ADME evaluation in drug discovery. 7. Prediction of oral absorption by correlation and classification

Predictive cheminformatics in drug discovery: statistical modeling for analysis of micro-array and gene expression data

The vast amounts of chemical and biological data available through robotic high-throughput assays and micro-array technologies require computational techniques for visualization, analysis, and predictive -modeling. Predictive cheminformatics and bioinformatics employ statistical methods to mine this data for hidden correlations and to retrieve molecules or genes with desirable biological activity from large databases, for the purpose of drug development. While many statistical methods are commonly employed and widely accessible, their proper use involves due consideration to data representation and preprocessing, model validation and domain of applicability estimation, similarity assessment, the nature of the structure-activity landscape, and model interpretation. This chapter seeks to review these considerations in light of the current state of the art in statistical modeling and to summarize the best practices in predictive cheminformatics.

Optimization of Pharmacokinetics through Manipulation of Physicochemical Properties in a Series of HCV Inhibitors

A novel series of HCV replication inhibitors based on a pyrido[3,2-d]pyrimidine core were optimized for pharmacokinetics (PK) in rats. Several associations between physicochemical properties and PK were identified and exploited to guide the design of compounds. In addition, a simple new metric that may aid in the prediction of bioavailability for compounds with higher polar surface area is described (3*HBD-cLogP).

ADME evaluation in drug discovery. 9. Prediction of oral bioavailability in humans based on molecular properties and structural fingerprints

Oral bioavailability is an essential parameter in drug screening cascades and a good indicator of the capability of the delivery of a given compound to the systemic circulation by oral administration. In the present work, we report a database of oral bioavailability of 1014 molecules determined in humans. A systematic examination of the relationships between various physicochemical properties and oral bioavailability were carried out to investigate the influence of these properties on oral bioavailability. A number of property-based rules for bioavailability classification were generated and evaluated. We found that no rule was an effective predictor for oral bioavailability because these simple rules cannot characterize the influence of important metabolic processes on bioavailability. Finally, the genetic function approximation (GFA) technique was employed to construct the multiple linear regression models for oral bioavailability using structural fingerprints as the basic parameters, together with several important molecular properties. The best model is able to predict human oral bioavailability with an r of 0.79, a q of 0.72, and a RMSE (root-mean-square error) of 22.30% of the compounds from the training set. The analysis of the descriptors chosen by GFA shows that the important structural fingerprints are primarily related to important intestinal absorption and well-known metabolic processes. The predictive power of the models was further evaluated using a separate test set of 80 compounds, and the consensus model can predict the oral bioavailability with r(test) = 0.71 and RMSE = 23.55% for the tested compounds. Since the necessary molecular properties and structural fingerprints can be calculated easily and quickly, the models we proposed here may help speed up the process of finding or designing compounds with improved oral bioavailability.

Influence of molecular properties on oral bioavailability of lipophilic drugs - mapping of bulkiness and different measures of polarity

The biopharmaceutical assessment of new drug candidates based on their chemical structure is important in drug discovery and development. The scope of this study is to focus on lipophilic drugs and to clarify the role of their chemical predictors on oral bioavailability in humans. First their chemical properties were calculated from molecular modeling and the bioavailability data was obtained from the literature. The data was then analyzed by a partial least square method including non-linear terms. Significant coefficients were identified from a group of polarity- and solubility-related properties. Contour plots were constructed mapping molecular weight together with different polarity factors. Depending on the molecular weight a maximal bioavailability was found at solubility parameters of about 31-35 (J/cm(3))(1/2) and HLB values of roughly 4-12. The mapping of lipophilic drugs also revealed that a solubility parameter of less than 20 (J/cm(3))(1/2) or a HLB of smaller than unity is critical for the drug-likeness of new compounds.

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.

Estimation of ADME properties with substructure pattern recognition

Over the past decade, absorption, distribution, metabolism, and excretion (ADME) property evaluation has become one of the most important issues in the process of drug discovery and development. Since in vivo and in vitro evaluations are costly and laborious, in silico techniques had been widely used to estimate ADME properties of chemical compounds. Traditional prediction methods usually try to build a functional relationship between a set of molecular descriptors and a given ADME property. Although traditional methods have been successfully used in many cases, the accuracy and efficiency of molecular descriptors must be concerned. Herein, we report a new classification method based on substructure pattern recognition, in which each molecule is represented as a substructure pattern fingerprint based on a predefined substructure dictionary, and then a support vector machine (SVM) algorithm is applied to build the prediction model. Therefore, a direct connection between substructures and molecular properties is built. The most important substructure patterns can be identified via the information gain analysis, which could help to interpret the models from a medicinal chemistry perspective. Afterward, this method was verified with two data sets, one for blood-brain barrier (BBB) penetration and the other for human intestinal absorption (HIA). The results demonstrated that the overall predictive accuracies of the best HIA model for the training and test sets were 98.5 and 98.8%, and the overall predictive accuracies of the best BBB model for the training and test sets were 98.8 and 98.4%, which confirmed the reliability of our method. In the additional validations, the predictive accuracies were 94 and 69.5% for the HIA and the BBB models, respectively. Moreover, some of the representative key substructure patterns which significantly correlated with the HIA and BBB penetration properties were also presented.

