Google’s DeepMind claims to have solved one of biology’s long-standing problems

Google’s DeepMind this week reported that it solved one of biology’s long-standing problems: predicting the 3D structure of proteins.

Because matter: Being able to determine the structure of proteins could help accelerate drug development and help researchers understand the underlying biology of the disease.

The problem: Predicting how physics organizes atoms into amino acids, giving rise to the twisted and folded structure of proteins, is one of biology’s most difficult challenges.

• The structure of a protein determines if and how it binds to other proteins and molecules, biological processes that underpin life.
• Structure also plays a role in how drugs bind to proteins in the body.
• “We’ve been stuck on this problem – how proteins fold – for nearly 50 years,” said John Moult, a professor at the University of Maryland and co-founder of Critical Assessment of Structure Prediction, or CASP, in a Press release.

What is happening: At Monday’s CASP meeting, DeepMind announced that AlphaFold 2 – its second competitor in the evaluation that has occurred every two years since 1994 – can reliably and accurately predict protein structures within the width of an atom.

• CASP teams receive sequences of proteins or parts of proteins over the course of a few months and present the expected structures.
• About two-thirds of the time, AlphaFold accurately predicted protein structure on par with X-ray crystallography and cryo-electron microscopy, proven experimental techniques for determining protein structures that are expensive, time-consuming and an art form. scientific.
• AlphaFold is a deep learning network trained on approximately 170,000 protein structures. One area where the system has struggled is with protein clusters that can distort the shape of the other, Nature news reported.

Keep in mind: Determining the structure of a protein is a big step, but only one in the process of developing new drugs.

• “But DeepMind’s methods could be a way to determine if a clinical trial will fail due to toxic reactions or other problems, at least in some cases,” writes NYT’s Cade Metz.