DeepMind is releasing a free comprehensive database of its predictions of the structure of nearly every protein known to science, the company, a subsidiary of Google’s parent company, Alphabet, announced today.
DeepMind transformed science in 2020 with its AlphaFold AI software, which produces highly accurate predictions of the structures of proteins — information that can help scientists understand how they work, help treat disease and develop drugs. It first began making AlphaFold’s predictions public last summer through a database built in collaboration with the European Molecular Biology Laboratory (EMBL). That first set comprised 98 percent of all human proteins.
Now the database expands to more than 200 million structures, “covering nearly every organism on Earth whose genome sequence has been established,” DeepMind said in a statement.
“You can think of it as the whole protein universe,” DeepMind CEO Demis Hassabis said at a news conference. “We are now at the dawn of a new era in digital biology.”
AlphaFold protein structures are already widely used by research teams around the world. They are mentioned in research into things like a malaria vaccine candidate and honeybee health. “We believe AlphaFold is the most significant contribution AI has made to advancing scientific knowledge to date,” Pushmeet Kohli, chief of AI for science at DeepMind, said in a statement.
Alphabet builds on the success of AlphaFold. Alphabet launched a company called Isomorphic Labs that will focus AI tools on drug discovery, and although separate from DeepMind, the two companies will work together. DeepMind has also set up a lab at The Francis Crick Institute, where researchers can conduct experiments to test the information from the AI system.
Easy access to predicted protein structures gives scientists a boost in research efforts across the scientific landscape — such as those trying to understand how complex processes work in the body or which molecules can be used to tackle things like pollution. “With this new addition of structures illuminating nearly the entire protein universe, we can expect more biological mysteries to be solved every day,” Eric Topol, founder and director of the Scripps Research Translational Institute, said in a statement.