Deepmind’s AI Predicts Shape of Nearly Every Known Protein

Highlights:

• More than 500,000 researchers and biologists have used the AlphaFold database to view over two million structures.
• In the upcoming years, DeepMind also plans to collaborate with groups from the World Health Organization and the Drugs for Neglected Diseases Initiative.

DeepMind, a Google subsidiary, recently revealed that it had successfully used Artificial Intelligence (AI) to predict the three-dimensional structures of almost all catalogued proteins known to science. There are about 200 million proteins in plants, bacteria, animals, and people combined.

“Essentially, you can think of it as covering the entire protein universe,” Demis Hassabis, founder and CEO of DeepMind, told reporters.

It’s possible because of AlphaFold, a ground-breaking AI system from DeepMind that includes an open-source database that allows researchers worldwide to use it for free and at their discretion. The scientific scene had changed significantly since AlphaFold’s official launch in July 2021, when only about 350,000 3D proteins were identified.

“More than 500,000 researchers and biologists have used the database to view over 2 million structures,” Hassabis said. “And these predictive structures have helped scientists make brilliant new discoveries.”

For instance, in April 2022, Yale University researchers called on AlphaFold’s database to help them achieve their objective of creating a brand-new, incredibly powerful malaria vaccine. And in July 2021, researchers at the University of Portsmouth employed the method to create enzymes that tackle pollution caused by single-use plastics.

These endeavors are just a small sample of AlphaFold’s ultimate reach. “In the past year alone, there have been over a thousand scientific articles on a broad range of research topics which use AlphaFold structures; I have never seen anything like it,” Sameer Velankar, DeepMind collaborator and team leader at the European Molecular Biology Laboratory’s Protein Data Bank, said in a press release.

Hassabis claims that others who have used the database include those attempting to further human understanding of Parkinson’s disease, those hoping to safeguard the well-being of honeybees, and even those hoping to obtain insightful knowledge into human evolution.

“AlphaFold is already changing the way we think about the survival of molecules in the fossil record, and I can see it will soon become a fundamental tool for researchers working not only in evolutionary biology but also in archaeology and other palaeo-sciences,” Beatrice Demarchi, an associate professor at the University of Turin, who recently used the system in a study on an ancient egg controversy, said in a press release.

In the upcoming years, DeepMind also plans to collaborate with groups from the World Health Organization and the Drugs for Neglected Diseases Initiative to identify treatments for poorly understood but widespread tropical diseases like Chagas disease and leishmaniasis.

“It will make many researchers around the world think about what experiments they could do,” Ewan Birney, DeepMind collaborator and deputy director of the EMBL, told reporters. “And think about what is going on in the organisms and the systems that they study.”