Check out the on-demand sessions from the Low-Code/No-Code Summit to learn how to successfully innovate and achieve efficiency by upskilling and scaling citizen developers. Watch now.
A study coauthored by researchers at MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) describes an open source system that introduces methods for designing, evaluating, and augmenting both new and existing vaccine designs. The system — OptiVax — leverages machine learning to select short strings of amino acids called peptides that are predicted to provide high population coverage for a vaccine.
Fewer than 12% of all drugs entering clinical trials end up in pharmacies, and it takes at least 10 years for medicines to complete the journey from discovery to the marketplace. Clinical trials alone take six to seven years, on average, putting the cost of R&D at roughly $2.6 billion, according to the Pharmaceutical Research and Manufacturers of America.
OptiVax might hold a key to reducing costs and expediting drug discovery, courtesy of its use of multiple predictive models. By identifying peptide fragments from a set of viral or tumor proteins and scoring the peptides for selection across various criteria, including their mutation rate in about 5,000 genomes, OptiVax can design a vaccine to maximize population coverage in several different geographic regions. Administering the peptide fragments in a vaccine can lead to immunity because the fragments stem from the virus.
OptiVax also takes into consideration the vast differences in people’s individual DNA. As the researchers explain, for a peptide to induce immunity, it must first bind within the groove of a class of major histocompatibility complex (MHC) molecules — molecules containing sets of genes that code for immune system cell surface proteins. The peptide must also be immunogenic, meaning it must activate T cells (the white blood cells that search out and destroy pathogens) when it’s bound by MHC proteins and displayed in the body.
Intelligent Security Summit
Learn the critical role of AI & ML in cybersecurity and industry specific case studies on December 8. Register for your free pass today.
The researchers used a complementary system — EvalVax — to predict coverage for OptiVax-generated vaccines and other competing vaccine designs. They found that two EvalVax-proposed COVID-19 vaccines would provide greater than 90% population coverage. On the other hand, they identified several from a group of 29 third-party designs that wouldn’t likely provide high coverage.
“We evaluated a common vaccine design based on the spike protein for COVID-19 that is currently in multiple clinical trials,” CSAIL Ph.D. students and paper coauthors Ge Liu and Brandon Carter said. “Based on our analysis, we developed an augmentation to improve its population coverage by adding peptides. If this works in animal models, the design could move to human clinical trials.”
Liu, Carter, and the other coauthors say they’re working with the National Institutes of Health to determine whether their approaches could be used for risk prediction using data from COVID-19 patients. Beyond this, they hope to apply the frameworks — OptiVax and EvalVax — to design vaccines for a range of infectious diseases.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Discover our Briefings.