Recent breakthroughs in artificial intelligence have led scientists to develop revolutionary new proteins. These advances are led by a team of scientists led by David Baker, a biochemist at the University of Washington (UW), Seattle, who reported developing molecules in seconds rather than months. This is expected to lead to many new vaccines, resistant biomaterials and treatments.
The first-ever medical vaccine, COVID-19, made from a new human-developed protein, has been approved by South Korean regulators, which is based on a spherical protein called a nanoparticle. A paper titled Robust Protein Sequence Development Based on Deep Learning Using ProteinMPN, published by biologists at Washington University of Medicine, explains how machine learning can be used to create protein molecules more precisely and quickly. The Baker lab has spent more than three decades creating new proteins using software called Rosetta. The process was divided into steps to produce the final protein. First, the researchers came up with a form for the new protein, mostly by combining pieces of other proteins. Then the program derives the sequence of amino acids corresponding to that shape. “Proteins are fundamental to all biology, but we know that all the proteins found in any plant, animal or microbe are much less than one percent of what is possible. With these new software tools, researchers should be able to find solutions to long-standing problems in medicine, energy and technology,” said David Baker, senior author and professor of biochemistry, University of Washington School of Medicine.
The team developed the protein with an approach called hallucination, where the researchers fed random amino acid sequences into a structure prediction network.
AlphaFold and a similar tool called RoseTAFold were trained to predict the structure of individual protein chains, leading to the discovery of using such networks to model the assembly of multiple interacting proteins.
