19 compounds (84% accuracy) predicted as active (≥20% inhibition) showed inhibition. Of these compounds, six (abamectin, asiaticoside, berbamine, doramectin, mobocertinib, and umbralisib) appear to be novel inhibitors of OATP1B1 not previously reported. These validated machine learning models can now be used to make predictions for drug-drug interactions for human OATP1B1 alongside other machine learning models for important drug transporters in our MegaTrans software.

Consensus Prediction of Human Gut Microbiota-Mediated Metabolism Susceptibility for Small Molecules by Machine Learning, Structural Alerts, and Dietary Compounds-Based Average Similarity Methods

The human gut microbiota (HGM) colonizing human gastrointestinal tract (HGT) confers a repertoire of dynamic and unique metabolic capacities that are not possessed by the host and therefore is tentatively perceived as an alternative metabolic ″organ″ besides the liver in the host. Nevertheless, the significant contribution of HGM to the overall human metabolism is often overlooked in the modern drug discovery pipeline. Hence, a systematic evaluation of HGM-mediated drug metabolism is gradually important, and its computational prediction becomes increasingly necessary. In this work, a new data set containing both the HGM-mediated metabolism susceptible (HGMMS) and insusceptible (HGMMI) compounds (329 vs 320) was manually curated. Based on this data set, the first machine learning (ML) model, a new structural alerts (SA) model, and the K-nearest neighboring dietary compounds-based average similarity (AS) model were proposed to directly predict the HGM-mediated metabolism susceptibility for small molecules, and exhibit promising performance on three independent test sets. Finally, consensus prediction (ML/SA/AS) for DrugBank molecules revealed an intriguing phenomenon that a typical Michael acceptor ″α,β-unsaturated carbonyl group″ is a very common warhead for the design of covalent inhibitors and inclined to be metabolized by HGM in anaerobic HGT to generate the reduced metabolite without the reactive warhead, which could be a new concern to medicinal chemists. To the best of our knowledge, we gleaned the first HGMMS/HGMMI data set, developed the first HGMMS/HGMMI classification model, implemented a relatively comprehensive program based on ML/SA/AS approaches, and found a new phenomenon on the HGM-mediated deactivation of an extensively used warhead for covalent inhibitors.

Roles of cytochrome P450 enzymes in pharmacology and toxicology: Past, present, and future

The development of the cytochrome P450 (P450) field has been remarkable in the areas of pharmacology and toxicology, particularly in drug development. Today it is possible to use the knowledge base and relatively straightforward assays to make intelligent predictions about drug disposition prior to human dosing. Much is known about the structures, regulation, chemistry of catalysis, and the substrate and inhibitor specificity of human P450s. Many aspects of drug-drug interactions and side effects can be understood in terms of P450s. This knowledge has also been useful in pharmacy practice, as well as in the pharmaceutical industry and medical practice. However, there are still basic and practical questions to address regarding P450s and their roles in pharmacology and toxicology. Another aspect is the discovery of drugs that inhibit P450 to treat diseases.

Deep Learning Based Drug Metabolites Prediction

Drug metabolism research plays a key role in the discovery and development of drugs. Based on the discovery of drug metabolites, new chemical entities can be identified and potential safety hazards caused by reactive or toxic metabolites can be minimized. Nowadays, computational methods are usually complementary tools for experiments. However, current metabolites prediction methods tend to have high false positive rates with low accuracy and are usually only used for specific enzyme systems. In order to overcome this difficulty, a method was developed in this paper by first establishing a database with broad coverage of SMARTS-coded metabolic reaction rule, and then extracting the molecular fingerprints of compounds to construct a classification model based on deep learning algorithms. The metabolic reaction rule database we built can supplement chemically reasonable negative reaction examples. Based on deep learning algorithms, the model could determine which reaction types are more likely to occur than the others. In the test set, our method can achieve the accuracy of 70% (Top-10), which is significantly higher than that of random guess and the rule-based method SyGMa. The results demonstrated that our method has a certain predictive ability and application value.

Virtual screening, Docking, ADMET and System Pharmacology studies on Garcinia caged Xanthone derivatives for Anticancer activity

Caged xanthones are bioactive compounds mainly derived from the Garcinia genus. In this study, a structure-activity relationship (SAR) of caged xanthones and their derivatives for anticancer activity against different cancer cell lines such as A549, HepG2 and U251 were developed through quantitative (Q)-SAR modeling approach. The regression coefficient (r²), internal cross-validation regression coefficient (q²) and external cross-validation regression coefficient (pred_r²) of derived QSAR models were 0.87, 0.81 and 0.82, for A549, whereas, 0.87, 0.84 and 0.90, for HepG2, and 0.86, 0.83 and 0.83, for U251 respectively. These models were used to design and screened the potential caged xanthone derivatives. Further, the compounds were filtered through the rule of five, ADMET-risk and synthetic accessibility. Filtered compounds were then docked to identify the possible target binding pocket, to obtain a set of aligned ligand poses and to prioritize the predicted active compounds. The scrutinized compounds, as well as their metabolites, were evaluated for different pharmacokinetics parameters such as absorption, distribution, metabolism, excretion, and toxicity. Finally, the top hit compound 1G was analyzed by system pharmacology approaches such as gene ontology, metabolic networks, process networks, drug target network, signaling pathway maps as well as identification of off-target proteins that may cause adverse reactions.

