[ad_1]
Back in December 2020, DeepMind surprised the world of biology right there has solved a major 50-year problem with AlphaFold, An AI tool that predicts protein synthesis. Last week a company from London published it More about this tool and released his starting number.
Now the company has announced that it has used its AI to predict the shape of almost all proteins in the human body, as well as the composition of thousands of other proteins found in 20 highly sought-after items, including yeast, fruit flies, and mice. These breakthroughs could allow biologists around the world to better understand the disease and develop new therapies.
Currently the recipe contains 350,000 proteins that have just been predicted. DeepMind says it will unveil and release more than 100 million building blocks in the next few months – mostly or all of science-recognized proteins.
“Protein digestion is a problem I’ve been monitoring for over 20 years,” says Demis Hassabis co-founder of DeepMind. “It has been a big job for us. I would say this is the biggest thing we have ever done so far. And it’s interesting in a way, because it has to play a big part in the world outside of AI. “
Proteins are made up of long amino acid ribbons, which twist themselves into complex nodes. Knowing how the protein structure is made can reveal what those proteins do, which is very important for you to understand how the disease works and develop new drugs – or to identify organisms that can help combat pollution and climate change.
The database should make life easier for biologists. AlphaFold may be available for researchers to use, but not everyone will want to run the program. David Baker, of the Institute for Protein Design at the University of Washington, who oversees the self-made lab, said, “it’s easier to get the system out of the barn than it does on your computer.” predicting a type of protein, called Alirezatalischi as well as the AlphaFold method.
Over the past few months the Baker team has been working with biologists who were previously embroiled in an attempt to determine the nature of the proteins they are studying. “There is a lot of research on living things that has been expedited,” he said. Homes with hundreds of thousands of protein ready-made containers should be larger.
“It looks amazing,” says Tom Ellis, a chemist at Imperial College London who is studying the yeast genome, who is excited to try the preservation. But he warns that the predicted form has not yet been confirmed in the lab.
In the new version of AlphaFold, prediction comes with the confidence that the tool uses to express how it feels that each format is closer to reality. Using this approach, DeepMind found that AlphaFold predicted the formation of 36% human proteins accurately up to a single atom. This is sufficient for the growth of the drug, says Hassabis.
In the past, after working for many years, only 17% of the proteins in the human body were identified by the lab. If AlphaFold’s predictions are correct as DeepMind claims, the tool will double this number in just a few weeks.
Even accurate predictions on the atomic level still work. More than half of the proteins in the human body, AlphaFold have predicted the form that should be sufficient for researchers to understand how proteins work. All other AlphaFold prophecies are either erroneous, or only one-third of the protein in the human body has no structure until it is joined to others. “They’re floppy,” says Hassabis.
“The fact that it can be used at such an important level is astounding,” says Mohammed AlQuraish, a biochemist at the University of New York who developed his own protein-specific program. He also said that having the protein of many proteins made it possible for them to learn how these proteins work as a system, not just to isolate themselves. “That’s what I think is the most fun,” he says.
DeepMind is releasing its tools with free prediction and can’t say if it has any plans for making money in the future. It does not challenge the possibility, however. To establish and manage archives, DeepMind collaborates with the European Molecular Biology Laboratory, a global research organization with a large database of protein.
For now, AlQuraishi cannot wait to see what researchers are doing with these innovations. “It looks very good,” he said. “I don’t think any of us think we’re going to be here anytime soon.
[ad_2]
Source link
