Yes Artificial intelligence (AI) EIt is able to build words and phrases with all the letters of the alphabet, why cannot the genome write using four letters that make up the DNA sequences? This question was asked by the team of scientists from Stanford University Before conducting a study that has just been published without steam, but with very interesting results. And, as they explain, they managed to get Viral genomes And with them to make Particles of the bacteriophage viruscapable of kalmanization of bacteria populations in the experiment.
It would be interesting at any time, but, first of all, at the time when Antibiotic resistance They become a serious problem. More and more bacteria have tools to evade the only drugs that can destroy them, so they become a more powerful and almost indestructible version. It’s as terrible as it seems, so Bacteriophage virusCable to infect and kill bacteria, can be a very effective tool.
Of course, they can be isolated and cultivated in the laboratory, but this is not easy. Remember that viruses can be reproduced only if they infect guest cells. In the case bacteriophagesThese cells are bacteria. Therefore, it would initially be necessary to grow bacteria to multiply these viruses. On these scientists, on the other hand, they managed to create bacteriophages themselves using The algorithm that looks like the one that gives a chat bots such as ChatgptThe field was very positive when trying to get bacteriophyrs on bacteria Escherichia coliAlthough we must remember that the study should still go through appropriate reviews.
How can you build bacteriophage viruses through AI?
AI works on the basis of training with millions and millions of data. To be able to create phrases, you previously read a lot of phrases. He knows how to collect letters to each other so that they make sense, and then how to knit the words received.
In the case of genes, only 4 letters are required that correspond to 4 nitrogen bases: adenin (a), guanine (G), cytosine (C) and Timin
Each species, and more specifically each person, has its own combination of these 4 letters. They are, so to speak, our training book. Therefore, if AI reads many of these training books, perhaps could make others from scratchThe field is what these scientists tried with the development of the algorithm that they baptized as Evo2The field was trained in many genetic sequences of bacteriophage viruses, so he learned How to combine letters To write your instructions. Then they allow him to build new sequences from scratch. On the other hand, they had a tool that could determine which of the received sequences really correspond to the bacteriophage. Remember that sometimes the chat What does not know is the inventionThis field is quite common in AI, so you should have a thousand eyes.
And now?
As soon as the sequences were filtered to choose only those that correspond to bacteriophage viruses, these scientists have begun to make them. Nucleotide sequences They can be synthesized in the laboratory. The most difficult thing is to get a sequence. Then they turned this new genetic material into Plasmids These are round DNA structures typical of bacteria. To introduce new genes into the bacterium, this form is usually given initially, since the introduction is easier.
The sequences of viruses of bacteriophages selected AI become a Trojan horse. This is glued between the genetic material of bacteria, which begins to read the instructions, as if they were it. The result is that the bacteria themselves are built inside the virus, which will cause its destruction.
How these scientists explained in statements IflscienceViruses have already tried bacteria Escherichia coli And the results were very good. Viruses extended the bacteria.
You have to go carefully
The authors of this study have already sent it for review. That is, they made it accessible to other scientists to check whether the correct methodology, and the results make sense. This review has not yet ended, so the results should be read very careful. In any case, this is undoubtedly a very interesting example of what can do. If the vapor review is successful, and you can continue to work in this line, we could face a very powerful future tool for Bacteria resistant to antibiotics.
Source: Hiper Textual
