Artificial intelligence is, at best, a controversial topic. AI models are taking away jobs from the tech sector, and the people who remain need AI for job security.. Then there’s all the damage caused by AI data centers, but every once in a while, AI finds a truly beneficial niche. Concrete example: the design of vaccines against the coronavirus. Well, at least some breeds of COVID-19. A team of virologists from the University of Cambridge tested a new vaccine and published the results in the Journal of Infection.
The team designed the vaccine to target the Sarbeco family of coronaviruses using AI known as digitally optimized synthetic vaccine (DIOSynVax) technology. Long story short, DIOSynVax created a sort of “super antigen” that was billed as what a Chemical & Engineering News editorial called a “universal coronavirus vaccine.” Oh, and the vaccine was “formulated” in DNA, allowing it to use “needle-free intradermal delivery” technology. Add this to the list of real-life inventions inspired by science fiction (especially the hypospray from “Star Trek”).
The trial consisted of injecting 39 participants, aged 18 to 50, with different doses of the vaccine, called pEVAC-PS. Researchers studied safety, tolerability, immunogenicity (whether it triggered an immune system response), and reactogenicity (whether it produced the expected physical responses). According to the results, the 39 volunteers “tolerated” all doses without any adverse effects. Not counting those expected on immunogenicity, which of course consisted of minor episodes of COVID-19.
One test completed, there is still much to do
Margjohnsonva/Getty Images
The pEVAC-PS vaccine trial is notable because it is the first AI-designed vaccine that has progressed to human testing. However, science is not based on one-off tests but rather on repetition. This is especially true in human testing. Although researchers determined that pEVAC-PS is safe, it was only a “phase 1 study” and was subject to several limitations.
For example, all participants were from the Southampton area of the UK (recruited from the NIHR Southampton Clinical Research Facility). Additionally, it was not possible to blind the researchers or participants (no, not literally; they just knew which test groups they belonged to). More importantly, some groups within the participant pool had not been exposed to specific COVID variants before the trial, which could have unfairly impacted their immune responses.
According to the research paper, further testing is needed to (and would) address several of these issues, particularly the last one. After all, pEVAC-PS wouldn’t really be a “universal coronavirus vaccine” if it didn’t work on the omicron BA.1 or BA.2 variants. Nevertheless, the researchers are confident that pEVAC-PS could help pave the way for future vaccine technology – mainly because DNA vaccines enable rapid prototyping and are easy to manufacture, and the intradermal delivery system reduces needle waste. Between this vaccine and the wonder materials invented by AI, there could be more and more research articles evaluating the effectiveness of AI-generated products.
