Modelación basada en multiclasificadores de compuestos químicos con actividad antimalárica
Archivos
Fecha
2016-07
Autores
Caballero Alfonso, Ana Yisel
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Central “Marta Abreu” de Las Villas. Facultad de Química Farmacia. Departamento de Farmacia
Resumen
La malaria o paludismo es una enfermedad causada por parásitos del género Plasmodium y se transmite al ser humano por la picadura de mosquitos infectados del género Anopheles. Esta patología es considerada como la primera causa de muerte en países tropicales, y aunque existen diversos tratamientos, estos poseen una elevada toxicidad. Los métodos computacionales basados en Relaciones Cuantitativas Estructura-Actividad han sido ampliamente empleados en el diseño de fármacos y son una alternativa racional al descubrimiento de fármacos. El objetivo del presente trabajo fue desarrollar modelos computacionales que permitan la identificación de hits antimaláricos. Con este propósito se recopiló de la literatura un conjunto de datos adecuado para la modelación y se sometieron a un estricto proceso de curación. Además se estudió el rendimiento de diversas metodologías de clasificación basadas en multiclasificadores y se seleccionó una de estas como la más apropiada para la identificación de hits antimaláricos en base a su desempeño. Durante el proceso de recopilación se obtuvieron datos de una buena calidad, ensayados por un único laboratorio y siguiendo un mismo protocolo. La curación de estos datos garantizó que estuvieran libres de cualquier error o ambigüedad que afectara la modelación. Por otra parte se encontró que los multiclasificadores que empleaban como método de selección Algoritmo Genético y como forma de agregación de salidas el Voto Mayoritario exhibían un desempeño superior.
Malaria or Paludism is a tropical disease caused by parasites of the Plasmodium genre and transmitted to humans through the bite of infected mosquitos of the Anopheles genre. This pathology is considered as the first cause of death in tropical countries and, despite several therapies exist, they have a high toxicity. Computational methods based on Quantitative Structure-Activity Relationship studies have been widely used for drug design and they stand as an alternative for drug discovery. The main goal of this research is to develop computational models for the identification of anti-malaric hit compounds. For this, a dataset suitable for the modeling of the anti-malaric activity of chemical compounds was compiled from the literature and subjected to a thorough curation process. In addition, the performance of a diverse set of ensemble-based classification methodologies was evaluated and one of these ensembles was selected based on its performance as the most suitable for the identification of anti-malaric hits. During the compilation of the dataset it was possible to obtain a high quality dataset that was curated to ensure that no noisy data would affect the modeling process. Among the explored ensemble-based methods, the one combining Genetic Algorithms for the selection of the base classifiers and Majority Vote for their aggregation showed the best performance.
Malaria or Paludism is a tropical disease caused by parasites of the Plasmodium genre and transmitted to humans through the bite of infected mosquitos of the Anopheles genre. This pathology is considered as the first cause of death in tropical countries and, despite several therapies exist, they have a high toxicity. Computational methods based on Quantitative Structure-Activity Relationship studies have been widely used for drug design and they stand as an alternative for drug discovery. The main goal of this research is to develop computational models for the identification of anti-malaric hit compounds. For this, a dataset suitable for the modeling of the anti-malaric activity of chemical compounds was compiled from the literature and subjected to a thorough curation process. In addition, the performance of a diverse set of ensemble-based classification methodologies was evaluated and one of these ensembles was selected based on its performance as the most suitable for the identification of anti-malaric hits. During the compilation of the dataset it was possible to obtain a high quality dataset that was curated to ensure that no noisy data would affect the modeling process. Among the explored ensemble-based methods, the one combining Genetic Algorithms for the selection of the base classifiers and Majority Vote for their aggregation showed the best performance.
Descripción
Palabras clave
Paludismo, Parásitos, Plasmodium, Diseño de Fármacos