In Nature Communications, authors describe a novel web platform for genomic surveillance of the SARS-CoV-2 virus called CoVerage, which could identify variants of concern (VOCs) up to 3 months before the World Health Organization (WHO) is able to classify the variants that can lead to surges of COVID-19 activity.
Within months of the start of the COVID-19 pandemic, new variants of the original wild-type virus emerged, and these VOCs have influenced virus dynamics, produced altered symptoms, and significantly enhanced immune escape, limiting vaccines ability to offer long-lasting protection against COVID-19.
CoVerage is based on observations from the long-term evolution of certain influenza viruses such as influenza A H3N2. In particular, the model looks for early changes in the virus’s spike proteins, which help the virus attach to human cells. Spike proteins are also the target for vaccines and therapies, the authors said.
Using the GISAID virus genome database, which has more than 16.5 million SARS-CoV-2 sequences uploaded and available, CoVerage looks at SARS-CoV-2 genome data by country of origin for strain dynamics and antigenic changes. The computer model then scans the sequences for changes to the spike proteins, and strains with significant alterations to the spike proteins are identified on a “heat maps.”
Predicts 2 to 3 months earlier than WHO
By continuously reading emerging potential variants of interest from country-wide lineage frequency dynamics, the authors were able to predict VOCs earlier than what is currently seen in practice. To test the model, the authors used it to retrospectively locate and identify known VOCs, including Omicron and JN.1.
“It was interesting to see that the virus variants that were also officially classified as important by the WHO showed significantly higher values in our analyses than other, less noticed variants,” said study author Alice McHardy, PhD, in a press release from the German Center for Infection Research. “These results underscore the ability of our method to effectively predict the emergence of health-relevant SARS-CoV-2 variants with a growth advantage—well before they reach their maximum frequency or are formally identified by the WHO as concerning.”
CoVerage accurately identified 88% of the VOIs and VOCs designated by the WHO since the establishment of SARS-CoV-2 in the human population.
“CoVerage accurately identified 88% of the VOIs and VOCs designated by the WHO since the establishment of SARS-CoV-2 in the human population, on average, more than two months before their official WHO designation,” the authors wrote.
When compared to other currently available web-based tools to identify VOC, CoVerage provided continuously updated resources for the detection of VOCs and links to alternative web-based resources for additional information on these selected lineages, the authors said.
McHardy said the tool, “…could provide valuable time to initiate in-depth analysis required for vaccine adjustments or take targeted measures to protect vulnerable groups, for example.”