Since 2020, the world has seen a multitude of SARS-CoV-2 variants with devastating effects.
Now, tapping on the advancements in the artificial intelligence (AI) space, Harvard Medical School (HMS) researchers have a developed a new tool that could make predictions about new viruses before they emerge.
The AI tool, EVEscape, has two elements. One, a model of evolutionary sequences that predicts changes in a virus, and two, detailed biological and structural information about the virus. These two help EVEscape predict the variants that are most likely to occur as the virus evolves, a press statement by HMS explains.
In the study, published recently in Nature, researchers show that if this tool had been deployed during the covid-19 pandemic, it could have predicted the most frequent mutations and detected variants the world should be most concerned about. Notably, the tool also made accurate predictions about other viruses, including HIV and influenza.
Currently, the researchers are using EVEscape to predict future variants of SARS-CoV-2. Every two weeks, they release a ranking of new variants to help scientists develop more effective vaccines and therapies. “We want to know if we can anticipate the variation in viruses and forecast new variants — because if we can, that’s going to be extremely important for designing vaccines and therapies,” senior author Debora Marks said in the HMS statement.
The researchers first developed EVE, short for evolutionary model of variant effect, to predict gene mutations that cause human diseases. When covid-19 hit the world and the SARS-CoV-2 virus kept changing its structure to escape vaccines, the researchers pivoted EVE to build EVEscape. “We’re taking biological information about how the immune system works and layering it on our learnings from the broader evolutionary history of the virus,” co-lead author Nicole Thadani explained in the statement.
In recent years, scientists have been experimenting with AI to reshape how they diagnose and detect diseases and health risks. For instance, in July, researchers found a new software that can significantly speed up the process of analysing bone density scans by processing more than 60,000 images in a single day, a Science Daily report said. Such AI tools and software, if used properly, could significantly benefit the healthcare sector.