Quantitative structure-pharmacokinetic relationships for drug clearance by using statistical learning methods

Another string to your bow: machine learning prediction of the pharmacokinetic properties of small molecules

INTRODUCTION

Prediction of pharmacokinetic (PK) properties is crucial for drug discovery and development. Machine-learning (ML) models, which use statistical pattern recognition to learn correlations between input features (such as chemical structures) and target variables (such as PK parameters), are being increasingly used for this purpose. To embed ML models for PK prediction into workflows and to guide future development, a solid understanding of their applicability, advantages, limitations, and synergies with other approaches is necessary.

AREAS COVERED

This narrative review discusses the design and application of ML models to predict PK parameters of small molecules, especially in light of established approaches including in vitro-in vivo extrapolation (IVIVE) and physiologically based pharmacokinetic (PBPK) models. The authors illustrate scenarios in which the three approaches are used and emphasize how they enhance and complement each other. In particular, they highlight achievements, the state of the art and potentials of applying machine learning for PK prediction through a comphrehensive literature review.

EXPERT OPINION

ML models, when carefully crafted, regularly updated, and appropriately used, empower users to prioritize molecules with favorable PK properties. Informed practitioners can leverage these models to improve the efficiency of drug discovery and development process.

Synthesis and Evaluation of Antimicrobial Activity of the Rearranged Abietane Prattinin A and Its Synthetic Derivatives

Synthesis of the natural product prattinin A and some new derivatives has been achieved using abietic acid. The final products and a selection of intermediates were evaluated for their antibacterial activity against three human pathogenic bacteria: E. coli, P. aeruginosa, and S. aureus. The results showed that the antibacterial activity varies depending on the chemical structure of the compounds. Notably, compound 27 exhibited the most potent activity against E. coli and P. aeruginosa, with a minimal inhibitory concentration (MIC) of 11.7 µg/mL, comparable to that of the standard antibiotic ciprofloxacin, and strong activity against S. aureus, with an MIC of 23.4 µg/mL. Furthermore, we assessed the stability of these derivative compounds as potential antimicrobial agents and determined their interactions with the crystal structure of the protein receptor mutant TEM-12 from E. coli (pdb:1ESU) using molecular docking via UCSF Chimera software 1.17.3. The results suggest that 27 has potential as a natural antibiotic agent.

Lipophilic Studies and In Silico ADME Profiling of Biologically Active 2-Aminothiazol-4(5H)-one Derivatives

Pseudothiohydantoin derivatives have a wide range of biological activities and are widely used in the development of new pharmaceuticals. Lipophilicity is a basic, but very important parameter in the design of potential drugs, as it determines solubility in lipids, nonpolar solvents, and makes it possible to predict the ADME profile. The aim of this study was to evaluate the lipophilicity of 28 pseudothiohydantoin derivatives showing the inhibition of 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1) using chromatographic methods. Experimentally, lipophilicity was determined by reverse phase thin layer chromatography (RP-TLC) and reverse phase high-performance liquid chromatography (RP-HPLC). In both methods, methanol was used as the organic modifier of the mobile phase. For each 2-aminothiazol-4(5H)-one derivative, a relationship was observed between the structure of the compound and the values of the lipophilicity parameters (log kw, RM0). Experimental lipophilicity values were compared with computer calculated partition coefficient (logP) values. A total of 27 of the 28 tested compounds had a lipophilicity value < 5, which therefore met the condition of Lipinski's rule. In addition, the in silico ADME assay showed favorable absorption, distribution, metabolism, and excretion parameters for most of the pseudothiohydantoin derivatives tested. The study of lipophilicity and the ADME analysis indicate that the tested compounds are good potential drug candidates.

Shear Strength Estimation of Reinforced Concrete Deep Beams Using a Novel Hybrid Metaheuristic Optimized SVR Models

This study looks to propose a hybrid soft computing approach that can be used to accurately estimate the shear strength of reinforced concrete (RC) deep beams. Support vector regression (SVR) is integrated with three novel metaheuristic optimization algorithms: African Vultures optimization algorithm (AVOA), particle swarm optimization (PSO), and Harris Hawks optimization (HHO). The proposed models, SVR-AVOA, -PSO, and -HHO, are designed and compared to reference existing models. Multi variables are used and evaluated to model and evaluate the deep beam’s shear strength, and the sensitivity of the selected variables in modeling the shear strength is assessed. The results indicate that the SVR-AVOA outperforms other proposed and existing models for the shear strength prediction. The mean absolute error of SVR-AVOA, SVR-PSO, and SVR-HHO are 43.17 kN, 44.09 kN, and 106.95 kN, respectively. The SVR-AVOA can be used as a soft computing technique to estimate the shear strength of the RC deep beam with a maximum error of ±3.39%. Furthermore, the sensitivity analysis shows that the deep beam’s key parameters (shear span to depth ratio, web reinforcement’s yield strength, concrete compressive strength, stirrups spacing, and the main longitudinal bars reinforcement ratio) are efficiently impacted in the shear strength detection of RC deep beam.