Comparison of Deep Learning With Multiple Machine Learning Methods and Metrics Using Diverse Drug Discovery Data Sets

Machine learning methods have been applied to many data sets in pharmaceutical research for several decades. The relative ease and availability of fingerprint type molecular descriptors paired with Bayesian methods resulted in the widespread use of this approach for a diverse array of end points relevant to drug discovery. Deep learning is the latest machine learning algorithm attracting attention for many of pharmaceutical applications from docking to virtual screening. Deep learning is based on an artificial neural network with multiple hidden layers and has found considerable traction for many artificial intelligence applications. We have previously suggested the need for a comparison of different machine learning methods with deep learning across an array of varying data sets that is applicable to pharmaceutical research. End points relevant to pharmaceutical research include absorption, distribution, metabolism, excretion, and toxicity (ADME/Tox) properties, as well as activity against pathogens and drug discovery data sets. In this study, we have used data sets for solubility, probe-likeness, hERG, KCNQ1, bubonic plague, Chagas, tuberculosis, and malaria to compare different machine learning methods using FCFP6 fingerprints. These data sets represent whole cell screens, individual proteins, physicochemical properties as well as a data set with a complex end point. Our aim was to assess whether deep learning offered any improvement in testing when assessed using an array of metrics including AUC, F1 score, Cohen's kappa, Matthews correlation coefficient and others. Based on ranked normalized scores for the metrics or data sets Deep Neural Networks (DNN) ranked higher than SVM, which in turn was ranked higher than all the other machine learning methods. Visualizing these properties for training and test sets using radar type plots indicates when models are inferior or perhaps over trained. These results also suggest the need for assessing deep learning further using multiple metrics with much larger scale comparisons, prospective testing as well as assessment of different fingerprints and DNN architectures beyond those used.

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.

Prediction of pharmacokinetic and toxicological parameters of a 4-phenylcoumarin isolated from geopropolis: In silico and in vitro approaches

In silico and in vitro methodologies have been used as important tools in the drug discovery process, including from natural sources. The aim of this study was to predict pharmacokinetic and toxicity (ADME/Tox) properties of a coumarin isolated from geopropolis using in silico and in vitro approaches. Cinnamoyloxy-mammeisin (CNM) isolated from Brazilian M. scutellaris geopropolis was evaluated for its pharmacokinetic parameters by in silico models (ACD/Percepta™ and MetaDrug™ software). Genotoxicity was assessed by in vitro DNA damage signaling PCR array. CNM did not pass all parameters of Lipinski's rule of five, with a predicted low oral bioavailability and high plasma protein binding, but with good predicted blood brain barrier penetration. CNM was predicted to show low affinity to cytochrome P450 family members. Furthermore, the predicted Ames test indicated potential mutagenicity of CNM. Also, the probability of toxicity for organs and tissues was classified as moderate and high for liver and kidney, and moderate and low for skin and eye irritation, respectively. The PCR array analysis showed that CNM significantly upregulated about 7% of all DNA damage-related genes. By exploring the biological function of these genes, it was found that the predicted CNM genotoxicity is likely to be mediated by apoptosis. The predicted ADME/Tox profile suggests that external use of CNM may be preferable to systemic exposure, while its genotoxicity was characterized by the upregulation of apoptosis-related genes after treatment. The combined use of in silico and in vitro approaches to evaluate these parameters generated useful hypotheses to guide further preclinical studies.

Antihypertensive drugs metabolism: an update to pharmacokinetic profiles and computational approaches

Drug discovery and development is a high-risk enterprise that requires significant investments in capital, time and scientific expertise. The studies of xenobiotic metabolism remain as one of the main topics in the research and development of drugs, cosmetics and nutritional supplements.

Antihypertensive drugs are used for the treatment of high blood pressure, which is one the most frequent symptoms of the patients that undergo cardiovascular diseases such as myocardial infraction and strokes. In current cardiovascular disease pharmacology, four drug clusters - Angiotensin Converting Enzyme Inhibitors, Beta-Blockers, Calcium Channel Blockers and Diuretics - cover the major therapeutic characteristics of the most antihypertensive drugs. The pharmacokinetic and specifically the metabolic profile of the antihypertensive agents are intensively studied because of the broad inter-individual variability on plasma concentrations and the diversity on the efficacy response especially due to the P450 dependent metabolic status they present.

Several computational methods have been developed with the aim to: (i) model and better understand the human drug metabolism; and (ii) enhance the experimental investigation of the metabolism of small xenobiotic molecules. The main predictive tools these methods employ are rule-based approaches, quantitative structure metabolism/activity relationships and docking approaches.

This review paper provides detailed metabolic profiles of the major clusters of antihypertensive agents, including their metabolites and their metabolizing enzymes, and it also provides specific information concerning the computational approaches that have been used to predict the metabolic profile of several antihypertensive drugs.

Bigger data, collaborative tools and the future of predictive drug discovery

Over the past decade we have seen a growth in the provision of chemistry data and cheminformatics tools as either free websites or software as a service commercial offerings. These have transformed how we find molecule-related data and use such tools in our research. There have also been efforts to improve collaboration between researchers either openly or through secure transactions using commercial tools. A major challenge in the future will be how such databases and software approaches handle larger amounts of data as it accumulates from high throughput screening and enables the user to draw insights, enable predictions and move projects forward. We now discuss how information from some drug discovery datasets can be made more accessible and how privacy of data should not overwhelm the desire to share it at an appropriate time with collaborators. We also discuss additional software tools that could be made available and provide our thoughts on the future of predictive drug discovery in this age of big data. We use some examples from our own research on neglected diseases, collaborations, mobile apps and algorithm development to illustrate these ideas.

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.

Estimating screening-level organic chemical half-lives in humans

Relatively few measured data are available for the thousands of chemicals requiring hazard and risk assessment. The whole body, total elimination half-life (HLT) and the whole body, primary biotransformation half-life (HLB) are key parameters determining the extent of bioaccumulation, biological concentration, and risk from chemical exposure. A one-compartment pharmacokinetic (1-CoPK) mass balance model was developed to estimate organic chemical HLB from measured HLT data in mammals. Approximately 1900 HLs for human adults were collected and reviewed and the 1-CoPK model was parametrized for an adult human to calculate HLB from HLT. Measured renal clearance and whole body total clearance data for 306 chemicals were used to calculate empirical HLB,emp. The HLB,emp values and other measured data were used to corroborate the 1-CoPK HLB model calculations. HLs span approximately 7.5 orders of magnitude from 0.05 h for nitroglycerin to 2 × 10(6) h for 2,3,4,5,2',3',5',6'-octachlorobiphenyl with a median of 7.6 h. The automated Iterative Fragment Selection (IFS) method was applied to develop and evaluate various quantitative structure-activity relationships (QSARs) to predict HLT and HLB from chemical structure and two novel QSARs are detailed. The HLT and HLB QSARs show similar statistical performance; that is, r(2) = 0.89, r(2-ext) = 0.72 and 0.73 for training and external validation sets, respectively, and root-mean-square errors for the validation data sets are 0.70 and 0.75, respectively.

