Artificial Intelligence Takes Over Vaccine Development
Published July 28, 2019 | Vaccination, Future Vaccines
Story Highlights
- A computer with artificial intelligence (AI) has designed a new “turbocharged” flu vaccine without human help.
- A research team from Flinders
University in Australia created the AI program, which they named SAM
(Search Algorithm for Ligands).
- The AI vaccine research work is being funded by the U.S. National Institutes of Health.
The computer program has been dubbed “SAM” (Search Algorithm for Ligands), reflecting its mission to “search the universe for all conceivable compounds to find a good human drug (also called a ligand).”2
“It has theoretical ability to acquire knowledge and then make new ideas,” said the team of researchers from Flinders University in Adelaide, Australia that first “taught” SAM by programming in a set of compounds known to stimulate the human immune system and another set of those known not to be active. SAM had to work out for itself which qualities it was looking for to create an active drug.2
“We essentially showed all of that to the AI program called SAM and then SAM came up with its own suggestion of what might be an effective adjuvant, which we then took and tested, and sure enough, it worked,” said lead researcher Professor Nikolai Petrovsky.2 The study team used a second program, which they called the “synthetic chemist” to generate “trillions of different chemical compounds that we then fed to SAM so that it could sift through all of these to find candidates that it thought might be good human immune drugs.”3
Using the candidates identified as most appropriate by the SAM program, study team tested the compounds on human blood samples and found that, not only did the program correctly recognize immune-activating drugs, the new compounds were more powerful than those that were previously known. Shown in animal testing to boost the effectiveness of the flu vaccine, the study team is now recruiting approximately 240 patients for 12-month human trials across the U.S.4
If the new vaccine is more effective than existing flu shots (only 29 percent effective for the 2019 season—a mere nine percent for the second half of the season5 6) the developers expect it will ultimately replace the existing flu vaccine. Prof. Petrovsky said, “If this is the case then the same technology we are using for flu vaccines can be applied to improve or develop many other vaccines.”7
Why is the New Vaccine Being Tested in the U.S.?
Research dollars in Australia typically support larger institutions than Flinders University, where this AI-derived flu vaccine was developed. When the proposal by Prof. Petrovsky was turned down by Australia’s National Health and Medical Research Council (NHMRC), Petrovsky turned to the U.S. National Institutes of Health (NIH). The NIH accepted the proposal and provided more than $50 million in funding for the project.8The use of AI is expected to significantly impact the field of vaccine research and development. Currently, vaccines are developed over many years by large pharmaceutical companies, with innumerable researchers working for years to analyze the potential candidates. Using their AI technology, Prof. Petrovsky’s small team developed the “turbocharged” vaccine in about two years and expects to have it on the market in three years.9
Explaining their decision to decline funding for the AI-lead development of the new flu vaccine, a NHMRC representative said, “Funding applications for health and medical research projects are subject to rigorous, expert peer review against published criteria, to ensure transparency, probity and fairness… Therefore only applications—including applications for the development of flu vaccines—of the highest quality are funded by NHMRC.”10
References:
No comments:
Post a Comment