Machine learning in pharmacometrics: Opportunities and challenges

The explosive growth in medical devices, imaging and diagnostics, computing, and communication and information technologies in drug development and healthcare has created an ever-expanding data landscape that the pharmacometrics (PMX) research community must now traverse. The tools of machine learning (ML) have emerged as a powerful computational approach in other data-rich disciplines but its effective utilization in the pharmaceutical sciences and PMX modelling is in its infancy. ML-based methods can complement PMX modelling by enabling the information in diverse sources of big data, e.g. population-based public databases and disease-specific clinical registries, to be harnessed because they are capable of efficiently identifying salient variables associated with outcomes and delineating their interdependencies. ML algorithms are computationally efficient, have strong predictive capabilities and can enable learning in the big data setting. ML algorithms can be viewed as providing a computational bridge from big data to complement PMX modelling. This review provides an overview of the strengths and weaknesses of ML approaches vis-à-vis population methods, assesses current research into ML applications in the pharmaceutical sciences and provides perspective for potential opportunities and strategies for the successful integration and utilization of ML in PMX.

Predicting blood-to-plasma concentration ratios of drugs from chemical structures and volumes of distribution in humans

Abstract

Despite their importance in determining the dosing regimen of drugs in the clinic, only a few studies have investigated methods for predicting blood-to-plasma concentration ratios (Rb). This study established an Rb prediction model incorporating typical human pharmacokinetics (PK) parameters. Experimental Rb values were compiled for 289 compounds, offering reliable predictions by expanding the applicability domain. Notably, it is the largest list of Rb values reported so far. Subsequently, human PK parameters calculated from plasma drug concentrations, including the volume of distribution (Vd), clearance, mean residence time, and plasma protein binding rate, as well as 2702 kinds of molecular descriptors, were used to construct quantitative structure–PK relationship models for Rb. Among the evaluated PK parameters, logVd correlated best with Rb (correlation coefficient of 0.47). Thus, in addition to molecular descriptors selected by XGBoost, logVd was employed to construct the prediction models. Among the analyzed algorithms, artificial neural networks gave the best results. Following optimization using six molecular descriptors and logVd, the model exhibited a correlation coefficient of 0.64 and a root-mean-square error of 0.205, which were superior to those previously reported for other Rb prediction methods. Since Vd values and chemical structures are known for most medications, the Rb prediction model described herein is expected to be valuable in clinical settings.

Graphical abstract

Supplementary informations

The online version of this article (10.1007/s11030-021-10186-7) contains supplementary material, which is available to authorized users.

An automated approach to identify scientific publications reporting pharmacokinetic parameters

Pharmacokinetic (PK) predictions of new chemical entities are aided by prior knowledge from other compounds. The development of robust algorithms that improve preclinical and clinical phases of drug development remains constrained by the need to search, curate and standardise PK information across the constantly-growing scientific literature. The lack of centralised, up-to-date and comprehensive repositories of PK data represents a significant limitation in the drug development pipeline.In this work, we propose a machine learning approach to automatically identify and characterise scientific publications reporting PK parameters from in vivo data, providing a centralised repository of PK literature. A dataset of 4,792 PubMed publications was labelled by field experts depending on whether in vivo PK parameters were estimated in the study. Different classification pipelines were compared using a bootstrap approach and the best-performing architecture was used to develop a comprehensive and automatically-updated repository of PK publications. The best-performing architecture encoded documents using unigram features and mean pooling of BioBERT embeddings obtaining an F1 score of 83.8% on the test set. The pipeline retrieved over 121K PubMed publications in which in vivo PK parameters were estimated and it was scheduled to perform weekly updates on newly published articles. All the relevant documents were released through a publicly available web interface (https://app.pkpdai.com) and characterised by the drugs, species and conditions mentioned in the abstract, to facilitate the subsequent search of relevant PK data. This automated, open-access repository can be used to accelerate the search and comparison of PK results, curate ADME datasets, and facilitate subsequent text mining tasks in the PK domain.

