QSAR analysis of hypoglycemic agents using the topological indices

QSAR study of some anti-hyperglycaemic sulphonylurea drugs

Sulphonylureas are widely used anti-hyperglycaemic drugs for the treatment of type 2 diabetes. The only published quantitative structure-activity relationship (QSAR) models for sulphonylurea drugs have been found to be questionable, for a number of reasons. We have re-analysed the human anti-hyperglycaemic potencies, acute mouse intraperitoneal toxicities (LD50) and plasma protein-binding abilities of the 15 drugs using multiple linear regression and obtained good QSAR models for each endpoint. The obtained QSARs all comply well with the Organisation for Economic Co-operation and Development (OECD) Guidelines for the Validation of (Q)SARs. We could not carry out external validation of our models for acute toxicity and plasma protein-binding because of the very small datasets available.

Trends and plot methods in MLR studies

Regression toward the mean effects are presented within the field of quantitative structure-activity relationship modeling and in situations in which multilinear regression techniques are considered for model building. The concept is related to the graphical aspect of some scatter plots (experimental vs fitted and fitted vs experimental values). These graphs demonstrate how the point cloud is not always symmetrically distributed along the so-called "ideal" or "desired" line, that is, the bisector of the first and third quadrants. The deviation from the ideal line is fixed, and it is also related to the coefficient of determination. An extrapolation of these regression effects is also discussed within the context of property predictions obtained via the leave-one-out cross-validation technique.

On the applicability of QSAR for recognition of miRNA bioorganic structures at early stages of organism and cell development: Embryo and stem cells

Quantitative structure-activity-relationship (QSAR) models have application in bioorganic chemistry mainly to the study of small sized molecules while applications to biopolymers remain not very developed. MicroRNAs (miRNAs), which are non-coding small RNAs, regulate a variety of biological processes and constitute good candidates to scale up the application of QSAR to biopolymers. The propensity of a small RNA sequence to act as miRNA depends on its secondary structure, which one can explain in terms of folding thermodynamic parameters. Then, thermodynamic QSAR can be used, for instance, for fast identification of miRNAs at early stages of development such as embryos and stem cells (called here esmiRNAs), and gain clarity inside cellular differentiation processes and diseases such as cancer. First, we calculated folding free energies (DeltaG), enthalpies (DeltaH), and entropies (DeltaS) as well as melting temperatures (T(m)) for 2623 small RNA sequences (including 623 esmiRNAs and 2000 negative control sequences). Next, we seek a QSAR classification model: esmiRNA=0.035 x T(m)-0.078 x DeltaS-8.748. The model correctly recognized 543 (87.2%) of esmiRNAs and 935 (93.5%) of non-esmiRNAs divided into both training and validation series. The model also recognized 908 out of 1000 additional negative control sequences. ROC curve analysis (area=0.93) demonstrated that the present model significantly differentiates from a random classifier. In addition, we map the influence of thermodynamic parameters over esmiRNA activity. Last, a double ordinate Cartesian plot of cross-validated residuals (first ordinate), standard residuals (second ordinate), and leverages (abscissa) defined the domain of applicability of the model as a squared area within +/-2 band for residuals and a leverage threshold of h=0.0074. The present is the first QSAR model for quickly accurate selection of new esmiRNAs with potential use in bioorganic and medicinal chemistry.

Pharmacokinetic parameter prediction from drug structure using artificial neural networks

Simple methods for determining the human pharmacokinetics of known and unknown drug-like compounds is a much sought-after goal in the pharmaceutical industry. The current study made use of artificial neural networks (ANNs) for the prediction of clearances, fraction bound to plasma proteins, and volume of distribution of a series of structurally diverse compounds. A number of theoretical descriptors were generated from the drug structures and both automated and manual pruning were used to derive optimal subsets of descriptors for quantitative structure-pharmacokinetic relationship models. Models were trained on one set of compounds and validated with another. Absolute predicted ability was evaluated using a further independent test set of compounds. Correlations for test compounds ranged from 0.855 to 0.992. Predicted values agreed closely with experimental values for total clearance, renal clearance, and volume of distribution, while predictions for protein binding were encouraging. The combination of descriptor generation, ANNs, and the speed and success of this technique compared with conventional methods shows strong potential for use in pharmaceutical product development.

Classification of some active HIV-1 protease inhibitors and their inactive analogues using some uncorrelated three-dimensional molecular descriptors and a fuzzy c-means algorithm

A fuzzy c-means algorithm was used to classify some 3D convex hull descriptors computed for 345 active HIV-1 protease inhibitors collected from literature and 437 inactive analogues searched from the MDL/ISIS database. The number of descriptors used to represent each compound was from 4 to 8, and they were uncorrelated using the principal component analysis. These uncorrelated descriptors were then divided into two groups and classified by the fuzzy c-means algorithm. The classification produced a clear-cut switch in membership functions computed for each uncorrelated descriptor at the group boundary. Compounds with nonswitching membership functions computed were treated as outliers, and they were counted for estimating the accuracy of the classification. The averaged accuracy of classification for the active inhibitor set was about 80% which was better than that directly classified by a linear discriminant function on the original 3D convex hull descriptors. The whole classification scheme was also applied to several sets of some conventional descriptors computed for each compound, but the averaged accuracy was around 58%. Further classification using some 3D convex hull descriptors searched from comparing the distribution of these descriptors was performed on a new data set composed of 289 outliers-deducted active inhibitors and 63 outliers identified from the inactive analogues through previous classification. This final classification identified 19 inactive analogues which were similar in structural and topological features to those of some highly active inhibitors classified together with them.

Quantitative structure/property relationship analysis of Caco-2 permeability using a genetic algorithm-based partial least squares method

Caco-2 cell monolayers are widely used systems for predicting human intestinal absorption. This study was carried out to develop a quantitative structure-property relationship (QSPR) model of Caco-2 permeability using a novel genetic algorithm-based partial least squares (GA-PLS) method. The Caco-2 permeability data for 73 compounds were taken from the literature. Molconn-Z descriptors of these compounds were calculated as molecular descriptors, and the optimal subset of the descriptors was explored by GA-PLS analysis. A fitness function considering both goodness-of-fit to the training data and predictability of the testing data was adopted throughout the genetic algorithm-driven optimization procedure. The final PLS model consisting of 24 descriptors gave a correlation coefficient (r) of 0.886 for the entire dataset and a predictive correlation coefficient (r(pred)) of 0.825 that was evaluated by a leave-some-out cross-validation procedure. Thus, the GA-PLS analysis proved to be a reasonable QSPR modeling approach for predicting Caco-2 permeability.