Within every cell, thousands of different proteins form the mechanisms that keep all living things alive, from humans and plants to microbacteria.Almost all illnesses, including cancer, dementia, and even infectious diseases such as: COVID-19 (new coronavirus infection) (# If there is no character limit, add parentheses when it first appears, Related to how these proteins work. Since the function of each protein is directly related to its three-dimensional shape, scientists around the world have been striving for half a century to find an accurate and quick way to discover the shape of any protein. I did.
Today (Monday), researchers at the 14th Community-Wide Experiment (CASP14) on an important assessment of protein structure prediction technology are announcing the discovery of an artificial intelligence (AI) solution to this challenge.
Created by London-based AI lab DeepMind, an AI program called AlphaFold is based on the work of hundreds of researchers around the world and has proven to be able to shape many proteins.It did so to the level of Accuracy It is comparable to what was achieved in costly and time-consuming laboratory experiments.
CASP14 is organized by Dr. John Moult (Chair) of the University of Maryland, USA. Dr. Krzysztof Fidelis, University of California, Davis, USA; Dr. Andriy Kryshtafovych, University of California, Davis, USA. Dr. Torsten Schwede of the University of Basel and SIB Swiss Institute of Bioinformatics, Switzerland. Dr. Maya Topf, Birkbeck, University of London, UK and CSSB (HPI and UKE) Hamburg, Germany.
Dr. Mort said: “Proteins are highly complex molecules, and their precise three-dimensional structure is key to many of the roles proteins play, such as insulin, which regulates sugar levels in the blood, and antibodies that help fight infections. Even the slightest rearrangement of these important molecules can have devastating effects on our health, so one of the most efficient ways to understand the disease and find new treatments is It is to study the proteins involved.
“Other species, including bacteria and viruses, have tens of thousands of human proteins and billions, but making just one form requires expensive equipment and can take years.
“Nearly 50 years ago, Christian B. Anfinsen was awarded the Nobel Prize for showing that he should be able to shape a protein based on the sequence of the protein. amino acid — Individual building blocks that make up a protein. That’s why our community of scientists is tackling the biennial CASP Challenge. “
Teams participating in the CASP Challenge will be awarded Amino acid A sequence of sets of about 100 proteins. Approximately 100 participating CASP teams from more than 20 countries try to do the same using computers while scientists study proteins in the laboratory and experimentally determine their shape. The results will be evaluated by an independent scientist.
Dr. Fideris said: “We have seen how the CASP approach has created strong collaboration between researchers working in this field of science and has accelerated the development of science.
“Since we first performed the challenge in 1994, each has seen a series of discoveries that solve this aspect of the problem, and as a result, computational models of protein structure have become increasingly useful in medical research. . “
In the latest round of the challenge, DeepMind’s AlphaFold program determined the shape of about two-thirds of the protein with accuracy comparable to laboratory experiments *. Alpha Fold was also highly accurate for most other proteins, but not at that level.
According to CASP organizers, this success is based on the achievements of both the DeepMind team and other participants in previous CASP rounds, with other teams participating in CASP 14 also having some very during this round. It states that it has created an accurate structure.
Dr. Kryshtafovych said: “What AlphaFold has achieved is truly amazing, and today’s announcement is a victory for DeepMind, but also a victory for team science. A unique and powerful way to collaborate with researchers around the world through CASP and for many years. With the contributions of many teams of scientists, we have reached this breakthrough. “
He adds: “The ability to quickly and accurately investigate the shape of a protein has the potential to revolutionize life science. Now that a single protein has almost solved the problem, a new way to determine the shape of a protein complex. It paves the way for the development of methods: a collection of proteins that work together to form many of the machines of life, and other applications. “
Professor Dame Janet Thornton, Honorary Director of the European Bioinformatics Institute (EMBL-EBI) at EMBL, which is not affiliated with CASP or DeepMind, said: From the smallest bacteria to plants, animals and humans, all living things are defined and powered by proteins that help them function at the molecular level.
“So far, this mystery remains unsolved, and determining a single protein structure often required years of experimental effort. Humans in solving this problem. It is tremendous to see the victory of curiosity, effort and intellect. The ability to better understand the structure of proteins and predict them using computers is with life, evolution and, of course, human health. It means a better understanding of the illness. “
* AlphaFold created a model of about two-thirds of the CASP14 target protein, with a global distance test score of over 90 out of 100. Above the 90-score threshold, the remaining difference between the model and the experimental structure is small and will be the size expected in the experiment. Artifacts and errors, and alternative low-energy local conformations. Note that these CASP targets are a single protein or domain, not the next frontier protein complex. The global distance test is a measure of how well the shape of a protein model matches the shape of a lab experiment. ZemlaA, Venclovas, Moult J, Fidelis K. Forecast processing and evaluation in CASP4. Proteins 2001; Suppl 5: 13-21; Zemla A. LGA: How to find 3D similarities in protein structure. Nucleic acid resolution 2003; 31 (13): 3370-3374).
Conference: 14th Community-Wide Experiment on Critical Evaluation of Techniques for Protein Structure Prediction.
Funding: The operation of CASP is partially supported by a grant from the National Institutes of Health, NIHR01GM100482.
Artificial Intelligence Solution to a 50-Year-Old Science Challenge Could “Revolutionize Medical Research” Source link Artificial Intelligence Solution to a 50-Year-Old Science Challenge Could “Revolutionize Medical Research”