Comparison of Machine Learning Models on Performance of Single- and Dual-Type Electrochromic Devices

This study shows that the model fitting based on machine learning (ML) from experimental data can successfully predict the electrochromic characteristics of single- and dual-type flexible electrochromic devices (ECDs) by using tungsten trioxide (WO3) and WO3/vanadium pentoxide (V2O5), respectively. Seven different regression methods were used for experimental observations, which belong to single and dual ECDs where 80% percent was used as training data and the remaining was taken as testing data. Among the seven different regression methods, K-nearest neighbor (KNN) achieves the best results with higher coefficient of determination (R 2) score and lower root-mean-squared error (RMSE) for the bleaching state of ECDs. Furthermore, higher R 2 score and lower RMSE for the coloration state of ECDs were achieved with Gaussian process regressor. The robustness result of the ML modeling demonstrates the reliability of prediction outcomes. These results can be proposed as promising models for different energy-saving flexible electronic systems.

Direct Comparison of Total Clearance Prediction: Computational Machine Learning Model versus Bottom-Up Approach Using In Vitro Assay

The in vitro-in vivo extrapolation (IVIVE) approach for predicting total plasma clearance (CL_tot) has been widely used to rank order compounds early in discovery. More recently, a computational machine learning approach utilizing physicochemical descriptors and fingerprints calculated from chemical structure information has emerged, enabling virtual predictions even earlier in discovery. Previously, this approach focused more on in vitro intrinsic clearance (CL_int) prediction. Herein, we directly compare these two approaches for predicting CL_tot in rats. A structurally diverse set of 1114 compounds with known in vivo CL_tot, in vitro CL_int, and plasma protein binding was used as the basis for this evaluation. The machine learning models were assessed by validation approaches using the time- and cluster-split training and test sets, and five-fold cross validation. Assessed by five-fold validation, the random forest regression (RF) and radial basis function (RBF) models demonstrated better prediction performance in eight attempted machine learning models. The CL_tot values predicted by the RF and RBF models were within two-fold of the observed values for 67.7 and 71.9% of cluster-split test set compounds, respectively, while the predictivity was worse in the time-split dataset. The predictivity of both models tended to be improved by incorporating in vitro parameters, unbound fraction in plasma (f_u,p), and CL_int. CL_tot prediction utilizing in vitro CL_int and the well-stirred model, correcting for the fraction unbound in blood, was substantially worse compared to machine learning approaches for the same cluster-split test set. The reason that CL_tot is underestimated by IVIVE is not fully explained by considering the calculated microsomal unbound fraction (cf_u,mic), extended clearance classification system (ECCS), and omitting high clearance compounds in excess of hepatic blood flow. The analysis suggests that in silico machine learning models may have the power to reduce reliance on or replace in vitro and in vivo studies for chemical structure optimization in early drug discovery.

ADMET profiling of geographically diverse phytochemical using chemoinformatic tools

Aim: Phytocompounds are important due to their uniqueness, however, only few reach the development phase due to their poor pharmacokinetics. Therefore, preassessing the absorption, distribution, metabolism, excretion and toxicity (ADMET) properties is essential in drug discovery. Methodology: Biologically diverse databases (Phytochemica, SerpentinaDB, SANCDB and NuBBE_DB) covering the region of India, Brazil and South Africa were considered to predict the ADMET using chemoinformatic tools (Qikprop, pkCSM and DataWarrior). Results: Screening through each of pharmacokinetic criteria resulted in identification of 24 compounds that adhere to all the ADMET properties. Furthermore, assessment revealed that five have potent anticancer biological activity against cancer cell lines. Conclusion: We have established an open-access database (ADMET-BIS) to enable identification of promising molecules that follow ADMET properties and can be considered for drug development.

