IC Seminar - August

Artificial intelligence for the discovery and design of antimicrobial peptides

Expone: Carlos Brizuela, Departamento de Computación y Centro de Investigación Científica y de Educación Superior de Ensenada, Baja California

Resumen:

The Digital Protein Design (DPD) problem can be posed as the generation of an amino acid sequence that will result in a desired tertiary structure. There are 20^(250) possible amino acid sequences of length 250, compared to the approximately 10^(12) extant proteins. This difference suggests the possible existence of a vast universe of proteins to explore and functions that are not observed in nature. The implications of solving the CDP problem are enormous and affect, among others, materials science, biology, biomedicine, and drug design. In the area of drug design, designing protein-based ligands for biological targets such as viruses, bacteria, fungi, parasites, or tumor cell, can give a significant boost to the development of new therapies with minimal side effects. In this talk, we will focus on two approaches, one based on combinatorial optimization and the other on machine learning, for the identification and design of antimicrobial peptides.