In November 2020, the AI group DeepMind announced it had developed a program called AlphaFold that could rapidly predict how chains of amino acids fold up into complex shapes:
Last year, DeepMind published the protein structures for 20 species — including nearly all 20,000 proteins expressed by humans — on an open database. Now it has finished the job, and released predicted structures for more than 200m proteins.
“Essentially, you can think of it as covering the entire protein universe. It includes predictive structures for plants, bacteria, animals, and many other organisms, opening up huge new opportunities for AlphaFold to have an impact on important issues, such as sustainability, food insecurity, and neglected diseases,” said Demis Hassabis, DeepMind’s founder and chief executive.
Scientists are already using some of its earlier predictions to help develop new medicines. In May, researchers led by Prof Matthew Higgins at the University of Oxford announced they had used AlphaFold’s models to help determine the structure of a key malaria parasite protein, and work out where antibodies that could block transmission of the parasite were likely to bind.
“Previously, we’d been using a technique called protein crystallography to work out what this molecule looks like, but because it’s quite dynamic and moves around, we just couldn’t get to grips with it,” Higgins said. “When we took the AlphaFold models and combined them with this experimental evidence, suddenly it all made sense. This insight will now be used to design improved vaccines which induce the most potent transmission-blocking antibodies.”