In silico approaches and tools for the prediction of drug metabolism and fate: A review

The fate of administered drugs is largely influenced by their metabolism. For example, endogenous enzyme-catalyzed conversion of drugs may result in therapeutic inactivation or activation or may transform the drugs into toxic chemical compounds. This highlights the importance of drug metabolism in drug discovery and development, and accounts for the wide variety of experimental technologies that provide insights into the fate of drugs. In view of the high cost of traditional drug development, a number of computational approaches have been developed for predicting the metabolic fate of drug candidates, allowing for screening of large numbers of chemical compounds and then identifying a small number of promising candidates. In this review, we introduce in silico approaches and tools that have been developed to predict drug metabolism and fate, and assess their potential to facilitate the virtual discovery of promising drug candidates. We also provide a brief description of various recent models for predicting different aspects of enzyme-drug reactions and provide a list of recent in silico tools used for drug metabolism prediction.

Evaluating the influence of half-life, milk:plasma partition coefficient, and volume of distribution on lactational exposure to chemicals in children

Women are exposed to multiple environmental chemicals, many of which are known to transfer to breast milk during lactation. However, little is known about the influence of the different chemical-specific pharmacokinetic parameters on children's lactational dose. Our objective was to develop a generic pharmacokinetic model and subsequently quantify the influence of three chemical-specific parameters (biological half-life, milk:plasma partition coefficient, and volume of distribution) on lactational exposure to chemicals and resulting plasma levels in children. We developed a two-compartment pharmacokinetic model to simulate lifetime maternal exposure, placental transfer, and lactational exposure to the child. We performed 10,000 Monte Carlo simulations where half-life, milk:plasma partition coefficient, and volume of distribution were varied. Children's dose and plasma levels were compared to their mother's by calculating child:mother dose ratios and plasma level ratios. We then evaluated the association between the three chemical-specific pharmacokinetic parameters and child:mother dose and level ratios through linear regression and decision trees. Our analyses revealed that half-life was the most influential parameter on children's lactational dose and plasma concentrations, followed by milk:plasma partition coefficient and volume of distribution. In bivariate regression analyses, half-life explained 72% of child:mother dose ratios and 53% of child:mother level ratios. Decision trees aiming to identify chemicals with high potential for lactational exposure (ratio>1) had an accuracy of 89% for child:mother dose ratios and 84% for child:mother level ratios. Our study showed the relative importance of half-life, milk:plasma partition coefficient, and volume of distribution on children's lactational exposure. Developed equations and decision trees will enable the rapid identification of chemicals with a high potential for lactational exposure.

Towards cheminformatics-based estimation of drug therapeutic index: Predicting the protective index of anticonvulsants using a new quantitative structure-index relationship approach

The overall efficacy and safety profile of a new drug is partially evaluated by the therapeutic index in clinical studies and by the protective index (PI) in preclinical studies. In-silico predictive methods may facilitate the assessment of these indicators. Although QSAR and QSTR models can be used for predicting PI, their predictive capability has not been evaluated. To test this capability, we developed QSAR and QSTR models for predicting the activity and toxicity of anticonvulsants at accuracy levels above the literature-reported threshold (LT) of good QSAR models as tested by both the internal 5-fold cross validation and external validation method. These models showed significantly compromised PI predictive capability due to the cumulative errors of the QSAR and QSTR models. Therefore, in this investigation a new quantitative structure-index relationship (QSIR) model was devised and it showed improved PI predictive capability that superseded the LT of good QSAR models. The QSAR, QSTR and QSIR models were developed using support vector regression (SVR) method with the parameters optimized by using the greedy search method. The molecular descriptors relevant to the prediction of anticonvulsant activities, toxicities and PIs were analyzed by a recursive feature elimination method. The selected molecular descriptors are primarily associated with the drug-like, pharmacological and toxicological features and those used in the published anticonvulsant QSAR and QSTR models. This study suggested that QSIR is useful for estimating the therapeutic index of drug candidates.

Modeling and Prediction of Solvent Effect on Human Skin Permeability using Support Vector Regression and Random Forest

PURPOSE

The solvent effect on skin permeability is important for assessing the effectiveness and toxicological risk of new dermatological formulations in pharmaceuticals and cosmetics development. The solvent effect occurs by diverse mechanisms, which could be elucidated by efficient and reliable prediction models. However, such prediction models have been hampered by the small variety of permeants and mixture components archived in databases and by low predictive performance. Here, we propose a solution to both problems.

METHODS

We first compiled a novel large database of 412 samples from 261 structurally diverse permeants and 31 solvents reported in the literature. The data were carefully screened to ensure their collection under consistent experimental conditions. To construct a high-performance predictive model, we then applied support vector regression (SVR) and random forest (RF) with greedy stepwise descriptor selection to our database. The models were internally and externally validated.