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.

Drug and drug candidate building block analysis

Drug likeness analysis is widely used in modern drug design. However, most drug likeness filters, represented by Lipinski's "Rule of 5", are based on drugs' simple structural features and some physiochemical properties. In this study, we conducted thorough structural analyses for two drug datasets. The first dataset, ADDS, is composed of 1240 FDA-approved drugs, and the second drug dataset, EDDS, is a nonredundant collection of FDA-approved drugs and experimental drugs in different phases of clinical trials from several drug databases (6932 entries). For each molecule, all possible fragments were enumerated using a brutal force approach. Three kinds of building blocks, namely, the drug scaffold, ring system, and the small fragment, were identified and ranked according to the frequencies of their occurrence in drug molecules. The major finding is that most top fragments are essentially common for both drug datasets; the top 50 fragments cover 52.6% and 48.6% drugs for ADDS and EDDS, respectively. The identified building blocks were further ranked according to their relative hit rates in the drug datasets and in a screening dataset, which is a nonredundant collection of screening compounds from many resources. In comparison with the previous reports in the field, we have identified many more high-quality building blocks. The results obtained in this study could provide useful hints to medicinal chemists in designing drug-like compounds as well as prioritizing screening libraries to filter out those molecules lack of functional building blocks.

Ionization-specific analysis of human intestinal absorption

This study presents a mechanistic QSAR analysis of human intestinal absorption of drugs and drug-like compounds using a data set of 567 %HIA values. Experimental data represent passive diffusion across intestinal membranes, and are considered to be reasonably free of carrier-mediated transport or other unwanted effects. A nonlinear model was developed relating %HIA to physicochemical properties of drugs (lipophilicity, ionization, hydrogen bonding, and molecular size). The model describes ion-specific intestinal permeability of drugs by both transcellular and paracellular routes, and also accounts for unstirred water layer effects. The obtained model was validated on two external data sets consisting of in vivo human jejunal permeability coefficients (P(eff)) and absorption rate constants (K(a)). Validation results demonstrate good predictive power of the model (RMSE = 0.35-0.45 log units for log K(a) and log P(eff)). High prediction accuracy together with clear physicochemical interpretation (log P, pK(a)) makes this model particularly suitable for use in property-based drug design.

SAR of a series of inhaled A(2A) agonists and comparison of inhaled pharmacokinetics in a preclinical model with clinical pharmacokinetic data

COPD is a major cause of mortality in the western world. A(2A) agonists are postulated to reduce the lung inflammation that causes COPD. The cardiovascular effects of A(2A) agonists dictate that a compound needs to be delivered by inhalation to be therapeutically useful. The pharmacological and pharmacokinetic SAR of a series of inhaled A(2A) agonists is described leading through to human pharmacokinetic data for a clinical candidate.

A modified uncorrelated linear discriminant analysis model coupled with recursive feature elimination for the prediction of bioactivity

To meet the requirements of providing accurate, robust, and interpretable prediction of bioactivity, a modified uncorrelated linear discriminant analysis (M-ULDA) model was developed. In addition, a feature selection method called recursive feature elimination (RFE), originally used for support vector machine (SVM), was introduced and modified to fit the scheme of ULDA. From the evaluation of six pharmaceutical datasets, the M-UDLA coupled with RFE showed better or comparable classification accuracy with respect to other well-studied methods such as SVM and decision trees. The RFE used for ULDA has the advantage of increasing the computational speed and provides useful insights into biochemical mechanisms related to pharmaceutical activity by significantly reducing the number of variables used for the final model.

Structure-ADME relationship: still a long way to go?

BACKGROUND

Theoretical models for predicting absorption, distribution, metabolism and excretion (ADME) properties play increasingly important roles in support of the drug development process.

OBJECTIVE

We briefly review the in silico prediction models for three important ADME properties, namely, aqueous solubility, human intestinal absorption, and oral bioavailability.

METHODS

Rather than giving detailed descriptions of the ADME prediction models, we focus on the discussions of the prediction accuracies of the in silico models.

