[ad_1]
DeepMind announced a big hit Monday: Its AI can accurately predict how proteins are made. What exactly does this mean and what will it bring us?
DeepMind bought Google in 2014 and a year later became a wholly owned subsidiary of Alphabet. Society has created a neural network that learns to mimic human behavior. She became famous in 2016, when her AlphaGo program defeated Go Lee Sedol live pro player.
DeepMind reports on its blog that the latest version of AI AlphaFold has solved the challenge of the organizers of CASP (Critical Assessment of protein Structure Prediction). Artificial intelligence can predict the forms in which proteins are composed.
Do you know that…
are proteins essential for life and support virtually all of its functions? These are large complex molecules made up of chains of amino acids. The function of a protein is strongly influenced by its 3D structure.
LookDeepMind teaches artificial intelligence to navigate an unknown environment
Researchers have been trying for nearly 50 years to find out what form proteins are made of. Christian Anfinsen first hypothesized in 1972 that the structure of a protein should be completely determined by the amino acid sequence. The effort to prove the hypothesis has led to experiments that require time and money. The researchers wanted to predict the 3D structure of a protein based only on its 1D amino acid sequence as a complementary alternative.
The big hurdle was that there are astronomically many ways a protein could theoretically assemble itself before settling into its final 3D structure. Thanks to DeepMind, it looks like it’s now successful. AlphaFold’s results represented a significant improvement in accuracy over previous years, and some of DeepMind’s predictions were so accurate that they could compete with experimental results at the atomic level.
Proteins are the building blocks of life and are responsible for a myriad of functions inside and outside the human body.
“It’s a big problem. “ said John Moult, a computational biologist at the University of Maryland at College Park. “In a way, the problem is solved.” It was Moult who founded CASP in 1994 to improve computational methods for the accurate prediction of protein structures. According to the New York Times, in some cases, forecasts can be generated within hours!
Structural biologist Janet Thornton of the European Molecular Biology Laboratory told Nature that the leap could elucidate the functions of thousands of proteins in the human body, the function of which has not been elucidated.
LookDeepMind and AI have come to the rescue of endangered animals
In any case, this is an important step forward that shows the importance and influence of artificial intelligence on scientific progress. The potential for accelerating development in the areas that explain and shape our world is evident. Companies and countries that invest in these technologies are likely to have an unbeatable advantage.
sourceDeepMind, Nature
Source link