RESULTS

The SVR achieved higher performance statistics than RF. The (externally validated) determination coefficient, root mean square error, and mean absolute error of SVR were 0.899, 0.351, and 0.268, respectively. Moreover, because all descriptors are fully computational, our method can predict as-yet unsynthesized compounds.

CONCLUSION

Our high-performance prediction model offers an attractive alternative to permeability experiments for pharmaceutical and cosmetic candidate screening and optimizing skin-permeable topical formulations.

In silico optimization of pharmacokinetic properties and receptor binding affinity simultaneously: a 'parallel progression approach to drug design' applied to β-blockers

The present work exploits the potential of in silico approaches for minimizing attrition of leads in the later stages of drug development. We propose a theoretical approach, wherein 'parallel' information is generated to simultaneously optimize the pharmacokinetics (PK) and pharmacodynamics (PD) of lead candidates. β-blockers, though in use for many years, have suboptimal PKs; hence are an ideal test series for the 'parallel progression approach'. This approach utilizes molecular modeling tools viz. hologram quantitative structure activity relationships, homology modeling, docking, predictive metabolism, and toxicity models. Validated models have been developed for PK parameters such as volume of distribution (log Vd) and clearance (log Cl), which together influence the half-life (t1/2) of a drug. Simultaneously, models for PD in terms of inhibition constant pKi have been developed. Thus, PK and PD properties of β-blockers were concurrently analyzed and after iterative cycling, modifications were proposed that lead to compounds with optimized PK and PD. We report some of the resultant re-engineered β-blockers with improved half-lives and pKi values comparable with marketed β-blockers. These were further analyzed by the docking studies to evaluate their binding poses. Finally, metabolic and toxicological assessment of these molecules was done through in silico methods. The strategy proposed herein has potential universal applicability, and can be used in any drug discovery scenario; provided that the data used is consistent in terms of experimental conditions, endpoints, and methods employed. Thus the 'parallel progression approach' helps to simultaneously fine-tune various properties of the drug and would be an invaluable tool during the drug development process.

Prediction of human clearance based on animal data and molecular properties

Human clearance is often predicted prior to clinical study from in vivo preclinical data by virtue of interspecies allometric scaling methods. The aims of this study were to determine the important molecular descriptors for the extrapolation of animal data to human clearance and further to build a model to predict human clearance by combination of animal data and the selected molecular descriptors. These important molecular descriptors selected by genetic algorithm (GA) were from five classes: quantum mechanical, shadow indices, E-state keys, molecular properties, and molecular property counts. Although the data set contained many outliers determined by the conventional Mahmood method, the variation of most outliers was reduced significantly by our final support vector machine (SVM) model. The values of cross-validated correlation coefficient and root-mean-squared error (RMSE) for leave-one-out cross-validation (LOOCV) of the final SVM model were 0.783 and 0.305, respectively. Meanwhile, the reliability and consistency of the final model were also validated by an external test set. In conclusion, the SVM model based on the molecular descriptors selected by GA and animal data achieved better prediction performance than the Mahmood method. This approach can be applied as an improved interspecies allometric scaling method in drug research and development.

pkCSM: Predicting Small-Molecule Pharmacokinetic and Toxicity Properties Using Graph-Based Signatures

Drug development has a high attrition rate, with poor pharmacokinetic and safety properties a significant hurdle. Computational approaches may help minimize these risks. We have developed a novel approach (pkCSM) which uses graph-based signatures to develop predictive models of central ADMET properties for drug development. pkCSM performs as well or better than current methods. A freely accessible web server (http://structure.bioc.cam.ac.uk/pkcsm), which retains no information submitted to it, provides an integrated platform to rapidly evaluate pharmacokinetic and toxicity properties.

Machine learning methods in chemoinformatics

Machine learning algorithms are generally developed in computer science or adjacent disciplines and find their way into chemical modeling by a process of diffusion. Though particular machine learning methods are popular in chemoinformatics and quantitative structure-activity relationships (QSAR), many others exist in the technical literature. This discussion is methods-based and focused on some algorithms that chemoinformatics researchers frequently use. It makes no claim to be exhaustive. We concentrate on methods for supervised learning, predicting the unknown property values of a test set of instances, usually molecules, based on the known values for a training set. Particularly relevant approaches include Artificial Neural Networks, Random Forest, Support Vector Machine, k-Nearest Neighbors and naïve Bayes classifiers.

