ABSTRACT Introduction: Predictions of environmental adverse effects and pharmacokinetic properties of compounds are areas of great interest. Objectives: The aim of this work is to develop computational models to predict ecotoxicological risk and to predict pharmacokinetic properties. Methods: In order to develop the computational models, several databases of toxicity, mode of toxic action, permeability across human barriers and chiral properties were collected. According to the specific study, statistical techniques as well as artificial intelligence and/or machine learning techniques were employed. We used several international software as well as software developed by our groups. Results: The main results are the development of new QSAR models to predict environmental toxicity of chemicals, new models to predict the permeability of compounds as well as the computational definition of new chiral descriptors. The results have been published in internationally recognized journals, presented in conferences and have obtained several national and international awards. Conclusions: The work permits increasing the collaboration with several universities from Cuba and other countries. The new developed models are available for the international scientific community.

RESUMEN Introducción: La evaluación de los efectos adversos de los contaminantes sobre los ecosistemas y la predicción de propiedades farmacocinéticas de compuestos, son campos de investigación de gran interés. Objetivos: desarrollar modelos computacionales que permitan la predicción, lo más acertada posible, del potencial ecotoxicológico de sustancias químicas de interés; así como la predicción de propiedades farmacocinéticas. Métodos: Para el desarrollo de los modelos predictivos se recopilaron bases de datos de toxicidad, mecanismos de acción toxica, permeabilidad a través de barreras biológicas y propiedades quirales. Se emplearon técnicas de estadística, técnicas de inteligencia artificial o de machine learning según el objetivo. Asimismo, se emplearon programas reconocidos internacionalmente y programas propios. Resultados: Como resultados más importantes se tuvieron nuevos modelos computacionales que permiten predecir mejor los efectos adversos de los compuestos sobre el medio ambiente y nuevos modelos para predecir el paso a través de membranas biológicas, así como la definición de nuevos descriptores quirales. Los resultados han sido publicados en revistas de alto impacto, presentados en eventos científicos y han obtenidos varios premios tanto nacionales como internacionales. Conclusiones: El trabajo permitió profundizar la colaboración con varias universidades nacionales y extranjeras. Los nuevos modelos desarrollados están a disposición de la comunidad científica internacional.

In the recent times, the race to cope with the increasing multidrug resistance of pathogenic bacteria has lost much of its momentum and health professionals are grasping for solutions to deal with the unprecedented resistance levels. As a result, there is an urgent need for a concerted effort towards the development of new antimicrobial drugs to stay ahead in the fight against the ever adapting bacteria. In the present report, antibacterial classification functions (models) based on the topological molecular computational design-computer aided ‘‘rational’’ drug design (TOMOCOMD-CARDD) atom-based non-stochastic and stochastic bilinear indices are presented. These models were built using the linear discriminant analysis (LDA) method over a balanced chemical compounds dataset of 2230 molecular structures, with a diverse range of structural and molecular mechanism modes. The results of this study indicated that the non-stochastic and stochastic bilinear indices provided excellent classification of the chemical compounds (with accuracies of 86.31% and 84.92%, respectively, in the training set). These models were further externally validated yielding correct classification percentages of 86.55% and 87.91% for the non-stochastic and stochastic bilinear models, respectively. Additionally, the obtained models were compared with those reported in the literature and demonstrated comparable results, although the latter were built over much smaller datasets and with much higher degrees of freedom. Finally, simulated ligand-based virtual screening of 116 compounds, recently identified as potential antibacterials, was performed yielding 86.21% and 83.62% of correct classification, respectively, and thus demonstrating the utility of the obtained TOMOCOMD-CARDD models in the search of novel compounds with desirable antibacterial activity.