DNA2.0 has developed ProteinGPS, a technology for functional navigation in protein space. The technology takes advantage of efficient gene synthesis combined with advancements in linear and nonlinear systems optimization. Protein engineering has classically been approached from two diametrically opposed directions: rational design and directed evolution. ProteinGPS instead uses established machine learning to provide a standard convention for protein space navigation. The method calculates the specific location of a protein variant in multidimensional space and places unique information rich variants at important crossroads within the space assessed. The resulting datasets are synthesized and used to map the functional protein hyper space and calculate new protein variant sequences that fulfill the functional constraints needed. The feasibility of manufacturing tens or hundreds of protein variants in which all amino acid changes are precisely defined allows the sequences and activities of these variants to be analyzed using machine learning techniques adapted from optimization tasks found in other engineering disciplines. Several recent examples of ProteinGPS based protein engineering are to be presented.

Extended Abstract Status: Not Uploaded