A Novel Strategy of Structural Similarity Based Consensus Modeling

A novel strategy of "structural similarity based consensus modeling" (SSCM) based on "model distance and guided model selection" (MD-QGMS) submodel set was proposed. The SSCM strategy is built upon a hypothesis, that is, similar compounds are most probably predicted more accurately by a same submodel among a model population, which can be concluded from the fact that models employing a different set of descriptors can predict compounds with specific structures more accurately. It is proved that the proposed SSCM strategy can remarkably improve the external prediction ability of QSAR models by employing two different datasets. In future, the proposed SSCM strategy may provide a new direction to develop more accurate predictive models.

Quorumpeps database: chemical space, microbial origin and functionality of quorum sensing peptides

Quorum-sensing (QS) peptides are biologically attractive molecules, with a wide diversity of structures and prone to modifications altering or presenting new functionalities. Therefore, the Quorumpeps database (http://quorumpeps.ugent.be) is developed to give a structured overview of the QS oligopeptides, describing their microbial origin (species), functionality (method, result and receptor), peptide links and chemical characteristics (3D-structure-derived physicochemical properties). The chemical diversity observed within this group of QS signalling molecules can be used to develop new synthetic bio-active compounds.

Quorumpeps database: chemical space, microbial origin and functionality of quorum sensing peptides

Quorum-sensing (QS) peptides are biologically attractive molecules, with a wide diversity of structures and prone to modifications altering or presenting new functionalities. Therefore, the Quorumpeps database (http://quorumpeps.ugent.be) is developed to give a structured overview of the QS oligopeptides, describing their microbial origin (species), functionality (method, result and receptor), peptide links and chemical characteristics (3D-structure-derived physicochemical properties). The chemical diversity observed within this group of QS signalling molecules can be used to develop new synthetic bio-active compounds.

Prediction of pharmacokinetic parameters

In silico tools specifically developed for prediction of pharmacokinetic parameters are of particular interest to pharmaceutical industry because of the high potential of discarding inappropriate molecules during an early stage of drug development itself with consequent saving of vital resources and valuable time. The ultimate goal of the in silico models of absorption, distribution, metabolism, and excretion (ADME) properties is the accurate prediction of the in vivo pharmacokinetics of a potential drug molecule in man, whilst it exists only as a virtual structure. Various types of in silico models developed for successful prediction of the ADME parameters like oral absorption, bioavailability, plasma protein binding, tissue distribution, clearance, half-life, etc. have been briefly described in this chapter.

QSARs for PBPK modelling of environmental contaminants

Physiologically-based pharmacokinetic (PBPK) models are increasingly finding use in risk assessment applications of data-rich compounds. However, it is a challenge to determine the chemical-specific parameters for these models, particularly in time- and resource-limiting situations. In this regard, SARs, QSARs and QPPRs are potentially useful for computing the chemical-specific input parameters of PBPK models. Based on the frequency of occurrence of molecular fragments (CH(3), CH(2), CH, C, C=C, H, benzene ring and H in benzene ring structure) and exposure conditions, the available QSAR-PBPK models facilitate the simulation of tissue and blood concentrations for some inhaled volatile organic chemicals. The application domain of existing QSARs for developing PBPK models is limited, due to lack of relevant data for diverse chemicals and mechanisms. Even though this approach is conceptually applicable to non-volatile and high molecular weight organics as well, it is more challenging to predict the other PBPK model parameters required for modelling the kinetics of these chemicals (particularly tissue diffusion coefficients, association constants for binding and oral absorption rates). As the level of our understanding of the mechanistic basis of toxicokinetic processes improves, QSARs to provide a priori predictions of key chemical-specific PBPK parameters can be developed to expedite the internal dose-based health risk assessments in data-poor situations.

Toxicokinetic Considerations in Predicting Toxicity

The ability of a compound to elicit a toxic effect within an organism is dependent upon three factors (i) the external exposure of the organism to the toxicant in the environment or via the food chain (ii) the internal uptake of the compound into the organism and its transport to the site of action in sufficient concentration and (iii) the inherent toxicity of the compound. The in silico prediction of toxicity and the role of external exposure have been dealt with in other chapters of this book. This chapter focuses on the importance of ‘internal exposure’ i.e. the absorption, distribution, metabolism and elimination (ADME) properties of compounds which determine their toxicokinetic profile. An introduction to key concepts in toxicokinetics will be provided, along with examples of modelling approaches and software available to predict these properties. A brief introduction will also be given into the theory of physiologically-based toxicokinetic modelling.

