Computational models to predict blood–brain barrier permeation and CNS activity

The blood-brain permeation of a structurally diverse set of 281 compounds was modeled using linear regression and a multivariate genetic partial least squares (G/PLS) approach. Key structural features affecting the logarithm of blood-brain partitioning (logBB) were captured through statistically significant quantitative structure-activity relationship (QSAR) models. These relationships reveal the importance of logP, polar surface area, and a variety of electrotopological indices for accurate predictions of logBB. The best models reveal an excellent correlation (r > 0.9) for a training set of 58 compounds. Likewise, the comparison of the average logBB values obtained from an ensemble of QSAR models with experimental values also verifies the statistical quality of the models (r > 0.9). The models provide good agreement (r approximately 0.7) between the predicted logBB values for 34 molecules in the external validation set and the experimental values. To further validate the models for use during the drug discovery process, a prediction set of 181 drugs with reported CNS penetration data was used. A >70% success rate is obtained by using any of the QSAR models in the qualitative prediction for CNS permeable (active) drugs. A lower success rate (approximately 60%) was obtained for the best model for CNS impermeable (inactive) drugs. Combining the predictions obtained from all the models (consensus) did not significantly improve the discrimination of CNS active and CNS inactive molecules. Finally, using the therapeutic classification as a guiding tool, the CNS penetration capability of over 2000 compounds in the Synthline database was estimated. The results were very similar to the smaller set of 181 compounds.     

Local and global quantitative structure-activity relationship modeling and prediction for the baseline toxicity

The predictive accuracy of the model is of the most concern for computational chemists in quantitative structure-activity relationship (QSAR) investigations. It is hypothesized that the model based on analogical chemicals will exhibit better predictive performance than that derived from diverse compounds. This paper develops a novel scheme called "clustering first, and then modeling" to build local QSAR models for the subsets resulted from clustering of the training set according to structural similarity. For validation and prediction, the validation set and test set were first classified into the corresponding subsets just as those of the training set, and then the prediction was performed by the relevant local model for each subset. This approach was validated on two independent data sets by local modeling and prediction of the baseline toxicity for the fathead minnow. In this process, hierarchical clustering was employed for cluster analysis, k-nearest neighbor for classification, and partial least squares for the model generation. The statistical results indicated that the predictive performances of the local models based on the subsets were much superior to those of the global model based on the whole training set, which was consistent with the hypothesis. This approach proposed here is promising for extension to QSAR modeling for various physicochemical properties, biological activities, and toxicities.     

Prediction of plasma protein binding of drugs using Kier–Hall valence connectivity indices and 4D-fingerprint molecular similarity analyses

Summary A 115 compound dataset for HSA binding is divided into the training set and the test set based on molecular similarity and cluster analyses. Both Kier–Hall valence connectivity indices and 4D-fingerprint similarity measures were applied to this dataset. Four different predictive schemes (SM, SA, SR, SC) were applied to the test set based on the similarity measures of each compound to the compounds in the training set. The first algorithmic scheme (SM) predicts the binding affinity of a test compound using only the most similar training set compound’s binding affinity. This scheme has relatively poor predictivity based both on Kier–Hall valence connectivity indices similarity measures and 4D-fingerprints similarity analyses. The other three algorithmic schemes (SM SR, SC), which assign a weighting coefficient to each of the top-ten most similar training set compounds, have reasonable predictivity of a test set. The algorithmic scheme which categorizes the most similar compounds into different weighted clusters predicts the test set best. The 4D-fingerprints provide 36 different individual IPE/IPE type molecular similarity measures. This study supports that some types of similarity measures are highly similar to one another for this dataset. Both the Kier–Hall valence connectivity indices similarity measures and the 4D-fingerprints have nearly same predictivity for this particular dataset.     

