Google DeepMind announced today that it has released a free database that makes predictions about the structure of almost every protein known to science.
DeepMind took the scientific community by storm in 2020 with its AlphaFold AI software, which makes highly accurate predictions about the structure of proteins, information that can help scientists understand how they work and thus help treat diseases and develop drugs.
Last summer, DeepMind first publicly released AlphaFold’s predictions through a database built in collaboration with the European Molecular Biology Laboratory (EMBL), and the initial set of data included 98 percent of all human proteins.
Today, DeepMind announced that the database is being expanded to more than 200 million structures, covering nearly every organism on the planet whose genome has been sequenced.
DeepMind CEO Demis Hassabis said in a press release, “You can think of it as covering the entire protein landscape, and we are now at the beginning of a new era of digital biology.”
IT House has learned that protein structures from AlphaFold are already being used extensively by research teams around the world, contributing to studies on malaria vaccine candidates and bee health, among others.
Pushmeet Kohli, head of scientific AI at DeepMind, said in a statement, “We believe AlphaFold is the most important contribution that artificial intelligence has made to advance scientific knowledge to date.”