Imagine if you only had to get one flu shot in your lifetime instead of a flu shot every year. Imagine if there was a universal vaccine that could help the immune system recognize specific antigens on cancer cells and destroy them.
Belinda Akpa, an associate professor in the Department of Chemical and Biomolecular Engineering, is involved in a multi-million-dollar collaborative initiative that is trying to make that a reality by using predictive modeling and machine learning tools to design proteins that can act as broadly effective antigens.
The Antigens Predicted for Broad Viral Efficacy through Computational Experimentation (APECx) is a program funded by the Advanced Research Projects Agency for Health (ARPA-H), an agency within the U.S. Department of Health and Human Services.
By integrating predictive modeling and machine learning tools with some of the traditional biological and clinical ways of developing therapeutics, the APECx platform aims to uncover the protein structures necessary to stimulate robust immune responses and design more broadly effective, safe, and accessible vaccines against bacterial pathogens, cancer, and other health threats.
Akpa is part of a team being led by Texas A&M University (TAMU) that is focused on data and tool integration across the program. Through a $1.1 million ARPA-H award as the principal investigator for the University of Tennessee, Knoxville, Akpa’s role is to develop systems models of what a therapeutic protein will do to a person’s immune system.
Her efforts are part of a cross-cutting computer pipeline that will integrate tools and data from four other APECx project teams that are developing antigens and delivery platforms for specific viral or bacterial families. The pipeline developed by the TAMU-led SPHERICAL team (Scientific Platform for High Efficacy antigen design via Robust Integration of Computational experiments, AI, and protein modeLing) will enable virtual design and screening of antigens by predicting which protein will provide an effective immune response to multiple forms of the same virus.
“Computing offers us the possibility of integrating many different questions in parallel. Ultimately, we want to know: If a mutation happens, what kind of antigen, or combination of antigens, is liable to have the immune system ready to meet that challenge,” Akpa said. “Using different kinds of models—some shaped by data, and others capturing expert knowledge of human physiology—we can take the things that we’ve learned in the real-world space and lab space and begin to project what the implications would be for the future evolution of the virus and responsiveness of the immune system. This helps us plan for the future and make better design decisions.”
Spreading the Positive Impacts
The impetus for launching APECx was to speed up the methods for vaccine development and help bring down the cost. Current methods are targeted narrowly to just one form of one virus at a time, and tools to effectively use computer models are not yet widely available for vaccine design.
“A central theme of this effort from the Texas A&M team is what we call uncertainty quantification,” Akpa said. “We know our models aren’t perfect, but if we can put bounds around the confidence we have in their predictions, we can now make risk-informed decisions that make sense and focus future data acquisition on things that will most efficiently reduce our uncertainty and improve performance.”
Along with potentially speeding up the process, the use of computer models could help uncover new discoveries and treatments.
“Some viruses persist in the body for a long time without causing obvious disease, but their effects can show up later on in life, causing things like cancers or shingles or something else. They lie dormant and they can re-emerge. So how do we stop that re-emergence?” Akpa said. “There are all these different things that you could potentially do if you understood enough about the immune system and could develop the right therapeutic to tell your body, ‘Be on the look-out. We don’t want this.’”
Akpa was recently discussing the project with a practitioner of Eastern medicine, which is traditionally a more holistic approach to addressing health problems by using practices like acupuncture, diet, and movement. The practitioner told Akpa she tries to minimize the number of vaccines she receives.
“She got excited at the possibility of having just one vaccine and not having to do one every year,” Akpa said. “It opens the possibilities for making these therapeutic interventions fit into people’s lifestyles and worldviews in a way that’s likely to increase adoption and spread their impact even further than might otherwise be the case.”
Contact
Rhiannon Potkey (rpotkey@utk.edu)