The latest version of the DeepMind AI system, AlphaFold, has been identified as a solution to this major challenge by the organizers of the biennial protein structure prediction challenge called Critical Assessment of Protein Structure Prediction (CASP).
“We’ve been dealing with this same problem – as proteins fold – for nearly 50 years,” said Dr. John Moult, Co-Founder and Chair of CASP, University of Maryland in the USA, in a statement.
“Seeing DeepMind come up with a solution for this, after personally working on this problem for so long and after so many stops and starts, wondering if we would ever be there , it ‘s a special time. ”
This breakdown demonstrates the potential impact of AI on scientific discovery and its ability to significantly accelerate progress in some of the most fundamental areas that define and shape our world, said DeepMind in a blog post Monday.
The shape of a protein is closely linked to its function, and the ability to predict this structure unlocks a greater understanding of what it does and how it works.
Many of the biggest challenges in the world, such as developing cures for diseases or finding enzymes that break down industrial waste, are fundamentally linked to proteins and the role they play. .
The ability to computer-predict the 3D structure of a protein has been a major challenge because of the number of ways in which it could theoretically fold a protein before embedding it into its final 3D structure with astronomy.
“Our methods may be particularly helpful for important classes of proteins, such as organ proteins, which are very difficult to crystallize and therefore challenging to test experimentally,” he said. UK-based AI company.
The results from the 14th CASP assessment were released on Monday.
DeepMind said they are preparing a paper on their system for submission to a peer-reviewed journal over time.