State-of-the-art techniques for predicting regions of high antibody variability from epitopes

• Grant number: 20/11194-4
• Support type: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): March 01, 2021
• Effective date (End): February 28, 2022
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Theory of Computation
• Principal researcher: João Meidanis
• Grantee: Henrique da Fonseca Simões
• Home Institution: Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

One of the main factors responsible for the evolutionary success of mammals is the flexibility of their immune system. Antigens, foreign and harmful agents, when on the surface or inside the body, are detected by the immune system through parts of their structures, the epitopes, producing an immediate response through NK cells and a slower, but more lethal, adaptive response, through B cells and T cells.The adaptive response is generated when B cell and T cell receptors bind to an epitope, generating the immune response, which then starts the production of new cells and antibodies, already adapted to the identified epitope. The adaptability of antibodies occurs in regions of hyper variability called CDRs (\ textit {Complementarity Determining Regions}). This response is memorized by the antibody-producing B cells, which allows a faster response of the adaptive system in a new contact with the same epitope.This research will analyze the main public databases of available epitopes and antibodies and evaluate the possibility of using state-of-the-art, deep learning techniques in predicting CDRs from epitopes. The construction of antibodies from epitopes is of great importance in the development of new treatments, for example, against COVID-19.