Related news
Google is something more than the most used web search engine in Spain; he is such a great technological and scientific giant that he had to create a parent company, Alphabet, to cover everything he did.
For example, the development of artificial intelligence has been one of the main priorities in recent years, especially applied to medicine; here he was successful in detecting cancer.
Another project that has gained a lot of strength is that of DeepMind, which became famous after defeating the best masters of Go, in addition to learning chess and video games.
Google for the secret of life
But from the beginning, Google made it clear that the future of DeepMind has gone far beyond becoming a game master; Today it was demonstrated with the announcement that its AlphaFold system has managed to solve one of the big problems facing the scientific community: protein folding.
The chemical composition of a protein does not matter; the most important part is how it folds itself, creating microstructures, because if it does not fold correctly, it will not work as it should in a biological system. This is especially vital for the development of new drugs. It does not matter if the composition is correct: if it does not fold as it should, it will not do any good.
Computer simulated protein folding
He has known this science for 50 years, but so far he has not been able to do much other than this, other than testing all combinations by “brute force”. This is how projects like Folding @ Home were born, which take advantage of the unused power of our computers and consoles to process the folding of proteins until we find the right solution. But even so, it takes too long, a few months.
Making DeepMind
What the DeepMind team has done with AlphaFold is to predict the folding of proteins with great precision; Specifically, it is able to know how to bend with a margin of error of only the diameter of an atom. This process can take days or even hours, so it is possible to get almost immediate results that can then be verified by researchers.
Protein folding from experiments (green) and AI (blue)
Omicrono
Although it is not an absolutely precise system, it is enough to let the scientists do the rest of the work. To understand the extent to which this advance can change everything, there is the example of a team of German researchers who had spent a decade trying to figure out the structure of a protein; once they received help from DeepMind, in just half an hour they managed to find the final structure.
DeepMind AI protein folding
A decade of work completed in half an hour; And this is only the beginning. In the future, it is expected that this human intervention will not be necessary and that computers will be able to obtain the necessary accuracy to predict the folding of any protein; effectively, revealing all its secrets and opening the doors to new developments.