High accuracy in silico sulfotransferase models

Predicting enzymatic behavior in silico is an integral part of our efforts to understand biology. Hundreds of millions of compounds lie in targeted in silico libraries waiting for their metabolic potential to be discovered. In silico "enzymes" capable of accurately determining whether compounds can inhibit or react is often the missing piece in this endeavor. This problem has now been solved for the cytosolic sulfotransferases (SULTs). SULTs regulate the bioactivities of thousands of compounds--endogenous metabolites, drugs and other xenobiotics--by transferring the sulfuryl moiety (SO3) from 3'-phosphoadenosine 5'-phosphosulfate to the hydroxyls and primary amines of these acceptors. SULT1A1 and 2A1 catalyze the majority of sulfation that occurs during human Phase II metabolism. Here, recent insights into the structure and dynamics of SULT binding and reactivity are incorporated into in silico models of 1A1 and 2A1 that are used to identify substrates and inhibitors in a structurally diverse set of 1,455 high value compounds: the FDA-approved small molecule drugs. The SULT1A1 models predict 76 substrates. Of these, 53 were known substrates. Of the remaining 23, 21 were tested, and all were sulfated. The SULT2A1 models predict 22 substrates, 14 of which are known substrates. Of the remaining 8, 4 were tested, and all are substrates. The models proved to be 100% accurate in identifying substrates and made no false predictions at Kd thresholds of 100 μM. In total, 23 "new" drug substrates were identified, and new linkages to drug inhibitors are predicted. It now appears to be possible to accurately predict Phase II sulfonation in silico.

Prediction of cytochrome P450 xenobiotic metabolism: tethered docking and reactivity derived from ligand molecular orbital analysis

Metabolism of xenobiotic and endogenous compounds is frequently complex, not completely elucidated, and therefore often ambiguous. The prediction of sites of metabolism (SoM) can be particularly helpful as a first step toward the identification of metabolites, a process especially relevant to drug discovery. This paper describes a reactivity approach for predicting SoM whereby reactivity is derived directly from the ground state ligand molecular orbital analysis, calculated using Density Functional Theory, using a novel implementation of the average local ionization energy. Thus each potential SoM is sampled in the context of the whole ligand, in contrast to other popular approaches where activation energies are calculated for a predefined database of molecular fragments and assigned to matching moieties in a query ligand. In addition, one of the first descriptions of molecular dynamics of cytochrome P450 (CYP) isoforms 3A4, 2D6, and 2C9 in their Compound I state is reported, and, from the representative protein structures obtained, an analysis and evaluation of various docking approaches using GOLD is performed. In particular, a covalent docking approach is described coupled with the modeling of important electrostatic interactions between CYP and ligand using spherical constraints. Combining the docking and reactivity results, obtained using standard functionality from common docking and quantum chemical applications, enables a SoM to be identified in the top 2 predictions for 75%, 80%, and 78% of the data sets for 3A4, 2D6, and 2C9, respectively, results that are accessible and competitive with other recently published prediction tools.

Role of electrostatic potential in the in silico prediction of molecular bioactivation and mutagenesis

Electrostatic potential (ESP) is a useful physicochemical property of a molecule that provides insights into inter- and intramolecular associations, as well as prediction of likely sites of electrophilic and nucleophilic metabolic attack. Knowledge of sites of metabolic attack is of paramount importance in DMPK research since drugs frequently fail in clinical trials due to the formation of bioactivated metabolites which are often difficult to measure experimentally due to their reactive nature and relatively short half-lives. Computational chemistry methods have proven invaluable in recent years as a means to predict and study bioactivated metabolites without the need for chemical syntheses, or testing on experimental animals. Additional molecular properties (heat of formation, heat of solvation and E(LUMO) - E(HOMO)) are discussed in this paper as complementary indicators of the behavior of metabolites in vivo. Five diverse examples are presented (acetaminophen, aniline/phenylamines, imidacloprid, nefazodone and vinyl chloride) which illustrate the utility of this multidimensional approach in predicting bioactivation, and in each case the predicted data agreed with experimental data described in the scientific literature. A further example of the usefulness of calculating ESP, in combination with the molecular properties mentioned above, is provided by an examination of the use of these parameters in providing an explanation for the sites of nucleophilic attack of the nucleic acid cytosine. Exploration of sites of nucleophilic attack of nucleic acids is important as adducts of DNA have the potential to result in mutagenesis.

Stable isotope- and mass spectrometry-based metabolomics as tools in drug metabolism: a study expanding tempol pharmacology

The application of mass spectrometry-based metabolomics in the field of drug metabolism has yielded important insights not only into the metabolic routes of drugs but has provided unbiased, global perspectives of the endogenous metabolome that can be useful for identifying biomarkers associated with mechanism of action, efficacy, and toxicity. In this report, a stable isotope- and mass spectrometry-based metabolomics approach that captures both drug metabolism and changes in the endogenous metabolome in a single experiment is described. Here the antioxidant drug tempol (4-hydroxy-2,2,6,6-tetramethylpiperidine-N-oxyl) was chosen because its mechanism of action is not completely understood and its metabolic fate has not been studied extensively. Furthermore, its small size (MW = 172.2) and chemical composition (C(9)H(18)NO(2)) make it challenging to distinguish from endogenous metabolites. In this study, mice were dosed with tempol or deuterated tempol (C(9)D(17)HNO(2)) and their urine was profiled using ultraperformance liquid chromatography coupled with quadrupole time-of-flight mass spectrometry. Principal component analysis of the urinary metabolomics data generated a Y-shaped scatter plot containing drug metabolites (protonated and deuterated) that were clearly distinct from the endogenous metabolites. Ten tempol drug metabolites, including eight novel metabolites, were identified. Phase II metabolism was the major metabolic pathway of tempol in vivo, including glucuronidation and glucosidation. Urinary endogenous metabolites significantly elevated by tempol treatment included 2,8-dihydroxyquinoline (8.0-fold, P < 0.05) and 2,8-dihydroxyquinoline-β-d-glucuronide (6.8-fold, P < 0.05). Urinary endogenous metabolites significantly attenuated by tempol treatment including pantothenic acid (1.3-fold, P < 0.05) and isobutrylcarnitine (5.3-fold, P < 0.01). This study underscores the power of a stable isotope- and mass spectrometry-based metabolomics in expanding the view of drug pharmacology.