Predicting drug pharmacokinetic properties using molecular interaction fields and SIMCA

We have developed a method that combines molecular interaction fields with soft independent modeling of class analogy (SIMCA) Wold:1977 to predict pharmacokinetic drug properties. Several additional considerations to those made in traditional QSAR are required in order to develop a successful QSPR strategy that is capable of accommodating the many complex factors that contribute to key pharmacokinetic properties such as ADME (absorption, distribution, metabolism, and excretion) and toxicology. An accurate prediction of oral bioavailability, for example, requires that absorption and first-pass hepatic elimination both be taken into consideration. To accomplish this, general properties of molecules must be related to their solubility and ability to penetrate biological membranes, and specific features must be related to their particular metabolic and toxicological profiles. Here we describe a method, which is applicable to structurally diverse data sets while utilizing as much detailed structural information as possible. We address the issue of the molecular alignment of a structurally diverse set of compounds using idiotropic field orientation (IFO), a generalization of inertial field orientation Clark:1998. We have developed a second flavor of this method, which directly incorporates electrostatics into the molecular alignment. Both variations of IFO produce a characteristic orientation for each structure and the corresponding molecular fields can then be analyzed using SIMCA. Models are presented for human intestinal absorption, blood-brain barrier penetration and bioavailability to demonstrate ways in which this tool can be used early in the drug development process to identify leads likely to exhibit poor pharmacokinetic behavior in pre-clinical studies, and we have explored the influence of conformation and molecular field type on the statistical properties of the models obtained.     

Three useful dimensions for domain applicability in QSAR models using random forest

One popular metric for estimating the accuracy of prospective quantitative structure-activity relationship (QSAR) predictions is based on the similarity of the compound being predicted to compounds in the training set from which the QSAR model was built. More recent work in the field has indicated that other parameters might be equally or more important than similarity. Here we make use of two additional parameters: the variation of prediction among random forest trees (less variation among trees indicates more accurate prediction) and the prediction itself (certain ranges of activity are intrinsically easier to predict than others). The accuracy of prediction for a QSAR model, as measured by the root-mean-square error, can be estimated by cross-validation on the training set at the time of model-building and stored as a three-dimensional array of bins. This is an obvious extension of the one-dimensional array of bins we previously proposed for similarity to the training set [Sheridan et al. J. Chem. Inf. Comput. Sci.2004, 44, 1912-1928]. We show that using these three parameters simultaneously adds much more discrimination in prediction accuracy than any single parameter. This approach can be applied to any QSAR method that produces an ensemble of models. We also show that the root-mean-square errors produced by cross-validation are predictive of root-mean-square errors of compounds tested after the model was built.     

Performance evaluation of CAESAR–QSAR output using PAHs as a case study

The set of 16 polycyclic aromatic hydrocarbon compounds was examined with the Internet available quantitative structure–activity relationship (QSAR) CAESAR models. For mutagenicity, carcinogenicity, developmental toxicity, and skin sensitization, the report includes the predicted classifications, the analysis of applicability domains, and the similarity sets, which consist of the similar compounds from the training sets. These results were further analyzed with chemometrical methods, that is, hierarchical clustering, principal component analysis, and self‐organizing maps, which were used for clustering and to define the cluster indicators. Such analysis assists the users in planning the application of QSAR models for hazard communication in regulatory compliance and in research of new active compounds. Copyright © 2013 John Wiley & Sons, Ltd.     

QSAR models based on isomorphic and nonisomorphic data fusion for predicting the blood brain barrier permeability

A QSAR model for predicting the blood brain barrier permeability (BBBP) in a large and heterogeneous variety of compounds (136 compounds) has been developed using approximate similarity (AS) matrices as predictors and PLS as multivariate regression technique. AS values fuse information of both the isomorphic similarity and nonisomorphic dissimilarity with the purpose of achieving an accurate predictive space. In addition to the fact of applying AS values to heterogeneous data sets, a new concept on graph isomorphism based on the extended maximum common subgraph (EMCS) is defined for the building of AS spaces considering the atoms and bonds, which are bridges between the isomorphic and nonisomorphic substructures. This new isomorphism detection has as objective to take into account the position and nature of the nucleus substituents, thus allowing the development of accurate models for large and diverse sets of compounds. After an outliers study, the training and test stages were made and the results obtained using several AS approaches were compared. Several validation processes were carried out by means of employing several test sets, and high predictive ability was obtained for all the cases (Q(2) = 0.81 and standard error in prediction, SEP = 0.29).