Jesús Vicente de Julián-Ortiz

Search of Chemical Scaffolds for Novel Antituberculosis Agents

A method to identify chemical scaffolds potentially active against Mycobacterium tuberculosis is presented. The molecular features of a set of structurally heterogeneous antituberculosis drugs were coded by means of structural invariants. Three techniques were used to obtain equations able to model the antituberculosis activity: linear discriminant analysis, multilinear regression, and shrinkage estimation–ridge regression. The model obtained was statistically validated through leave-n-out test, and an external set and was applied to a database for the search of new active agents. The selected compounds were assayed in vitro, and among those identified as active stand reserpine, N,N,N′,N′-tetrakis-(2-pyridylmethyl)-ethylenediamine (TPEN), trifluoperazine, pentamidine, and 2-methyl-4,6-dinitro-phenol (DNOC). They show activity comparable to or superior to ethambutol, used in combination with other drugs for the prevention and treatment of Mycobacterium avium complex and drug-resistant tuberculosis.

Virtual generation of agents against Mycobacterium tuberculosis. A QSAR study

A QSAR approach based on the use of various topological indices as new theoretical molecular descriptors was applied to the study of a set of 64 anti-tuberculosis agents involving the substituted benzoxazines and phenylquinazolines. In order to evaluate the reliability of the proposed linear QSAR model, several statistical tests were proposed. The resulting model was subsequently applied to a wider virtual molecular library, which, together with the original set of 64 molecules with known activities contained another 512 molecules for which the predictions were made. Based on this prediction some new structures were proposed as especially promising candidates for active anti-tuberculotic drugs.

Testing selected optimal descriptors with artificial neural networks

Eleven properties have been modeled with the objective of checking the importance for model purposes of mixed descriptors made of empirical parameters, molecular connectivity indices and random numbers. The mixed descriptors with random indices have a descriptive character which is satisfactorily confirmed by the leave-one-out method of statistical analysis. The introduction of a partition of the set of compounds into training and evaluation sets decreases drastically the probability to find a mixed descriptor with random indices with good model quality. Two properties, the magnetic susceptibility and the elutropic values, insist on having optimal descriptors with random indices. The overall model study underlines the importance of semiempirical descriptors made of experimental parameters and molecular connectivity indices, as well as the importance of a perturbation parameter that has been introduced into the valence delta number to encode the contribution of the depleted hydrogen atoms. The use of complete graphs to encode the core electrons of higher-row atoms is also underlined. The model quality of the mixed descriptors obtained with combinatorial regressive methods has also been tested with three-layered feed-forward artificial neural network (ANN) methods. This methodology not only confirms the validity of the descriptors but also improves their model quality widening, thus, their predictive ability.

Applications of Chemoinformatics in Predictive Toxicology for Regulatory Purposes, Especially in the Context of the EU REACH Legislation

ChemoinformaticsmethodologiessuchasQSAR/QSPRhavebeenusedfordecadesindrugdiscovery projects,especiallyforthefindingofnewcompoundswiththerapeuticpropertiesandtheoptimization ofADMEpropertiesonchemicalseries.Theapplicationofcomputationaltechniquesinpredictive toxicologyismuchmorerecent,andtheyareexperiencinganincreasinglyinterestbecauseofthenew legalrequirementsimposedbynationalandinternationalregulations.Inthepharmaceuticalfield, theUSFoodandDrugAdministration(FDA)supporttheuseofpredictivemodelsforregulatory decision-makingwhenassessing thegenotoxic andcarcinogenicpotentialofdrug impurities. In Europe,theREACHlegislationpromotestheuseofQSARinordertoreducethehugeamountof animaltestingneededtodemonstratethesafetyofnewchemicalentitiessubjectedtoregistration, providedtheymeetspecificconditionstoensuretheirqualityandpredictivepower.Inthisreview, theauthorssummarizethestateofartofinsilicomethodsforregulatorypurposes,withespecial emphasisonQSARmodels. KEywoRdS Computational Toxicology, Docking, QSAR, REACH, Read-Across, Virtual Screening

Molecular rearrangement of an Aza-Scorpiand macrocycle induced by pH: A computational study

Rearrangements and their control are a hot topic in supramolecular chemistry due to the possibilities that these phenomena open in the design of synthetic receptors and molecular machines. Macrocycle aza-scorpiands constitute an interesting system that can reorganize their spatial structure depending on pH variations or the presence of metal cations. In this study, the relative stabilities of these conformations were predicted computationally by semi-empirical and density functional theory approximations, and the reorganization from closed to open conformations was simulated by using the Monte Carlo multiple minimum method.

QSPR with descriptors based on averages of vertex invariants. An artificial neural network study

New type of indices, the mean molecular connectivity indices (MMCI), based on nine different concepts of mean are proposed to model, together with molecular connectivity indices (MCI), experimental parameters and random variables, eleven properties of organic solvents. Two model methodologies are used to test the different descriptors: the multilinear least-squares (MLS) methodology and the Artificial Neural Network (ANN) methodology. The top three quantitative structure–property relationships (QSPR) for each property are chosen with the MLS method. The indices of these three QSPRs were used to train the ANNs that selected the best training sets of indices to estimate the evaluation sets of compounds. The best ANN relationships for most properties are of the semiempirical types that include mean molecular connectivity indices (MMCI), molecular connectivity indices (MCI) and experimental parameters. Refractive index, RI, viscosity, η, and surface tension, γ, prefer a semiempirical relationship made of MCI and an experimental parameter only. In our previous study with no MMCI, random variables contributed to semiempirical relationships for two properties at the ANN level (MS, and El), here the use of MMCIs undo the contribution of such variables. Most of the MMCIs that contribute to improve the model of the properties are valence-delta-dependent (δv), that is, they encode both the hydrogen atom contribution and the core electrons of higher-row atoms.

Discriminating Drug-Like Compounds by Partition Trees with Quantum Similarity Indices and Graph Invariants

The search for new drug candidates in databases is of paramount importance in pharmaceutical chemistry. The selection of molecular subsets is greatly optimized and much more promising when potential drug-like molecules are detected a priori. In this work, about one hundred thousand molecules are ranked following a new methodology: a drug/non-drug classifier constructed by a consensual set of classification trees. The classification trees arise from the stochastic generation of training sets, which in turn are used to estimate probability factors of test molecules to be drug-like compounds. Molecules were represented by Topological Quantum Similarity Indices and their Graph Theoretical counterparts. The contribution of the present paper consists of presenting an effective ranking method able to improve the probability of finding drug-like substances by using these types of molecular descriptors.