AI system solves 50-year-old protein folding problem in hours

According to, an artificial intelligence company that is famed for designing computer systems to outsmart humans at gaming has now made a huge advancement in biological science.

The company, DeepMind, is owned by Alphabet Inc., the same company that owns Google. DeepMind has created a system to predict how proteins fold into their 3D shapes, a complex problem that has been plaguing researchers for decades.

Figuring out the structure of a protein can take decades of experimentation, and current protein folding computer simulations come up short on accuracy. But, a report by the New York Times shows that DeepMind’s system, known as AlphaFold, required only a few hours to accurately predict the structure of a protein.

Proteins are large molecules that are essential for life. They are made up of a string of chemical compounds called amino acids. These ‘strings’ fold in very intricate ways to create unique 3D structures that determine what the protein can do.

Nearly fifty years ago, scientists hypothesized that you could predict the structure of the protein solely by knowing its sequence of amino acids. Twenty-five years ago, scientists created an international competition to compare various methods of predicting protein structure known as CASP, Critical Assessment of Protein Structure Prediction.

For the competition, teams are given the amino acid sequence of around one hundred proteins, the structures of which are known but have not been published. The predictions are scored on a scale of zero to one hundred, with ninety considered to be on par with the accuracy of experimental methods.

AlphaFold had a median score of 92.5, up from a score of less than 60 in its first CASP competition in 2018. Despite this, the system isn’t perfect. According to Nature News, AlphaFold did not perform well in modeling groups of proteins that interact with one another.

“I think it’s fair to say this will be very disruptive to the protein-structure-prediction field. I suspect many will leave the field as the core problem has arguably been solved,” Mohammed AlQuraishi, a computational biologist at Columbia University told Nature News. “It’s a breakthrough of the first order, certainly one of the most significant scientific results of my lifetime.” 




Leave a Reply