Artificial intelligence opens up new ways to fight coronavirus7. February 2021
Artificial intelligence opens up new ways to fight coronavirus
Washington, Feb. 7, 2021
Combating COVID-19 mutations and developing new vaccines could be lightning fast thanks to a new AI framework developed at USC.
Background: A team of researchers from the USC Viterbi School of Engineering has developed a new method that can counter emerging mutations of the coronavirus, speed vaccine development and stop the spread of the pathogen.
The method allows analysis of potential mutations in the virus and ensures that the best possible vaccines are quickly identified. Based on this, solutions are possible that give people a major advantage over the evolving contagion. The team’s machine learning model can do vaccine design cycles that used to take months or years in seconds and minutes, the study says.
“This AI framework, applied to the specifics of this virus, can deliver vaccine candidates in seconds and move them quickly into clinical trials to achieve preventive medical therapies without compromising safety,” said Paul Bogdan, associate professor of electrical and computer engineering at USC Viterbi and corresponding author of the study. “Furthermore, this can be adapted to help us stay one step ahead of coronavirus as it mutates around the world.”
The findings appear in Nature Research’s Scientific Reports. When applied to SARS-CoV-2 – the virus that causes COVID-19 – the computer model eliminated 95% of the pathogen’s compounds.
The AI-assisted method predicted 26 potential vaccines that would work against the coronavirus. From these, the scientists identified the best 11, from which they constructed a multiepitope vaccine that targets the spike proteins the coronavirus uses to bind and enter a host cell. The vaccine targets the region of infection to disrupt the spike protein, neutralizing the virus’ ability to replicate.
In addition, engineers can construct a new multi-epitope vaccine for a new virus in less than a minute and validate its quality within an hour. In contrast, current virus control methods require growing the pathogen in the lab, inactivating it, and injecting the virus that caused a disease. The process is time-consuming and takes more than a year; meanwhile, the disease continues to spread.
The AI-powered method developed by USC will be particularly useful at this stage of the pandemic because the coronavirus is beginning to mutate in populations around the world. Some scientists are concerned that the mutations could minimize the effectiveness of Pfizer and Moderna vaccines now being distributed. Recent variants of the virus that have emerged in the United Kingdom, South Africa and Brazil appear to be spreading more easily, which scientists say will quickly lead to many more cases, deaths and hospitalizations.
Bogdan pointed out that if SARS-CoV-2 becomes uncontrollable because current vaccines have lost their effectiveness , or if new vaccines are needed to deal with other emerging viruses, then the method can be used to quickly design other preventive mechanisms. In the study, we described that the USC scientists used only one B-cell epitope and one T-cell epitope, while applying a larger data set and more numerous combinations can develop a more comprehensive and rapid vaccine design tool. According to the study, the method can make accurate predictions with more than 700,000 different proteins in the dataset.
“The proposed vaccine design framework can address the three most commonly observed mutations and be extended to deal with other potentially unknown mutations,” Bogdan said.
The raw data for the research came from a huge bioinformatics database called the Immune Epitope Database (IEDB), where scientists around the world have collected data on coronavirus and other diseases. The IEDB contains more than 600,000 known epitopes from about 3,600 different species, along with the Virus Pathogen Resource, a complementary repository of information on pathogenic viruses. The genome and spike protein sequence of SARS-CoV-2 is from the National Center for Biotechnology Information.