RESULTS/CONCLUSION

We find that the robustness and predictive capability of the ADME models are directly associated with the complexity of the ADME property. For the ADME properties involving complex phenomena, such as bioavailability, the in silico models usually cannot give satisfactory predictions. Moreover, the lack of large and high-quality data sets also greatly hinder the reliability of ADME predictions. While considerable progress has been achieved in ADME predictions, many challenges remain to be overcome.

Prediction of human intestinal absorption by GA feature selection and support vector machine regression

QSAR (Quantitative Structure Activity Relationships) models for the prediction of human intestinal absorption (HIA) were built with molecular descriptors calculated by ADRIANA.Code, Cerius² and a combination of them. A dataset of 552 compounds covering a wide range of current drugs with experimental HIA values was investigated. A Genetic Algorithm feature selection method was applied to select proper descriptors. A Kohonen's self-organizing Neural Network (KohNN) map was used to split the whole dataset into a training set including 380 compounds and a test set consisting of 172 compounds. First, the six selected descriptors from ADRIANA.Code and the six selected descriptors from Cerius² were used as the input descriptors for building quantitative models using Partial Least Square (PLS) analysis and Support Vector Machine (SVM) Regression. Then, another two models were built based on nine descriptors selected by a combination of ADRIANA.Code and Cerius² descriptors using PLS and SVM, respectively. For the three SVM models, correlation coefficients (r) of 0.87, 0.89 and 0.88 were achieved; and standard deviations (s) of 10.98, 9.72 and 9.14 were obtained for the test set.

Drug absorption modeling as a tool to define the strategy in clinical formulation development

The purpose of this mini review is to discuss the use of physiologically-based drug absorption modeling to guide the formulation development. Following an introduction to drug absorption modeling, this article focuses on the preclinical formulation development. Case studies are presented, where the emphasis is not only the prediction of absolute exposure values, but also their change with altered input values. Sensitivity analysis of technologically relevant parameters, like the drug's particle size, dose and solubility, is presented as the basis to define the clinical formulation strategy. Taking the concept even one step further, the article shows how the entire design space for drug absorption can be constructed. This most accurate prediction level is mainly foreseen once clinical data is available and an example is provided using mefenamic acid as a model drug. Physiologically-based modeling is expected to be more often used by formulators in the future. It has the potential to become an indispensable tool to guide the formulation development of challenging drugs, which will help minimize both risks and costs of formulation development.

Explorations into modeling human oral bioavailability

Explorations into modeling human oral bioavailability started with a whole dataset of 772 drug compounds. First, training set and test set were chosen based on Kohonen's self-organizing Neural Network (KohNN). Then, a quantitative model of the whole dataset was built using multiple linear regression (MLR) analysis. This model had limited predictability emphasizing that a variety of pharmacokinetic factors influence human oral bioavailability. In order to explore whether better models can be built when the compounds share some ADME properties, four subsets were chosen from the whole dataset to build quantitative models and better models were obtained by MLR analysis. These studies show that, indeed, good models for predicting human oral bioavailability can be obtained from datasets sharing certain pharmacokinetic properties.

Hologram QSAR model for the prediction of human oral bioavailability

A drug intended for use in humans should have an ideal balance of pharmacokinetics and safety, as well as potency and selectivity. Unfavorable pharmacokinetics can negatively affect the clinical development of many otherwise promising drug candidates. A variety of in silico ADME (absorption, distribution, metabolism, and excretion) models are receiving increased attention due to a better appreciation that pharmacokinetic properties should be considered in early phases of the drug discovery process. Human oral bioavailability is an important pharmacokinetic property, which is directly related to the amount of drug available in the systemic circulation to exert pharmacological and therapeutic effects. In the present work, hologram quantitative structure-activity relationships (HQSAR) were performed on a training set of 250 structurally diverse molecules with known human oral bioavailability. The most significant HQSAR model (q(2)=0.70, r(2)=0.93) was obtained using atoms, bond, connection, and chirality as fragment distinction. The predictive ability of the model was evaluated by an external test set containing 52 molecules not included in the training set, and the predicted values were in good agreement with the experimental values. The HQSAR model should be useful for the design of new drug candidates having increased bioavailability as well as in the process of chemical library design, virtual screening, and high-throughput screening.

The influence of drug-like concepts on decision-making in medicinal chemistry

The application of guidelines linked to the concept of drug-likeness, such as the 'rule of five', has gained wide acceptance as an approach to reduce attrition in drug discovery and development. However, despite this acceptance, analysis of recent trends reveals that the physical properties of molecules that are currently being synthesized in leading drug discovery companies differ significantly from those of recently discovered oral drugs and compounds in clinical development. The consequences of the marked increase in lipophilicity--the most important drug-like physical property--include a greater likelihood of lack of selectivity and attrition in drug development. Tackling the threat of compound-related toxicological attrition needs to move to the mainstream of medicinal chemistry decision-making.