Trends and plot methods in MLR studies

Extra precision glide docking, free energy calculation and molecular dynamics studies of 1,2-diarylethane derivatives as potent urease inhibitors

For the latter half of the twentieth century, most medical professionals considered bacterial infection to be a primary cause of gastrointestinal ulcers in human beings. In 1994, the World Health Organization (WHO) recognized Helicobacter pylori, the bacterium most closely linked to ulcer development, as a type I carcinogen. Biological research has shown that there is a positive correlation between the number of species in the Helicobacter genus and the number of medical conditions associated with Helicobacter infection, both of which are increasing rapidly. N-Benzylaniline derivatives, frequently used in industrial manufacturing, are being considered as a strong candidate for ongoing drug modeling in search of novel therapies. The basic goal behind this study was to determine the potency of experimentally proved data, and to determine favorable substituents to enhance potency, and thereafter to support this finding through theoretical modification of the existing base skeleton by addition of suitable substituents. Ligands were investigated thoroughly by paying attention to the urease-inhibitory properties present in the selected series. Initially, docking was performed on ligands with protein to produce efficient docking poses. Molecular dynamics (MD) simulations were also performed to precisely understand the interactions between ligands and proteins. Thereafter, MM-GBSA was used in order to validate the methods and results. Good interaction was observed with amino acids Arg338, Ala169, Asp223, His322, and Asn168. This study also revealed that the electron rich hydroxyl group (-OH) substituent plays an important role during bond formation. In addition, various hydrogen bonds, ionic bonds, and pi-pi stacking bonds make significant contributions towards urease inhibition. Therefore, further research utilizing electron-rich moieties may lead to novel and efficacious urease inhibitors.

Beware of R(2): Simple, Unambiguous Assessment of the Prediction Accuracy of QSAR and QSPR Models

The statistical metrics used to characterize the external predictivity of a model, i.e., how well it predicts the properties of an independent test set, have proliferated over the past decade. This paper clarifies some apparent confusion over the use of the coefficient of determination, R(2), as a measure of model fit and predictive power in QSAR and QSPR modeling. R(2) (or r(2)) has been used in various contexts in the literature in conjunction with training and test data for both ordinary linear regression and regression through the origin as well as with linear and nonlinear regression models. We analyze the widely adopted model fit criteria suggested by Golbraikh and Tropsha ( J. Mol. Graphics Modell. 2002 , 20 , 269 - 276 ) in a strict statistical manner. Shortcomings in these criteria are identified, and a clearer and simpler alternative method to characterize model predictivity is provided. The intent is not to repeat the well-documented arguments for model validation using test data but rather to guide the application of R(2) as a model fit statistic. Examples are used to illustrate both correct and incorrect uses of R(2). Reporting the root-mean-square error or equivalent measures of dispersion, which are typically of more practical importance than R(2), is also encouraged, and important challenges in addressing the needs of different categories of users such as computational chemists, experimental scientists, and regulatory decision support specialists are outlined.

Real external predictivity of QSAR models. Part 2. New intercomparable thresholds for different validation criteria and the need for scatter plot inspection

The evaluation of regression QSAR model performance, in fitting, robustness, and external prediction, is of pivotal importance. Over the past decade, different external validation parameters have been proposed: Q(F1)(2), Q(F2)(2), Q(F3)(2), r(m)(2), and the Golbraikh-Tropsha method. Recently, the concordance correlation coefficient (CCC, Lin), which simply verifies how small the differences are between experimental data and external data set predictions, independently of their range, was proposed by our group as an external validation parameter for use in QSAR studies. In our preliminary work, we demonstrated with thousands of simulated models that CCC is in good agreement with the compared validation criteria (except r(m)(2)) using the cutoff values normally applied for the acceptance of QSAR models as externally predictive. In this new work, we have studied and compared the general trends of the various criteria relative to different possible biases (scale and location shifts) in external data distributions, using a wide range of different simulated scenarios. This study, further supported by visual inspection of experimental vs predicted data scatter plots, has highlighted problems related to some criteria. Indeed, if based on the cutoff suggested by the proponent, r(m)(2) could also accept not predictive models in two of the possible biases (location, location plus scale), while in the case of scale shift bias, it appears to be the most restrictive. Moreover, Q(F1)(2) and Q(F2)(2) showed some problems in one of the possible biases (scale shift). This analysis allowed us to also propose recalibrated, and intercomparable for the same data scatter, new thresholds for each criterion in defining a QSAR model as really externally predictive in a more precautionary approach. An analysis of the results revealed that the scatter plot of experimental vs predicted external data must always be evaluated to support the statistical criteria values: in some cases high statistical parameter values could hide models with unacceptable predictions.

Model of twelve properties of a set of organic solvents with graph-theoretical and/or experimental parameters

Twelve properties of a highly heterogeneous class of organic solvents have been modeled with a graph-theoretical molecular connectivity modified (MC) method, which allows to encode the core electrons and the hydrogen atoms. The graph-theoretical method uses the concepts of simple, general, and complete graphs, where these last types of graphs are used to encode the core electrons. The hydrogen atoms have been encoded by the aid of a graph-theoretical perturbation parameter, which contributes to the definition of the valence delta, delta(v), a key parameter in molecular connectivity studies. The model of the twelve properties done with a stepwise search algorithm is always satisfactory, and it allows to check the influence of the hydrogen content of the solvent molecules on the choice of the type of descriptor. A similar argument holds for the influence of the halogen atoms on the type of core electron representation. In some cases the molar mass, and in a minor way, special "ad hoc" parameters have been used to improve the model. A very good model of the surface tension could be obtained by the aid of five experimental parameters. A mixed model method based on experimental parameters plus molecular connectivity indices achieved, instead, to consistently improve the model quality of five properties. To underline is the importance of the boiling point temperatures as descriptors in these last two model methodologies.