Вештачка интелигенција

EVEscape: Predicting Virus Variants with Artificial Intelligence

Summary

Scientists have developed a groundbreaking tool called EVEscape, powered by artificial intelligence (AI), that aims to predict different versions of viruses, known as variants, before they emerge. Instead of relying on conventional methods, such as laboratory testing, EVEscape utilizes AI […]

EVEscape: Predicting Virus Variants with Artificial Intelligence

Scientists have developed a groundbreaking tool called EVEscape, powered by artificial intelligence (AI), that aims to predict different versions of viruses, known as variants, before they emerge. Instead of relying on conventional methods, such as laboratory testing, EVEscape utilizes AI algorithms to assess the likelihood of a variant evading the body’s natural defense mechanisms.

This innovative tool, developed by researchers from the University of Oxford in the United Kingdom and the Harvard Medical School in Boston, Massachusetts, analyzes extensive data on virus evolution over time. It also considers detailed biological and structural information related to the virus. This cutting-edge method falls under the category of “generative artificial intelligence,” a term referring to technology designed to operate at the level of human intelligence.

The recent study, published in the journal Nature, demonstrates how the upgraded version of the previously created tool, known as EVE, now adapted for studying genetic mutations causing human diseases, can accurately predict mutations associated with conditions like cancer and cardiovascular illnesses.

During the COVID-19 pandemic, researchers seized the opportunity to transform EVE into EVEscape, a novel tool specifically designed to forecast viral variants. To achieve this, the team “rewound the clock” back to January 2020, just before the onset of the pandemic, and posed a question to EVEscape: predict the mutations that the SARS-CoV-2 virus, responsible for COVID-19, would undergo.

Remarkably, the tool accurately forecasted which mutations would emerge and become prevalent. It even predicted the potential loss of effectiveness in antibody-based therapies as the virus mutated to evade them.

Compared to traditional laboratory testing methods, EVEscape offers faster and more efficient predictions. It eliminates the need to wait for relevant antibodies to appear in the population and become available for testing. This capability allows the system to analyze tens of thousands of new virus variants that emerge each week and identify those most likely to become problematic.

Sarah Gurev, a student of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology (MIT) and the leader of the research, emphasized that “by swiftly identifying the threat level posed by new variants, we can aid early public health decision-making.”

The research team continuously utilizes EVEscape to examine real-time data on the development of new SARS-CoV-2 variants. They actively share their findings with the public and international agencies like the World Health Organization (WHO).

Furthermore, experiments conducted by the researchers demonstrated that EVEscape successfully predicts mutational behavior in other common viruses, including HIV and influenza.

Debora Marks, a professor of Systems Biology at the Blavatnik Institute of Harvard Medical School and the study’s lead author, highlighted the tool’s potential in designing vaccines and therapies, and stated that they will continue enhancing its ability to predict new variants across various viruses.

For more information, please visit [domain link].

Frequently Asked Questions (FAQ)