Software aided approaches to structure-based metabolite identification in drug discovery and development

Technological advances in mass spectrometry (MS) such as accurate mass high resolution instrumentation have fundamentally changed the approach to systematic metabolite identification over the past decade. Despite technological break-through on the instrumental side, metabolite identification still requires tedious manual data inspection and interpretation of huge analytical datasets. The process of metabolite identification has become largely facilitated and partly automated by cheminformatics approaches such as knowledge base metabolite prediction using, for example, Meteor, MetaDrug, MetaSite and StarDrop that are typically applied pre-acquisition. Likewise, emerging new technologies in postacquisition data analysis like mass defect filtering (MDF) have moved the technology driven analytical methodology to metabolite identification toward generic, structure-based workflows. The biggest challenge for automation however remains the structural assignment of drug metabolites. Software-guided approaches for the unsupervised metabolite identification still cannot compete with expert user manual data interpretation yet. Recently MassMetaSite has been introduced for the automated ranked output of metabolite structures based on the combination of metabolite prediction and interrogation of analytical mass spectrometric data. This approach and others are promising milestones toward an unsupervised process to metabolite identification and structural characterization moving away from a sample focused per-compound approach to a structure-driven generic workflow.

Modeling and Simulation of Intracellular Drug Transport and Disposition Pathways with Virtual Cell

The development of computational approaches for modeling the spatiotemporal dynamics of intracellular, small molecule drug concentrations has become an increasingly important area of pharmaceutical research. For systems pharmacology, the system dynamics of subcellular transport can be coupled to downstream pharmacological effects on biochemical pathways that impact cell structure and function. Here, we demonstrate how a widely used systems biology modeling package - Virtual Cell - can also be used to model the intracellular, passive transport pathways of small druglike molecules. Using differential equations to represent passive drug transport across cellular membranes, spatiotemporal changes in the intracellular distribution and concentrations of exogenous chemical agents in specific subcellular organelles were simulated for weakly acidic, neutral, and basic molecules, as a function of the molecules' lipophilicity and ionization potentials. In addition, we simulated the transport properties of small molecule chemical agents in the presence of a homogenous extracellular concentration or a transcellular concentration gradient. We also simulated the effects of cell type-dependent variations in the intracellular microenvironments on the distribution and accumulation of small molecule chemical agents in different organelles over time, under influx and efflux conditions. Lastly, we simulated the transcellular transport of small molecule chemical agents, in the presence of different apical and basolateral microenvironments. By incorporating existing models of drug permeation and subcellular distribution, our results indicate that Virtual Cell can provide a user-friendly, open, online computational modeling platform for systems pharmacology and biopharmaceutics research, making mathematical models and simulation results accessible to a broad community of users, without requiring advanced computer programming knowledge.

Computational tools and resources for metabolism-related property predictions. 1. Overview of publicly available (free and commercial) databases and software

Metabolism has been identified as a defining factor in drug development success or failure because of its impact on many aspects of drug pharmacology, including bioavailability, half-life and toxicity. In this article, we provide an outline and descriptions of the resources for metabolism-related property predictions that are currently either freely or commercially available to the public. These resources include databases with data on, and software for prediction of, several end points: metabolite formation, sites of metabolic transformation, binding to metabolizing enzymes and metabolic stability. We attempt to place each tool in historical context and describe, wherever possible, the data it was based on. For predictions of interactions with metabolizing enzymes, we show a typical set of results for a small test set of compounds. Our aim is to give a clear overview of the areas and aspects of metabolism prediction in which the currently available resources are useful and accurate, and the areas in which they are inadequate or missing entirely.

Pharmacophore, QSAR, and ADME based semisynthesis and in vitro evaluation of ursolic acid analogs for anticancer activity

In the present work, QSAR models for predicting the activities of ursolic acid analogs against human lung (A-549) and CNS (SF-295) cancer cell lines were developed by a forward stepwise multiple linear regression method using a leave-one-out approach. The regression coefficient (r(2)) and the cross-validation regression coefficient (rCV(2)) of the QSAR model for cytotoxic activity against the human lung cancer cell line (A-549) were 0.85 and 0.80, respectively. The QSAR study indicated that the LUMO energy, ring count, and solvent-accessible surface area were strongly correlated with anticancer activity. Similarly, the QSAR model for cytotoxic activity against the human CNS cancer cell line (SF-295) also showed a high correlation (r(2) = 0.99 and rCV(2) = 0.96), and indicated that dipole vector and solvent-accessible surface area were strongly correlated with activity. Ursolic acid analogs that were predicted to be active against these cancer cell lines by the QSAR models were semisynthesized and characterized on the basis of their (1)H and (13)C NMR spectroscopic data, and were then tested in vitro against the human lung (A-549) and CNS (SF-295) cancer cell lines. The experimental results obtained agreed well with the predicted values.

