British Columbia biologists develop computer model to detect strains of COVID-19
A computer model developed by Boston College biologists could change the future of COVID-19 strain detection by helping to discover new antibodies and cellular mutations in the virus, Babak Momeni says.
“This kind of modeling can have many different applications, including finding new antibodies or optimizing antibodies,” said Momeni, a researcher on the project and an assistant professor of biology at BC.
The British Columbia research team created computer models of interactions between viruses and their receptors, hopefully allowing scientists to get ahead of the virus, Momeni said.
Marco Zaccaria, a British Columbia postdoctoral fellow and researcher on the team, said the team can now provide an estimate of Omicron mutations that could lead to even stronger variants.
“I think we are able to facilitate the identification of the highest affinity antibodies for the actual peak of the Omicron variant, as well as the most likely future changes it may undergo, in a try to stay a few steps ahead of the evolution of the coronavirus,” Zaccaria said.
Momeni said the team used models of atoms, called quantum mechanical modeling, to calculate the electronic structure between molecules and how they interact.
“One of the aspects of the model is that it looks at individual amino acids and focuses on the role of that amino acid in the interaction between molecules, which is very useful in assessing the role of mutations,” said Momeni.
According to Momeni, analyzing amino acids — the molecules that make up proteins — is also beneficial for artificially mutating cellular interactions using computer models. The binding of these molecules and their mutations help predict variants of COVID-19 and other infectious diseases, but there are limitations, Momeni explained.
“Crystal structures tell you where the actual atoms of a given molecule are in space,” Momeni said. “With COVID, we’re lucky that a lot of people are working on it and some of these crystal structures were available, but in other cases there’s work to be done to find the actual crystal structure. [for other infectious diseases].”
One of the main predictions of the biologists’ report, titled “Surveying the Mutational Landscape of the Omicron Variant of SARS-CoV-2 via Ab Initio Quantum mechanical modeling”, This is how the Omicron variant has stronger binding than previous strains of COVID-19, Momeni said.
“We can identify an additional mutation that could further increase Omicron’s binding to its human receptor,” Momeni said. “And that suggests there’s still room for an even stronger bond.”
Before BC biologists focused on strains of COVID-19, Momeni said the team worked with French researcher Luigi Genovese on simulating enzymes that break down toxins. Later, when COVID-19 first spread rapidly, the team refocused its tools to focus on the different variants emerging at the time.
“There were foundations, but we were looking at other interactions of biological significance,” Momeni said. “It seemed that the same tools could be applied to this question and [we] started to focus on that.
The research team is now working in collaboration with the Alternative Energies and Atomic Energy Commission (CEA). Comprised of a group of physicists, including Genovese, the CEA partners provide the actual simulations with supercomputers that run data in a relatively short period of time, Momeni said.
“We’re collaborating with them to walk them through this process of what would be the right simulations to run and how to analyze the data and how to present it so it’s accessible to biologists,” Momeni said.
Momeni also said that without the physicists’ focus on quantum mechanical simulations, the work would not be possible.
“Our project necessarily draws on the two distinct sets of expertise provided by an evolutionary biologist and a quantum physicist,” Zaccaria said. “The biologist must identify the most insightful questions to drive scientific inquiry, and the physicist must be able to distill the answers from the information that can be extracted from ab initio modeling, which is massive.”
Other scientists on the project include Welkin Johnson, chair of the British Columbia Department of Biology, and Michael Farzan, a professor at Scripps Research.
The next steps for the research team will be to focus on engineering antibodies and treating other viral diseases, most likely starting with HIV, Zaccaria said.
Momeni expects the team to work on validating their model in the future to better understand the infectivity of COVID-19 variants.
“So as new variants emerge, it will be ongoing work to simulate these structures and confirm that our model agrees with what we can see experimentally in terms of the role of each mutation,” Momeni said. .
Featured Image by Nicole Vagra/Editor