What a difference a year makes. Twelve months ago, artificial intelligence (AI) company DeepMind stunned many scientists by releasing the predicted structures of around 350,000 proteins, part of work known as scienceBreakthrough 2021. Yesterday, DeepMind and its partners went much, much further. The company discovered the possible structures of nearly all known proteins, more than 200 million from bacteria to humans, an astonishing achievement for AI and a potential treasure trove for drug development and evolutionary studies.

“We are now inferring the structures for the entire protein universe,” Demis Hassabis, DeepMind’s founder and CEO, said at a press conference in London.

The structural bounty comes from AlphaFold, one of the new AI programs that has solved the protein folding problem, the long-standing challenge of accurately deriving the 3D shapes of proteins from their amino acid sequences. The new predicted structures of AlphaFold were released yesterday in an existing database through a partnership with the European Bioinformatics Institute of the European Molecular Biology Laboratory (EMBL-EBI). The database “has given structural biologists this powerful new tool where you can look at the 3D structure of a protein almost as easily as you can do a keyword search on Google,” Hassabis said.

Eric Topol, director of the Scripps Translational Research Institute, echoed the surprise of many outside scientists. “AlphaFold is the unique and important breakthrough in life science that demonstrates the power of AI,” he tweeted. “With this new addition of structures illuminating almost the entire protein universe, we can expect more biological mysteries to be solved every day.”

The release of the DeepMind structure is “extraordinary”, Ewan Birney, EMBL’s deputy director general, said at the press conference. “It will make many researchers around the world think about what experiments they can do now.”

The proteins resolved by AlphaFold come from organisms ranging from bacteria to plants to vertebrates, including mice, zebrafish and humans. Kathryn Tunyasuvunakool, a DeepMind research scientist, said it took AlphaFold roughly 10 to 20 seconds to make each protein prediction. The company had to work closely with EMBL-EBI, she noted, to figure out how to represent the large number of structures in the database.

DeepMind says more than 500,000 researchers have already used the database since its launch last year. Hassabis envisioned a “new era in digital biology” in which drug developers could move from AI-predicted protein structures relevant to any medical condition to using AI to design small molecules that affect them protein – and therefore treat a disease.

Others are using the structure predictions to develop vaccine candidates, to investigate basic questions in biology, such as how the so-called nuclear pore complex controls which molecules enter a cell’s nucleus, or to examine evolution of proteins when life first evolved.

Hassabis, however, cautioned that loosening the structures is just a starting point. “There’s still a lot of biology and a lot of chemistry that needs to be done.”

By admin

Leave a Reply

Your email address will not be published. Required fields are marked *