Predicting inhibitors of acetylcholinesterase by regression and classification machine learning approaches with combinations of molecular descriptors

PURPOSE

Acetylcholinesterase (AChE) is both a therapeutic target for Alzheimer's disease and a target for organophosphorus, carbamates and chemical warfare agents. Prediction of the likelihood of compounds interacting with this enzyme is therefore important from both therapeutic and toxicological perspectives.

MATERIALS AND METHODS

Support vector machine classification and regression models with molecular descriptors derived from Shape Signatures and the Molecular Operating Environment (MOE) application software were built and tested using a set of piperidine AChE inhibitors (N = 110).

RESULTS

The combination of the alignment free Shape Signatures and 2D MOE descriptors with the Support Vector Regression method outperforms the models based solely on 2D and internal 3D (i3D) MOE descriptors, and is comparable with the best previously reported PLS model based on CoMFA molecular descriptors (r(2)(test,SVR) = 0.48 vs. r(2)(test,PLS) = 0.47 from Sutherland et al. J Med Chem 47:5541-5554, 2004). Support Vector Classification algorithms proved superior to a classifier based on scores from the molecular docking program GOLD, with the overall prediction accuracies being Q(SVC(10CV)) = 74% and Q(SVC(LNO)) = 67% vs. Q(GOLD) = 56%.

CONCLUSIONS

These new machine learning models with combined descriptor schemes may find utility for predicting novel AChE inhibitors.

Current mathematical methods used in QSAR/QSPR studies

This paper gives an overview of the mathematical methods currently used in quantitative structure-activity/property relationship (QASR/QSPR) studies. Recently, the mathematical methods applied to the regression of QASR/QSPR models are developing very fast, and new methods, such as Gene Expression Programming (GEP), Project Pursuit Regression (PPR) and Local Lazy Regression (LLR) have appeared on the QASR/QSPR stage. At the same time, the earlier methods, including Multiple Linear Regression (MLR), Partial Least Squares (PLS), Neural Networks (NN), Support Vector Machine (SVM) and so on, are being upgraded to improve their performance in QASR/QSPR studies. These new and upgraded methods and algorithms are described in detail, and their advantages and disadvantages are evaluated and discussed, to show their application potential in QASR/QSPR studies in the future.

Transdermal penetration behaviour of drugs: CART-clustering, QSPR and selection of model compounds

A set of 116 structurally very diverse compounds, mainly drugs, was characterized by 1630 molecular descriptors. The biological property modelled in this study was the transdermal permeability coefficient logK(p). The main objective was to find a limited set of suitable model compounds for skin penetration studies. The classification and regression trees (CART) approach was applied and the resulting groups were discussed in terms of their role as possible model compounds and their determining descriptors. A second objective was to model transdermal penetration as a function of selected descriptors in quantitative structure-property relationships (QSPR) using a boosted CART (BRT) approach and multiple linear regression (MLR) analysis, where regression models were obtained by stepwise selection of the best descriptors. Evaluation of the standard statistical, as well as descriptor-number dependent, regression quality attributes yielded a maximal 10-dimensional MLR model. The CART and MLR models were subjected to an external validation with a test set of 12 compounds, not included in the original learning set of 104 compounds, to assess the predictive power of the models.

Quantitative structure-pharmacokinetic/pharmacodynamic relationships

Quantitative structure-activity relationships have long been considered a vital component of drug discovery and development, providing insight into the role of molecular properties in the biological activity of similar and unrelated compounds. Recognition that in vitro bioassay and/or pre-clinical activity are insufficient for anticipating which compounds are suitable leads for further development has shifted the focus toward integrated pharmacokinetic (PK) and pharmacodynamic (PD) processes. Over the last decade, considerable progress has been made in constructing empirical and mechanistic quantitative structure-PK relationships (QSPKR), as well as diverse mechanism-based pharmacodynamic models of drug effects. In this review, traditional and contemporary approaches to developing QSPKR models are discussed, along with selected examples of attempts to couple QSPKR and pharmacodynamic models to anticipate the intensity and time-course of the pharmacological effects of new or related compounds, or quantitative structure-pharmacodynamic relationships modeling. Such models are in accordance with the goals of systems biology and the ideal of designing drugs and delivery systems from first principles.