Science

DeepMind unlocks structure of 200 m proteins in scientific breakthrough | DeepMind

By deciphering the structure of almost every protein known to science, artificial intelligence has paved the way for the development of new drugs or technologies to combat global problems such as hunger or pollution.

Proteins are the building blocks of life. Their 3D structure, formed by chains of amino acids folded into complex shapes, largely determines their function. Once you know how a protein folds, you can begin to understand how it works and how to change its behavior. Although DNA provides the instructions for building a chain of amino acids, predicting how they interact to form a 3D shape has been more difficult, and until recently scientists only deciphered a fraction of the 200m or so proteins known to science.

AI group in November 2020 DeepMind announced that it has developed a program called AlphaFold that can rapidly estimate this information using an algorithm. Since then, it has been examining the genetic code of every organism whose genome has been sequenced and predicting the structures of the hundreds of millions of proteins that make up their collective genome.

Last year, DeepMind published protein structures for 20 species, including almost all 20,000 proteins are expressed by humans – in the open database. It has now completed the work and released the predicted structures for more than 200 m proteins.

“In fact, you can think of it as covering the entire protein universe. It includes predictive structures for plants, bacteria, animals and many other organisms and opens up huge new opportunities for AlphaFold to impact important issues such as sustainability, food security and neglected diseases,” said Demis Hassabis, founder of DeepMind. Chief executive.

Scientists are already using some of his earlier predictions to help develop new drugs. In May, researchers led by Professor Matthew Higgins at the University of Oxford announced they used AlphaFold models to determine the structure of a key malaria parasite protein and worked out where antibodies that could block transmission of the parasite might bind.

“We used to use a technique called protein crystallography to find out what this molecule looks like, but because it’s quite dynamic and moving, we just couldn’t get a handle on it,” Higgins said. “When we took the AlphaFold models and combined them with this experimental evidence, everything suddenly made sense. This insight will now be used to design improved vaccines that induce the most potent transmission-blocking antibodies.

Sign up to Issue One, our free daily newsletter – every day at 7am BST

AlphaFold models are also being used by scientists at the University of Portsmouth’s Center for Enzyme Innovation to identify enzymes from the natural world that can be modified to digest and recycle plastics. “These huge structures took us quite a long time to go through the database, but it opened up this new series of three-dimensional shapes that we’d never seen before that could actually break down plastics,” he said. work “There is a complete paradigm shift. We can really accelerate where we go from here, and that helps us focus those precious resources on the things that matter.”

Professor Dame Janet Thornton, Group Leader and Senior Research Fellow at European Molecular Biology The lab’s European Bioinformatics Institute said: “AlphaFold protein structure predictions are already used in countless ways. I expect that this latest update will create an avalanche of new and exciting discoveries in the months and years to come, all thanks to the data being open for everyone to use.”

About the author

admin

Leave a Comment