Pioneering COVID-19 Research Came (In Half) From A 25-Yr-Outdated In Traverse City
Last November, a team of 28 researchers from the United States and the United Kingdom won a special award of $ 10,000 from the Association for Computing Machinery in recognition of groundbreaking AI simulations of the COVID-19 virus. Their research, shared thousands of times throughout the scientific community, could have played a key role in helping scientists develop vaccines effectively. One member of the 28-person team is an alum from Northwestern Michigan College (NMC) who remotely contributed to the project from here in Traverse City.
Meet Alex Brace, a 25-year-old University of Chicago graduate student who cut his teeth in the MINT department at NMC. Brace has been a Traverse City resident since sixth grade and graduated from Traverse City West Senior High School before earning an associate certificate in engineering from NMC in 2017 and ultimately a bachelor’s degree in computer science from the University of Michigan. He credits teachers at West and professors at NMC for taking him from a student who “didn’t really want to study” to someone deeply inspired by STEM.
Last year, as the pandemic hit the US and the world, Brace had the opportunity to test his “technical credibility” in fighting the virus. The summer before joining U of M, Brace had completed a research internship at Oak Ridge National Laboratory in Tennessee. Oak Ridge was founded in the 1940s as part of the Manhattan Project and is home to Summit, a system that was the fastest supercomputer in the world until last year. This internship introduced Brace to his mentor, Arvind Ramanathan, a computational biologist who, after moving from Oak Ridge to Argonne National Laboratory in the Chicago area, would bring Brace on board to play a role in the COVID-19 simulation project.
The research – published under an article entitled “AI-controlled multi-scale simulations illuminate mechanisms of SARS-CoV-2 spike dynamics” – dealt with simulating and tracking the interaction of the coronavirus with human cells. The team created a computer simulation of a SARS-CoV-2 virus envelope containing 305 million atoms, which was rendered in a realistic biological manner with a resolution that even the most powerful microscope in the world would not be able to achieve. By running this virtual model of the virus through a series of massive computer simulations, the researchers were able to get a clear picture of what the virus looks like and how it behaves. To support the project, Brace brought in his skills in computer programming, artificial intelligence, data analysis and supercomputing.
For the purposes of this project, according to Brace, the team looked at a particularly important aspect of the coronavirus.
“You saw the coronavirus picture, which looks like some kind of circle with all these little prongs,” explains Brace. “These prongs are called spike proteins. That is the atomic structure [of COVID-19] that binds to human cells. This project has essentially characterized how this bond occurs. “
Spike proteins act as the primary machinery of infection for COVID-19, enabling the virus to bind to a receptor on human cell membranes. Once that happens, the virus injects genetic material into the cell and uses the host cell to make copies of itself. According to Brace, through months of research, the team was able to highlight key occurrences of the spike protein, which allowed the SARS-CoV-2 virus to attach to and infect human cells. These revelations came thanks to countless data-intensive simulations by a number of the world’s fastest supercomputers (including Summit) that used the laws of physics to replicate the way viruses move in search of human cells.
“We really got into the question of what that looks like on the physical, atomic level.” Brace talks about the spike protein and how it enables COVID-19 to wage war on host cells. “And then, given that understanding, you can try designing some molecules to prevent this binding [between the spike protein and the human cell]. Many researchers and industry use our data to inform their vaccine development and drug design. “
By November, the research had been shared more than 8,000 times by academic and industrial groups alike. It has also been published, cited, or mentioned in several peer-reviewed publications, scientific journals, and major media outlets such as the New York Times, Scientific American, Forbes, and Wired.
The award that the project received, the ACM Gordon Bell Special Prize for COVID-19 research on the basis of high-performance computers, will only be awarded in 2020 and 2021 and is intended to provide “outstanding research achievements for understanding the COVID-19 pandemic “Through the use of high-performance computers. “The normal Gordon Bell Prize, which is awarded annually to“ recognize excellence in high-performance computing, ”is widely regarded as the Nobel Prize in the supercomputer world.
Brace, who made such a career-defining achievement at the beginning of his career, is already looking forward to the next steps. Now he works as a research fellow at the Argonne Institute – albeit remotely from Traverse City – and works with the same team to research and simulate potential drug molecules that could be used to treat COVID-19. He was also recently inducted into the University of Chicago’s computer science doctoral program. His interests, which range from automation to sustainable energy, could lead him on various career paths in the STEM, engineering and computer science fields. For now, though, he says he’s mostly grateful to be with inspiring mentors – a pattern in his life that began in northern Michigan.
“I think it all comes from my teachers and my professors [in Traverse City]”He tells The Ticker.