AI helped create a vaccine against all strains of coronavirus at once
(ORDO NEWS) — American researchers have developed a promising universal vaccine against any strain of the SARS-CoV-2 coronavirus.
In an experiment on genetically modified mice, the drug protected experimental animals from severe illness and death.
The vast majority of clinically tested and experimental Covid-19 vaccines use the S-protein as a “target”.
It is located on the “thorn” of the SARS-CoV-2 coronavirus, and the body of patients usually produces the most antibodies to it.
The trouble is that their effectiveness becomes less and less low over time: this component of the virus changes greatly with each new strain.
And it doesn’t matter if a person’s immunity is natural or formed as a result of vaccination.
In addition, the production of antibodies decreases by itself some time after vaccination. So there is no way to do without booster injections.
When antibodies do not cope – there are few of them or the antigen has changed – the virus enters the cells and does its dirty work.
But the immune system has another weapon in store for this case, so to speak, the second stage: cellular immunity.
With the help of T-lymphocytes, the body detects cells in which the virus “parasites” and destroys them. To do this, T-lymphocytes must “know” which proteins secreted by infected cells to respond to.
They are called HLA (human leukocyte antigens) and play the role of a complex traffic light that shows the immune system which cells of the body are “ours” and which are no longer very good.
Using this mechanism, it is also possible to create a vaccine, but it is much more difficult than just getting the body to produce specific antibodies.
And after such a vaccination, a person will still get sick. It just won’t be contagious, and the symptoms will be much milder. But immunity should be maintained for a longer period, and it can be formed more universal.
Because, unlike antibodies, it is possible to make the target of such a vaccination those proteins that are hidden under the virion shell, or binding to them is difficult for other reasons (for example, they seem to be outside, but covered with “thorns”). These components of the virus are often much less variable.
The problem is to choose the right HLA combinations, corresponding to the same proteins for different strains, and such that they occur in as many diseased people as possible.
A task with many parameters that requires large computing power and the analysis of colossal amounts of structured data. Exactly one in which artificial intelligence is strong.
Researchers from the Massachusetts Institute of Technology, the University of Texas, Boston University, and a number of American and Canadian scientific institutions have implemented the above approach.
They created a promising “panvariant” MIT-T-COVID vaccine that contains the genetic material of the most conserved epitopes (antigen fragments) of SARS-CoV-2.
The selection of suitable proteins was performed using a machine learning algorithm. In an experiment on genetically modified mice, the drug caused a strong immune response that protected the animals from severe illness and death.
On the seventh day after vaccination, in those coronavirus-infected mice that received MIT-T-COVID, almost a quarter (on average 24 percent) of all cells in the lungs were T-lymphocytes.
For comparison, in the control group, which received the Pfizer-BioNTech Comirnaty vaccine, the proportion of immune cells in the lungs did not exceed 6.5 percent (median 4.35 percent).
It is worth noting that on the second day after the injection, the picture was different: both vaccines provided approximately 12 percent of T-lymphocytes from the total number of cells in the lungs of experimental rodents.
That is, the drug is still only a “draft” and a demonstration of the concept, it is still very far from being used in medical practice.
This experiment showed that a universal vaccine is possible, works and is effective in preventing the most dangerous outcomes of coronavirus infection.
Detailed results of the work of American scientists are published in the peer-reviewed journal Frontiers in Immunology.
In the future, the researchers plan to optimize the drug, as well as test its effectiveness in combination with vaccines that provoke the production of antibodies.
An important area of work is checking the safety of universal vaccination, eliminating the possibility of an overly powerful immune response.
Contact us: [email protected]