[ad_1]
One of biology’s biggest mysteries has been solved using artificial intelligence, experts announced.
Predicting how a protein folds into a unique three-dimensional shape has puzzled scientists for half a century.
London-based artificial intelligence laboratory DeepMind has largely solved the problem, organizers of a scientific challenge say.
Their program determined the shape of the proteins to a level of accuracy comparable to expensive and time-consuming laboratory methods, they say.
The discovery is expected to accelerate research into a number of diseases, including Covid-19.
Dr Andriy Kryshtafovych, of the University of California (UC), Davis in the United States, one of the panel of scientific judges, described the result as “truly remarkable”.
“Being able to study the shape of proteins quickly and accurately has the potential to revolutionize the life sciences,” he said.
DeepMind’s Demis Hassabis on The Life Scientific (BBC Radio 4)
What are proteins?
Proteins are present in all living things where they play a central role in the chemical processes essential for life.
Made up of strings of amino acids, they fold in an infinite number of ways into elaborate shapes that hold the key to how they perform their vital functions.
Many diseases are linked to the role of proteins in catalyzing chemical reactions (enzymes), fighting diseases (antibodies) or acting as chemical messengers (hormones such as insulin).
“Even small rearrangements of these vital molecules can have catastrophic effects on our health, so one of the most effective ways to understand the disease and find new treatments is to study the proteins involved,” said Dr. John Moult of the University of Maryland, United States, the president of the college of scientific judges.
“There are tens of thousands of human proteins and many billions in other species, including bacteria and viruses, but processing the shape of just one requires expensive equipment and can take years.”
How does the challenge work?
In 1972, Christian Anfinsen was awarded the Nobel Prize for his work showing that it should be possible to determine the shape of proteins based on the sequence of their amino acid blocks.
Every two years, dozens of teams from more than 20 countries blindly attempt to predict the shape of a set of about 100 proteins from their amino acid sequences using computers.
At the same time, 3-D structures are processed in the laboratory by biologists using traditional techniques such as X-ray crystallography and NMR spectroscopy, which determine the position of each atom relative to each other in the protein molecule.
A team of scientists from Casp (the community-wide experiment on critical evaluation of protein structure prediction techniques) then compares these predictions with 3-D structures resolved using experimental methods.
Casp uses a metric known as the global distance test to evaluate accuracy, which ranges from 0 to 100. A score of about 90, obtained by DeepMind’s AlphaFold program, is considered comparable to laboratory techniques.
What happened this year?
In the final round of the challenge, Casp-14, AlphaFold determined the shape of about two-thirds of the proteins with precision comparable to laboratory experiments.
The evaluators stated that accuracy with most of the other proteins was also high, although not quite at that level.
AlphaFold is based on a concept called deep learning. In this process, the structure of a folded protein is represented as a spatial graph.
The program then “learns” using information on known 3-D forms of proteins contained in the public protein database.
The AI program was able to do in days what could take years on the lab bench.
How will this information be used?
Knowing the 3-D structure of a protein is important in drug design and understanding human diseases, including cancer, dementia, and infectious diseases.
One example is Covid-19, where scientists studied how the spike protein on the surface of the Sars-CoV-2 virus interacts with receptors in human cells.
Professor Andrew Martin of University College London (UCL), a former Casp participant and evaluator, told BBC News: ‘Understanding how a protein sequence folds into three dimensions is really one of the fundamental questions of biology.
“The whole way a protein works depends on its three-dimensional structure and its function is relevant to everything related to health and disease.
“By knowing the three-dimensional structures of proteins, we can help design drugs and deal with health problems, be they infections or hereditary diseases.”
Professor Janet Thornton of the EMBL’s European Bioinformatics Institute said that how proteins fold to create “exquisitely unique three-dimensional structures” is one of biology’s greatest mysteries.
“A better understanding of protein structures and the ability to predict them using a computer means a better understanding of life, evolution and, of course, human health and disease,” he explained.
What happens next?
Other scientists will want to examine the data to determine how accurate the AI method is and how well it performs at a very detailed level.
There is still a knowledge gap, including understanding how multiple proteins fit together and how proteins interact with other molecules, such as DNA and RNA.
“Now that the problem has been largely solved for individual proteins, the way is open for the development of new methods for determining the shape of protein complexes – collections of proteins that work together to form much of the mechanism of life and for other applications, “said Dr. Kryshtafovich.
Follow Helen Twitter.
[ad_2]
Source link
