Identification of Composite Biomarkers for the distinct clinical forms of Chagas Disease

• Grant number: 18/20473-4
• Support Opportunities: Scholarships in Brazil - Master
• Start date: January 01, 2019
• End date: June 18, 2019
• Field of knowledge: Biological Sciences - Immunology - Applied Immunology
• Principal Investigator: Edecio Cunha Neto
• Grantee: Natalia Bueno Pereira
• Host Institution: Instituto do Coração Professor Euryclides de Jesus Zerbini (INCOR). Hospital das Clínicas da Faculdade de Medicina da USP (HCFMUSP). Secretaria da Saúde (São Paulo - Estado). São Paulo , SP, Brazil

Abstract

Chagas disease (CHD) is a highly debilitating disease and affects around 1.2 million people in Brazil. Chronic chagasic cardiopathy (CCC) is the most severe clinical disease manifestation and affects 30% of patients, while 60% remain in the indeterminate form. There are no biomarkers of classification of the clinical forms and prognostic of progression to CCC, and these are urgently needed for the clinical management of patients. Our research group has a cohort, REDSII (Retrovirus Epidemiology Donor Study II), with 600 chagasic patients from the various clinical forms of the disease. This cohort was explored, not concomitantly, by different methods for identification of biomarkers such as cardiac and inflammatory plasma, parasitism, global analysis of genetic polymorphisms (GWAS), reactivity to panel of T. cruzi antigens and transcriptomic analysis of whole blood. However, the discriminatory power of classes of these individual markers proved to be unsatisfactory. It is known that composite biomarkers have been shown to be more efficient and discriminatory than individual ones. In this context, the objective of this project is to identify composite biomarkers that allow to discriminate patients from different clinical forms of ChD by the integration of all databases of markers already tested from REDSII cohort . To do so, after the normalization of each set of experimental data, we will use a bioinformatics approach through the use of several machine learning algorithms, implemented in different predictive methods, for the determination of more efficient and universal composite biomarkers that can classify the clinical forms of ChD.