Empleo de herramientas computacionales de Inteligencia Artificial en la identificación de nuevos compuestos con potencial actividad contra Leishmania major
Fecha
2021
Autores
Moreno López, Frank Luis
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Central ``Marta Abreu`` de Las Villas. Facultad de Química y Farmacia. Departamento de Farmacia
Resumen
La leishmaniosis se define como un conjunto de enfermedades de presentación clínica muy variada producidas por parásitos intracelulares obligados, pertenecientes al género Leishmania. Han sido clasificadas por la Organización Mundial de la Salud en la categoría I de enfermedades infecciosas y forman parte de las patologías tropicales olvidadas o desatendidas. La Leishmania major afecta principalmente a niños menores de cinco años y ha tenido un incremento en la aparición de leishmaniosis cutánea y visceral. La búsqueda de nuevas alternativas terapéuticas sigue siendo un reto y los estudios in silico constituyen herramientas altrernativas para dar solución a esta problemática. En la presente investigación se emplean WEKA, descriptores moleculares 0D-2D del software Dragón y técnicas de Inteligencia Artificial con este fin. Se confeccionó una nueva base de datos y se emplearon cuatro modelos con las técnicas: k vecinos más cercanos, árboles de clasificación, perceptrón multicapas y máquina vectorial de soporte (IBk, J48, MLP y SVM, respectivamente), para su cribado virtual. El sistema multiclasificador ensamblado de voto, empleado en la identificación de potenciales compuestos activos contra promastigotes de Leishmania major, mostró mejores resultados que los modelos por separado, y permitió la identificación de forma óptima de 34 nuevos candidatos a través del cribado virtual.
Palabras Clave: Leishmania major, estudios in silico, técnicas de Inteligencia Artificial, QSAR
Leishmaniasis is defined as a set of diseases with a highly varied clinical presentation produced by obligate intracellular parasites, belonging to the genus Leishmania. They have been classified by the World Health Organization in category I of infectious diseases and are part of the neglected or neglected tropical pathologies. Leishmania major mainly affects children under five years of age and has had an increase in the appearance of cutaneous and visceral leishmaniasis. The search for new therapeutic alternatives continues to be a challenge and in silico studies are alternative tools to solve this problem. In the present investigation WEKA, molecular descriptors 0D-2D of the Dragon software and Artificial Intelligence techniques are used for this purpose. A new database was prepared and four models were used with the techniques: k nearest neighbors, classification trees, multilayer perceptron and support vector machine (IBk, J48, MLP and SVM, respectively), for their virtual screening. The assembled voting multiclass system, used in the identification of potential active compounds against promastigotes of Leishmania major, showed better results than the separate models, and allowed the optimal identification of 34 new candidates through virtual screening. Key words: Leishmaniasis, Leishmania major, in silico studies, Artificial Intelligence techniques, QSAR
Leishmaniasis is defined as a set of diseases with a highly varied clinical presentation produced by obligate intracellular parasites, belonging to the genus Leishmania. They have been classified by the World Health Organization in category I of infectious diseases and are part of the neglected or neglected tropical pathologies. Leishmania major mainly affects children under five years of age and has had an increase in the appearance of cutaneous and visceral leishmaniasis. The search for new therapeutic alternatives continues to be a challenge and in silico studies are alternative tools to solve this problem. In the present investigation WEKA, molecular descriptors 0D-2D of the Dragon software and Artificial Intelligence techniques are used for this purpose. A new database was prepared and four models were used with the techniques: k nearest neighbors, classification trees, multilayer perceptron and support vector machine (IBk, J48, MLP and SVM, respectively), for their virtual screening. The assembled voting multiclass system, used in the identification of potential active compounds against promastigotes of Leishmania major, showed better results than the separate models, and allowed the optimal identification of 34 new candidates through virtual screening. Key words: Leishmaniasis, Leishmania major, in silico studies, Artificial Intelligence techniques, QSAR
Descripción
Palabras clave
Estudios in Silico, Técnicas de Inteligencia Artificial, Descriptores Moleculares, Estudios QSAR