Quantitative Property-Property Relationship for Screening-Level Prediction of Intrinsic Clearance of Volatile Organic Chemicals in Rats and Its Integration within PBPK Models to Predict Inhalation Pharmacokinetics in Humans

The objectives of this study were (i) to develop a screening-level Quantitative property-property relationship (QPPR) for intrinsic clearance (CL_int) obtained from in vivo animal studies and (ii) to incorporate it with human physiology in a PBPK model for predicting the inhalation pharmacokinetics of VOCs. CL_int, calculated as the ratio of the in vivo V_max (μmol/h/kg bw rat) to the K_m (μM), was obtained for 26 VOCs from the literature. The QPPR model resulting from stepwise linear regression analysis passed the validation step (R² = 0.8; leave-one-out cross-validation Q² = 0.75) for CL_int normalized to the phospholipid (PL) affinity of the VOCs. The QPPR facilitated the calculation of CL_int (L PL/h/kg bw rat) from the input data on log P_ow, log blood: water PC and ionization potential. The predictions of the QPPR as lower and upper bounds of the 95% mean confidence intervals (LMCI and UMCI, resp.) were then integrated within a human PBPK model. The ratio of the maximum (using LMCI for CL_int) to minimum (using UMCI for CL_int) AUC predicted by the QPPR-PBPK model was 1.36 ± 0.4 and ranged from 1.06 (1,1-dichloroethylene) to 2.8 (isoprene). Overall, the integrated QPPR-PBPK modeling method developed in this study is a pragmatic way of characterizing the impact of the lack of knowledge of CL_int in predicting human pharmacokinetics of VOCs, as well as the impact of prediction uncertainty of CL_int on human pharmacokinetics of VOCs.

Identification and characterization of reactive metabolites in natural products-driven drug discovery

Toxicity of natural products arising from their metabolic biotransformation into reactive chemical intermediates is an important reason for high attrition rates in early drug discovery efforts. Screening promising natural products for their likelihood to form such metabolites is therefore an important step in identifying potential liabilities in the drug development process. However, such screening is complicated by the need to have test methods that are sensitive, reliable, accurate, efficient, and cost-effective enough to allow for routine identification and characterization of the reactive metabolites. These metabolites are typically formed in minute quantities, usually through minor metabolic pathways, and, due to their highly reactive and therefore transient chemical nature, pose considerable analytical challenges in attempts to determine their properties. Understanding the formation of reactive metabolites may be used as the basis for synthetic chemical modification of parent natural products aimed at bypassing such harmful bioactivation. This paper highlights the general principles and protocols commonly used to predict and study the formation of reactive metabolites in vitro and how the data obtained from such studies can be used in the development of safer drugs from natural products.

Computational prediction of metabolism: sites, products, SAR, P450 enzyme dynamics, and mechanisms

Metabolism of xenobiotics remains a central challenge for the discovery and development of drugs, cosmetics, nutritional supplements, and agrochemicals. Metabolic transformations are frequently related to the incidence of toxic effects that may result from the emergence of reactive species, the systemic accumulation of metabolites, or by induction of metabolic pathways. Experimental investigation of the metabolism of small organic molecules is particularly resource demanding; hence, computational methods are of considerable interest to complement experimental approaches. This review provides a broad overview of structure- and ligand-based computational methods for the prediction of xenobiotic metabolism. Current computational approaches to address xenobiotic metabolism are discussed from three major perspectives: (i) prediction of sites of metabolism (SOMs), (ii) elucidation of potential metabolites and their chemical structures, and (iii) prediction of direct and indirect effects of xenobiotics on metabolizing enzymes, where the focus is on the cytochrome P450 (CYP) superfamily of enzymes, the cardinal xenobiotics metabolizing enzymes. For each of these domains, a variety of approaches and their applications are systematically reviewed, including expert systems, data mining approaches, quantitative structure–activity relationships (QSARs), and machine learning-based methods, pharmacophore-based algorithms, shape-focused techniques, molecular interaction fields (MIFs), reactivity-focused techniques, protein–ligand docking, molecular dynamics (MD) simulations, and combinations of methods. Predictive metabolism is a developing area, and there is still enormous potential for improvement. However, it is clear that the combination of rapidly increasing amounts of available ligand- and structure-related experimental data (in particular, quantitative data) with novel and diverse simulation and modeling approaches is accelerating the development of effective tools for prediction of in vivo metabolism, which is reflected by the diverse and comprehensive data sources and methods for metabolism prediction reviewed here. This review attempts to survey the range and scope of computational methods applied to metabolism prediction and also to compare and contrast their applicability and performance.

Systematic drug repositioning based on clinical side-effects

Drug repositioning helps fully explore indications for marketed drugs and clinical candidates. Here we show that the clinical side-effects (SEs) provide a human phenotypic profile for the drug, and this profile can suggest additional disease indications. We extracted 3,175 SE-disease relationships by combining the SE-drug relationships from drug labels and the drug-disease relationships from PharmGKB. Many relationships provide explicit repositioning hypotheses, such as drugs causing hypoglycemia are potential candidates for diabetes. We built Naïve Bayes models to predict indications for 145 diseases using the SEs as features. The AUC was above 0.8 in 92% of these models. The method was extended to predict indications for clinical compounds, 36% of the models achieved AUC above 0.7. This suggests that closer attention should be paid to the SEs observed in trials not just to evaluate the harmful effects, but also to rationally explore the repositioning potential based on this “clinical phenotypic assay”.

A systems biology strategy for predicting similarities and differences of drug effects: evidence for drug-specific modulation of inflammation in atherosclerosis

Background

Successful drug development has been hampered by a limited understanding of how to translate laboratory-based biological discoveries into safe and effective medicines. We have developed a generic method for predicting the effects of drugs on biological processes. Information derived from the chemical structure and experimental omics data from short-term efficacy studies are combined to predict the possible protein targets and cellular pathways affected by drugs.

