- Deep Learning Algorithm Could Enhance Genomic Sequencing
“Once you have the virus’ genetic sequence with next generation sequencing (NGS), then you can start asking more questions,” said Dirk Dittmer, PhD, professor of microbiology and immunology at the UNC School of Medicine, and senior author of the study.
“Where have we seen this exact sequence before? Did it come from a different state or country? When did this patient travel there and who else may have it?”
Next generation sequencing methods can help enhance diagnostic testing accuracy as well. Because most of the testing developed for COVID-19 looks for one portion of the gene sequence that causes the virus, if that one sequence mutates the test is no longer accurate.
While the researchers did find mutations in the virus’ genetic sequence during their study, none of the variations were located in the portion of the virus that’s targeted in common diagnostic testing.
“We are concerned about future mutations though,” Dittmer said. “It is inherent in a virus’ nature to mutate. Changes in other areas of the genetic sequence can not only disrupt testing, but hinder the effectiveness of vaccines.”
The study is the largest to focus on both rural and suburban communities. The team reconstructed the mutational landscape of cases seen at the UNC Medical Center, a tertiary clinical care center. Researchers analyzed 175 samples from confirmed COVID-19-positive patients between March 30 and May 8.
Of the samples tested, 57 percent carried the spike D614G variant noted in similar studies. This variant is associated with a higher genome copy number, and it has become more prevalent throughout the pandemic. The genetic variations found in these samples also support the hypotheses that most cases in North Carolina originated from people traveling within the US and not internationally.
Researchers will use a grant from the NC Policy Collaboratory based at UNC Chapel Hill to continue leveraging next generation sequencing to track COVID-19 through the remainder of 2020. The team’s goal is to enroll every patient at UNC Hospitals with flu or respiratory symptoms for COVID-19 diagnostic testing.
The group will sequence these samples and compile them to form a comprehensive profile of any virus that these patients carry. This information will further help researchers in combating COVID-19 and any novel coronaviruses.
The team will also continue to collaborate with multiple other laboratories at UNC-Chapel Hill to stay current on what changes, if any, should be made to testing protocols and potential vaccine development.
“Because we are only looking at one gene sequence for the virus, we have told the FDA that we will continually monitor for changes in this gene sequence so that we can be assured that our test is still reliable,” said Melissa Miller, PhD, director of UNC Medical Center Microbiology and Molecular Microbiology Laboratories and a co-author of the study. “NGS will help us do that.”
Other organizations have leveraged genomic sequencing to better understand and prepare for the spread of the virus. In May, the Nevada Genomics Center at the University of Nevada, Reno (UNR) announced that it would begin genomic sequencing on samples from northern and southern Nevada that initially tested positive for COVID-19.
“Probably the largest benefit will be to learn what strains of the virus are in northern and southern Nevada,” said Paul Hartley, PhD, director of the Nevada Genomics Center, one of several core laboratory facilities at the University.
“The coronavirus genome can acquire mutations, and with the right amount of data, one can do epidemiological studies to understand how the virus may have spread.”
By Jessica Kent