The method speeds up the analysis of vaccines by dramatically cutting down vaccine design cycles.

Researchers at the University of Southern California (USC) have developed a method to accelerate vaccine development to counter potential new mutations of the coronavirus. Using AI, the researchers can vaccine analysis and identify the best medical therapy for treatment. (USC)

AI-assisted computer model from USC

New variants of the deadly coronavirus have appeared in the United Kingdom, Brazil, and South Africa. Questions arise whether the vaccines developed by companies such as Pfizer and Moderna would be equally effective against the new variants.

The new avatars of the virus spread faster and could quickly lead to more infections, hospitalizations, and even deaths.

Moreover, what of other, more deadly mutations that could potentially surface in the future?

What we need is a system to design vaccines to rapidly counter these evolving threats. The new machine learning model from the USC scientists will accomplish vaccine design cycles that previously took months or years, in the space of minutes or seconds.

Researchers tested the model on SARS-CoV-2 — the virus that causes COVID-19. It quickly eliminated 95% of the compounds that were candidates to treat the virus.

Instead, it pinpointed the best, most viable options by predicting 26 potential vaccines that could prove effective against the pathogen. Scientists further refined this selection. They homed in on the best 11 that could target and disrupt the deadly spike proteins on the virus. These are instrumental in its rapid spread.

Going one up on the virus

“This AI framework, applied to the specifics of this virus, can provide vaccine candidates within seconds and move them to clinical trials quickly to achieve preventive medical therapies without compromising safety,” said Paul Bogdan, associate professor of electrical and computer engineering at USC Viterbi and corresponding author of the study. “Moreover, this can be adapted to help us stay ahead of the coronavirus as it mutates around the world.”

Encouragingly, Bogdan said the method will prove useful if current vaccines prove inadequate or if scientists need to develop new vaccines to tackle the virus.

“The proposed vaccine design framework can tackle the three most frequently observed mutations and be extended to deal with other potentially unknown mutations,” he said.

Related Story:  AI Found Bariticinib As A Potential Treatment For COVID-19