Results

Validation of the method with anti-atherosclerotic compounds (fenofibrate, rosuvastatin, LXR activator T0901317) demonstrated a great conformity between the computationally predicted effects and the wet-lab biochemical effects. Comparative genome-wide pathway mapping revealed that the biological drug effects were realized largely via different pathways and mechanisms. In line with the predictions, the drugs showed differential effects on inflammatory pathways (downstream of PDGF, VEGF, IFNγ, TGFβ, IL1β, TNFα, LPS), transcriptional regulators (NFκB, C/EBP, STAT3, AP-1) and enzymes (PKCδ, AKT, PLA2), and they quenched different aspects of the inflammatory signaling cascade. Fenofibrate, the compound predicted to be most efficacious in inhibiting early processes of atherosclerosis, had the strongest effect on early lesion development.

Conclusion

Our approach provides mechanistic rationales for the differential and common effects of drugs and may help to better understand the origins of drug actions and the design of combination therapies.

Computational strategies for metabolite identification in metabolomics

Most metabolomic data are characterized by complex spectra or chromatograms containing hundreds of peaks or features. While there are many methods for aligning or comparing these spectral features, there are few approaches for actually identifying which peaks match to which compounds. Indeed, one of the biggest unmet needs in the field of metabolomics lies in the problem of compound identification. This review describes some of the newly emerging computational strategies in metabolomics that are being used to aid in the identification of metabolites from biofluid mixtures analyzed by NMR and MS. The most successful compound-identification strategies typically involve matching spectral features of the unknown compound(s) to curated spectral databases of reference compounds. This approach is known as the identification of 'known unknowns'. However, the identification of truly novel compounds (the 'unknown unknowns') is particularly challenging and requires the use of computer-aided structure elucidation methods being applied to the purified compound. The strengths and limitations of these approaches as applied to different analytical technologies (GC-MS, LC-MS and NMR) will be discussed, as will prospects for potential improvements to existing strategies.

Advances in structure elucidation of small molecules using mass spectrometry

The structural elucidation of small molecules using mass spectrometry plays an important role in modern life sciences and bioanalytical approaches. This review covers different soft and hard ionization techniques and figures of merit for modern mass spectrometers, such as mass resolving power, mass accuracy, isotopic abundance accuracy, accurate mass multiple-stage MS(n) capability, as well as hybrid mass spectrometric and orthogonal chromatographic approaches. The latter part discusses mass spectral data handling strategies, which includes background and noise subtraction, adduct formation and detection, charge state determination, accurate mass measurements, elemental composition determinations, and complex data-dependent setups with ion maps and ion trees. The importance of mass spectral library search algorithms for tandem mass spectra and multiple-stage MS(n) mass spectra as well as mass spectral tree libraries that combine multiple-stage mass spectra are outlined. The successive chapter discusses mass spectral fragmentation pathways, biotransformation reactions and drug metabolism studies, the mass spectral simulation and generation of in silico mass spectra, expert systems for mass spectral interpretation, and the use of computational chemistry to explain gas-phase phenomena. A single chapter discusses data handling for hyphenated approaches including mass spectral deconvolution for clean mass spectra, cheminformatics approaches and structure retention relationships, and retention index predictions for gas and liquid chromatography. The last section reviews the current state of electronic data sharing of mass spectra and discusses the importance of software development for the advancement of structure elucidation of small molecules. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1007/s12566-010-0015-9) contains supplementary material, which is available to authorized users.

The use of machine learning and nonlinear statistical tools for ADME prediction

Absorption, distribution, metabolism and excretion (ADME)-related failure of drug candidates is a major issue for the pharmaceutical industry today. Prediction of ADME by in silico tools has now become an inevitable paradigm to reduce cost and enhance efficiency in pharmaceutical research. Recently, machine learning as well as nonlinear statistical tools has been widely applied to predict routine ADME end points. To achieve accurate and reliable predictions, it would be a prerequisite to understand the concepts, mechanisms and limitations of these tools. Here, we have devised a small synthetic nonlinear data set to help understand the mechanism of machine learning by 2D-visualisation. We applied six new machine learning methods to four different data sets. The methods include Naive Bayes classifier, classification and regression tree, random forest, Gaussian process, support vector machine and k nearest neighbour. The results demonstrated that ensemble learning and kernel machine displayed greater accuracy of prediction than classical methods irrespective of the data set size. The importance of interaction with the engineering field is also addressed. The results described here provide insights into the mechanism of machine learning, which will enable appropriate usage in the future.

In Silico Toxicology in Drug Discovery — Concepts Based on Three-dimensional Models

Animal testing is still compulsory worldwide, for the approval of drugs and chemicals produced in large quantities. Computer-assisted ( in silico) technologies are considered to be efficient alternatives to in vivo experiments, and are therefore endorsed by many regulatory agencies, e.g. for use in the European REACH initiative. Advantages of in silico methods include: the possible study of hypothetical compounds; their low cost; and the fact that such virtual experiments are typically based on human data, thus making the question of interspecies transferability obsolete. Since the mid-1990s, computer-based technologies have become an indispensable tool in drug discovery — used primarily to identify small molecules displaying a stereospecific and selective binding to a regulatory macromolecule. Since toxic effects are still responsible for some 20% of the late-stage failures, there is a continuing need for in silico concepts which can be used to estimate a compound's ADMET ( adsorption, distribution, metabolism, elimination, toxicity) properties — in particular, toxicity. The aim of this paper is to provide an insight into computational technologies that allow for the prediction of toxic effects triggered by pharmaceuticals. As most adverse and toxic effects are mediated by unwanted interactions with macromolecules involved in biological regulatory systems, we have focused on methodologies that are based on three-dimensional models of small molecules binding to such entities, and discuss the results at the molecular level.

