Dissertation defense (January 30, 2020): Francimary Procopio Garcia de Oliveira
Student: Francimary Procopio Garcia de Oliveira
Title: Data Integration in Drug Target Detection for Schistosoma mansoni
Advisor: Kele Teixeira Belloze
Committee: Kele Teixeira Belloze (president), Rafaelli de Carvalho Coutinho (CEFET/RJ), Ana Carolina Ramos Guimaraes (FIOCRUZ)
Day/Time: January 30, 2020/ 11:00h
Room: to be defined
Abstract:
Schistosomiasis caused by Schistosoma mansoni is a significant neglected disease for its occurrence in the world. However, there is only one drug recommended by the World Health Organization for its treatment. Therefore, searching for alternative drug targets in the fight against the disease is important. This work aims to identify possible new targets for S. mansoni drugs. The methodology takes an approach based on biological data integration, which is scattered across various public databases, in the use of homology and orthology concepts to identify essentiality and druggability proteins attributes. The machine learning method was also addressed to identify the essentiality attribute of S. mansoni proteins from the following essential and non-essential model organisms protein bases.
The intended homology-based methodology steps, which used essentiality and drugability characteristics, were achieved by obtaining a list of 15 S. mansoni drug target candidate proteins. The machine learning method indicated the Random Forest classifier as best performance, with 79% accuracy and 1,412 proteins reported as essential in the S. mansoni proteins prediction activity. The performed comparative analysis between two methods, based on homology and based on machine learning, presented a list of 6 proteins best rated.
