News18» News»Buzz»Artificial Intelligence Predicts How Proteins Fold from a Chain of Amino Acids in Ground-breaking Find
1-MIN READ

Artificial Intelligence Predicts How Proteins Fold from a Chain of Amino Acids in Ground-breaking Find

Representative image.

Representative image.

Synthesizing proteins with the desired structure could speed development of enzymes to make biofuels and degrade waste plastic.

Artificial intelligence (AI) has gradually become part of our lives, from Siri to Alexa and now it has also solved one of biology's grand challenges.

According to a recent report, AI has predicted how proteins fold from a chain of amino acids into 3D shapes that carry out life's tasks. Science Magazine report says that this week, organizers of a protein-folding competition announced the achievement by researchers at DeepMind, a U.K.-based AI company.

The method used by DeepMind will have far-reaching effects, including dramatically speeding the creation of new medications according to its organisers.

Janet Thornton, director emeritus of the European Bioinformatics Institute said that what the DeepMind team has managed to achieve is fantastic and will change the future of structural biology and protein research.

According to John Moult, a structural biologist at the University of Maryland, Shady Grove, and co-founder of the competition, Critical Assessment of Protein Structure Prediction (CASP), this is a 50-year-old problem and he never thought he would see this in his lifetime.

To simplify the significance: proteins are used by the body, tens of thousands of different proteins, each a string of dozens to hundreds of amino acids. The order of the amino acids dictates how all of them push and pull between them and give rise to proteins' complex 3D shapes, which, in turn, determine how they will function.

It will also help researchers devise drugs that can lodge in proteins' crevices. And being able to synthesize proteins with the desired structure could speed development of enzymes to make biofuels and degrade waste plastic.

However, John Jumper, who heads AlphaFold's development at DeepMind, believes the predictions were still too coarse. Science Magazine quotes him saying that they knew how far they were from biological relevance.

The team combined deep learning with an attention algorithm that replicates the way a person might assemble a jigsaw puzzle.

It connected pieces in groups and in this case clumps of amino acids, and then searched for ways to join the clumps in a larger whole.

The AI worked with a computer network built around 128 machine learning processors. DeepMind trained the algorithm on all 170,000 or so known protein structures.