Improving T-Cell Epitope Prediction in Livestock

Our environment contains microbes that might prove fatal without our immune system. When a microbe breaches one of our body’s physical barriers (e.g. skin), the immune system recognizes key molecular signatures (called epitopes) in the invader and dispatches potent defence cells to the site. The first step, recognizing the threat, is complex, but deciphering the rules of immune recognition allows us to develop better diagnostics and vaccines. We propose to develop a computational tool to predict a specific subset of epitopes, called T-cell epitopes, using artificial intelligence (AI) to decipher the rules of recognition. The AI techniques used to predict epitopes typically require large datasets of actual epitopes, which are lacking for livestock species. We propose to compensate for the paucity of data in livestock by leveraging data from similar species (e.g. human) and using principles from structural biology. TO BE CONT’D

Faculty Supervisor:

Anthony Kusalik

Student:

Partner:

École nationale vétérinaire, agroalimentaire et de l'alimentation, Nantes-Atlantique

Discipline:

Computer science

Sector:

University:

University of Saskatchewan

Program:

Globalink Research Award