A mapping of drug space from the viewpoint of small molecule metabolism

Small molecule drugs target many core metabolic enzymes in humans and pathogens, often mimicking endogenous ligands. The effects may be therapeutic or toxic, but are frequently unexpected. A large-scale mapping of the intersection between drugs and metabolism is needed to better guide drug discovery. To map the intersection between drugs and metabolism, we have grouped drugs and metabolites by their associated targets and enzymes using ligand-based set signatures created to quantify their degree of similarity in chemical space. The results reveal the chemical space that has been explored for metabolic targets, where successful drugs have been found, and what novel territory remains. To aid other researchers in their drug discovery efforts, we have created an online resource of interactive maps linking drugs to metabolism. These maps predict the "effect space" comprising likely target enzymes for each of the 246 MDDR drug classes in humans. The online resource also provides species-specific interactive drug-metabolism maps for each of the 385 model organisms and pathogens in the BioCyc database collection. Chemical similarity links between drugs and metabolites predict potential toxicity, suggest routes of metabolism, and reveal drug polypharmacology. The metabolic maps enable interactive navigation of the vast biological data on potential metabolic drug targets and the drug chemistry currently available to prosecute those targets. Thus, this work provides a large-scale approach to ligand-based prediction of drug action in small molecule metabolism.

Distinction between human cytochrome P450 (CYP) isoforms and identification of new phosphorylation sites by mass spectrometry

In mammals, Cytochrome P450 (CYP) enzymes are bound to membranes of the endoplasmic reticulum and mitochondria, where they are responsible for the oxidative metabolism of many xenobiotics as well as organic endogenous compounds. In humans, 57 isoforms were identified which are classified based on sequence homology. In the present work, we demonstrate the performance of a mass spectrometry-based strategy to simultaneously detect and differentiate distinct human Cytochrome P450 (CYP) isoforms including the highly similar CYP3A4, CYP3A5, CYP3A7, as well as CYP2C8, CYP2C9, CYP2C18, CYP2C19, and CYP4F2, CYP4F3, CYP4F11, CYP4F12. Compared to commonly used immunodetection methods, mass spectrometry overcomes limitations such as low antibody specificity and offers high multiplexing possibilities. Furthermore, CYP phosphorylation, which may affect various biochemical and enzymatic properties of these enzymes, is still poorly analyzed, especially in human tissues. Using titanium dioxide resin combined with tandem mass spectrometry for phosphopeptide enrichment and sequencing, we discovered eight human P450 phosphorylation sites, seven of which were novel. The data from surgical human liver samples establish that the isoforms CYP1A2, CYP2A6, CYP2B6, CYP2E1, CYP2C8, CYP2D6, CYP3A4, CYP3A7, and CYP8B1 are phosphorylated in vivo. These results will aid in further investigation of the functional significance of protein phosphorylation for this important group of enzymes.

Role of biotransformation studies in minimizing metabolism-related liabilities in drug discovery

Metabolism-related liabilities continue to be a major cause of attrition for drug candidates in clinical development. Such problems may arise from the bioactivation of the parent compound to a reactive metabolite capable of modifying biological materials covalently or engaging in redox-cycling reactions leading to the formation of other toxicants. Alternatively, they may result from the formation of a major metabolite with systemic exposure and adverse pharmacological activity. To avert such problems, biotransformation studies are becoming increasingly important in guiding the refinement of a lead series during drug discovery and in characterizing lead candidates prior to clinical evaluation. This article provides an overview of the methods that are used to uncover metabolism-related liabilities in a pre-clinical setting and offers suggestions for reducing such liabilities via the modification of structural features that are used commonly in drug-like molecules.

LC-MS-based metabolomics in drug metabolism

Xenobiotic metabolism, a ubiquitous natural response to foreign compounds, elicits initiating signals for many pathophysiological events. Currently, most widely used techniques for identifying xenobiotic metabolites and metabolic pathways are empirical and largely based on in vitro incubation assays and in vivo radiotracing experiments. Recent work in our lab has shown that LC-MS-based metabolomic techniques are useful tools for xenobiotic metabolism research since multivariate data analysis in metabolomics can significantly rationalize the processes of xenobiotic metabolite identification and metabolic pathway analysis. In this review, the technological elements of LC-MS-based metabolomics for constructing high-quality datasets and conducting comprehensive data analysis are examined. Four novel approaches of using LC-MS-based metabolomic techniques in xenobiotic metabolism research are proposed and illustrated by case studies and proof-of-concept experiments, and the perspective on their application is further discussed.

Strategy for genotoxicity testing—Metabolic considerations

The report from the 2002 International Workshop on Genotoxicity Tests (IWGT) Strategy Expert Group emphasized metabolic considerations as an important area to address in developing a common strategy for genotoxicity testing. A working group convened at the 2005 4th IWGT to discuss this area further and propose practical strategy recommendations. To propose a strategy, the working group reviewed: (1) the current status and deficiencies, including examples of carcinogens "missed" in genotoxicity testing, established shortcomings of the standard in vitro induced S9 activation system and drug metabolite case examples; (2) the current status of possible remedies, including alternative S9 sources, other external metabolism systems or genetically engineered test systems; (3) any existing positions or guidance. The working group established consensus principles to guide strategy development. Thus, a human metabolite of interest should be represented in genotoxicity and carcinogenicity testing, including evaluation of alternative genotoxicity in vitro metabolic activation or test systems, and the selection of a carcinogenicity test species showing appropriate biotransformation. Appropriate action triggers need to be defined based on the extent of human exposure, considering any structural knowledge of the metabolite, and when genotoxicity is observed upon in vitro testing in the presence of metabolic activation. These triggers also need to be considered in defining the timing of human pharmaceutical ADME assessments. The working group proposed two strategies to consider; a more proactive approach, which emphasizes early metabolism predictions to drive appropriate hazard assessment; and a retroactive approach to manage safety risks of a unique or "major" metabolite once identified and quantitated from human clinical ADME studies. In both strategies, the assessment of the genotoxic potential of a metabolite could include the use of an alternative or optimized in vitro metabolic activation system, or direct testing of an isolated or synthesized metabolite. The working group also identified specific areas where more data or experiences need to be gained to reach consensus. These included defining a discrete exposure action trigger for safety assessment and when direct testing of a metabolite of interest is warranted versus the use of an alternative in vitro activation system, a universal recommendation for the timing of human ADME studies for drug candidates and the positioning of metabolite structural knowledge (through in silico systems, literature, expert analysis) in supporting metabolite safety qualification. Lastly, the working group outlined future considerations for refining the initially proposed strategies. These included the need for further evaluation of the current in vitro genotoxicity testing protocols that can potentially perturb or reduce the level of metabolic activity (potential alterations in metabolism associated with both the use of some solvents to solubilize test chemicals and testing to the guidance limit dose), and proposing broader evaluations of alternative metabolic activation sources or engineered test systems to further challenge the suitability of (or replace) the current induced liver S9 activation source.

