A model would have managed to accurately predict the appearance of the Omicron subvariant and the Alpha variant. Technology could give public health programs advance notice of potentially dangerous coronavirus lineages.
Code name “PyR0”
A technology has exploited more than 6.4 million sequences of SARS-CoV-2, the virus that causes the coronavirus. His mission was to find patterns among mutations that help a new viral strain spread around the world. This is how, thanks to “machine learning”, PyR0 analyzed how different lineages emerged and spread between December 2019 and January 2022. “Machine learning” is a form of artificial intelligence that makes it possible to carry out predictions from the data. “We found that by modeling mutations rather than just lineages, the model was smarter and learned faster”, explains Jacob Lemieux, researcher in infectious diseases, with Scientific American. According to him, “the faster you learn the properties of a bloodline, the more you know how much to worry about”. For example, using data up to mid-December 2021, PyR0 predicted that Omicron’s BA.2 subvariant, still rare in much of the world at the time, would soon spread rapidly. . In March 2022, BA.2 effectively became the dominant strain globally. “We can’t necessarily say what’s going to happen next in terms of mutations” points to the researcher. He specifies, “we can tell what is going to happen in terms of which lineages are most likely to increase in frequency”. Concretely, PyR0 models how different combinations of mutations in different lineages of the virus affect the growth rate of individual viral variants in the population.
Perfecting the vaccine and therapies
According to the team of scientists, successful lines are driven by a small number of mutations, and the rest just follow suit. Thus, this ability of the model to quickly analyze entire genomes could help determine which areas of the virus genome to study in order to develop future therapies. You should know that most vaccines against the coronavirus target the spike protein of the virus. The latter uses it to enter the cells. However, mutations in this protein seem to allow certain variants to escape the body’s immune response against the virus. And this, even in the event of vaccination or a previous infection. The PyR0 model revealed that it was not the number of spike protein mutations that made a strain more evolutionarily fit; but rather some specific mutations of the protein. “The reason we are able to do [ces prédictions] is that people around the world sequence the virus and label the sequences with the date and region of collection,” completes the researcher. He adds, “So we know, in different regions, which lineages are increasing in frequency relative to others. This information is extremely valuable – we could not have created our model without this type of information”.
Follow Geeko on Facebook, Youtube and Instagram to not miss any news, tests and tips.