DeepMind, an artificial-intelligence company owned by Google, has announced a scientific breakthrough which could have a major influence on drug discovery and disease research.
The company's AI program AlphaFold brought together experts from a variety of fields — including machine learning, structural biology, and physics — to break new ground in helping scientists understand the behavior of microscopic proteins.
Proteins are essential to the basic functionings of biology, driving how humans, viruses, and other organisms work.
A protein comprises sequences of amino acids which "fold" into shapes. Predicting which shape a protein will take — the "protein-folding problem"— is highly complicated, and researchers have been puzzled for decades. Comprehension of protein structures would give researchers a better understanding of how everything from the human body to viruses like the coronavirus work, in turn boosting research and drug discovery.
DeepMind on Monday said AlphaFold could predict a protein's shape to a high degree of accuracy, which would essentially solve the protein-folding problem far earlier than experts had predicted.
The company put AlphaFold through its paces at CASP, the Critical Assessment of protein Structure Prediction, a global competition set up to encourage groundbreaking research in the field.
"What AlphaFold has achieved is truly remarkable, and today's announcement is a win for DeepMind, but it's also a triumph for team science," Krzysztof Fidelis, a senior CASP organizer, said. "Being able to investigate the shape of proteins quickly and accurately has the potential to revolutionize life sciences."
Experts said DeepMind's model could help better target treatments to individual patients, with doctors prescribing the most appropriate medication before any nasty side effects flared up.
Demis Hassabis, the CEO of DeepMind, said at an online press briefing: "We're best known for our work on games, but it's important to realise they were always meant as a stepping stone on our path."
"It's tremendous to see the triumph of human curiosity, endeavor, and intelligence in solving this problem," added professor Janet Thornton, a leading figure at the European Bioinformatics Institute who is not affiliated with CASP or DeepMind. "A better understanding of protein structures and the ability to predict them using a computer means a better understanding of life, evolution, and, of course, human health and disease."
Join the conversation about this story »
NOW WATCH: A cleaning expert reveals her 3-step method for cleaning your entire home quickly