Predicting the Oxidative Metabolism of Statins: An Application of the MetaSite® Algorithm

PURPOSE

This study was undertaken to examine the MetaSite algorithm by comparing its predictions with experimentally characterized metabolites of statins produced by cytochromes P450 (CYPs).

METHODS

Seven statins were investigated, namely atorvastatin, cerivastatin, fluvastatin, pitavastatin and pravastatin which are (or were) used in their active hydroxy-acid form, and lovastatin and simvastatin which are used as the lactone prodrug. But given the fast lactone-hydroxy-acid equilibrium undergone by statins, both forms were investigated for each of the seven drugs. The MetaSite version 2.5.3 used here contains the homology 3D-models of CYP1A2, CYP2C19, CYP2C9, CYP2D6 and CYP3A4. In addition, we also used the crystallographic 3D-structure of human CYP2C9 and CYP3A4. To allow a better interpretation of results, the probability function PsMi calculated by MetaSite (namely the probability of atom i to be a site of metabolism) was explicitly decomposed into its two components, namely a recognition score Ei (the accessibility of atom i) and the chemical reactivity Ri of atom i toward oxidation reactions.

RESULTS

The current version of MetaSite is known to work best with prior experimental knowledge of the cytochrome(s) P450 involved. And indeed, experimentally confirmed sites of oxidation were correctly given a high priority by MetaSite. In particular 77% of correct predictions (including false positive but, as discussed, this is not necessarily a shortcoming) were obtained when considering the first five metabolites indicated by MetaSite.

CONCLUSION

To the best of our knowledge, this is the first independent report on the software. It is expected to contribute to the development of improved versions, but above all it demonstrates that the usefulness of such softwares critically depends on human experts.

Algorithms for network analysis in systems-ADME/Tox using the MetaCore and MetaDrug platforms

The authors have previously applied two integrated platforms, MetaCore and MetaDrug, for the assembly and analysis of human biological networks as a useful method for the integration and functional interpretation of high-throughput experimental data. The present study demonstrates in detail the specific algorithms that are used in both software platforms. Using a standard set of genes as input, namely CYP3A4 (an enzyme), PXR (a nuclear hormone receptor), MDR1 (a transporter) and hERG (an ion channel) related to the absorption, distribution, metabolism, excretion and toxicity (ADME/Tox) of xenobiotics, we have now generated networks with each algorithm. The relative advantages and disadvantages of these algorithms are explained using these examples as well as appropriate instances of utility to illustrate further the particular circumstances for their use. In addition, the benefits of the different network algorithms are identified when compared with algorithms available in other products, where this information is available.

Methods for Predicting Human Drug Metabolism

Drug metabolism information is a necessary component of drug discovery and development. The key issues in drug metabolism include identifying: the enzyme(s) involved, the site(s) of metabolism, the resulting metabolite(s), and the rate of metabolism. Methods for predicting human drug metabolism from in vitro and computational methodologies and determining relationships between the structure and metabolic activity of molecules are also critically important for understanding potential drug interactions and toxicity. There are numerous experimental and computational approaches that have been developed in order to predict human metabolism which have their own limitations. It is apparent that few of the computational tools for metabolism prediction alone provide the major integrated functions needed to assist in drug discovery. Similarly the different in vitro methods for human drug metabolism themselves have implicit limitations. The utilization of these methods for pharmaceutical and other applications as well as their integration is discussed as it is likely that hybrid methods will provide the most success.

Classification of Metabolites with Kernel-Partial Least Squares (K-PLS)

Numerous experimental and computational approaches have been developed to predict human drug metabolism. Since databases of human drug metabolism information are widely available, these can be used to train computational algorithms and generate predictive approaches. In turn, they may be used to assist in the identification of possible metabolites from a large number of molecules in drug discovery based on molecular structure alone. In the current study we have used a commercially available database (MetaDrug) and extracted a fraction of the human drug metabolism data. These data were used along with augmented atom descriptors in a predictive machine learning model, kernel-partial least squares (K-PLS). A total of 317 molecules, including parent drugs and their primary and secondary (sequential) metabolites, were used to build these models corresponding to individual metabolism rules, representing the formation of discrete metabolites, e.g., N-dealkylation. Each model was internally validated to assess the capability to classify other molecules that were left out. Using receiver operator curve statistics models for N-dealkylation, O-dealkylation, aromatic hydroxylation, aliphatic hydroxylation, O-glucuronidation, and O-sulfation gave area under the curve values from 0.75 to 0.84 and were able to predict between 61 and 79% active molecules upon leave-one-out testing. This preliminary study indicates that K-PLS and possibly other similar machine learning methods (such as support vector machines) can be applied to predicting human drug metabolite formation in a classification manner. Improvements can be achieved using considerably larger datasets that contain more positive examples for the less frequently occurring metabolite rules, as well as the external evaluation of novel molecules.

Structure-based methods for the prediction of drug metabolism

There is a tantalising possibility that we may be able to predict the metabolism of a drug directly from its structure, thus obviating the requirement for animal tests in this area. There are a number of techniques that can be used to estimate a range of events associated with metabolism, and may allow us to achieve this aim. This paper considers the role of (quantitative) structure-activity relationships, and pharmacophore and homology modelling in the prediction of metabolism. Examples are also presented where such approaches have been formalised into expert systems. Clearly, many advances have been made in this area in recent years. Discussed herein is the importance of fully integrating the diverse systems and approaches available to fulfil the aspiration to predict metabolism directly from structure.