Seven months ago, White House COVID-19 response coordinator Jeffrey Zients declared the scale of the U.S. effort to sequence the pandemic coronavirus “totally unacceptable.” At that point, the country had managed to read the genetic blueprints of just 0.36% of all the coronavirus samples collected from U.S. noses and throats—far too few to detect rare, potentially dangerous mutations early and prepare to fight them.
But today, even as new U.S. coronavirus cases exceed 160,000 a day—a number not seen since late January—public health experts have a much better sense of their foe’s changing genetics. By July, aided by $1.7 billion in funding from the American Rescue Plan, expanded sequencing operations were reading out 8.6% of new coronavirus cases.
Weekly submissions to the leading international sequence repository, GISAID, “right now are well north of 40,000,” says Duncan MacCannell, who heads the Centers for Disease Control and Prevention (CDC) program that coordinates U.S. genomic surveillance. By comparison, it took 11 months—until November 2020—for the country’s first 40,000 sequences to be deposited. “We are doing amazingly” in numbers of samples sequenced, says Trevor Bedford, a computational biologist at the Fred Hutchinson Cancer Research Center. “I have no concerns about a new variant slipping by unnoticed.”
At the moment, sequencers around the country are seeing a deluge of Delta. The highly contagious variant accounted for 99.1% of cases in the United States by late August, up from 7.5% in late May, according to CDC. “We saw Delta sweep in in the blink of an eye,” says integrative biologist Lauren Ancel Meyers of the University of Texas, Austin. “But that can change.”
For example, coronavirus watchers noted in July that another variant, Mu, accounted for 9% of 101 sequenced cases in Florida’s Jackson Memorial Health System and the University of Miami’s UHealth Tower. Mu was first spotted in Colombia, where it made up 69% of cases in early July; last week the World Health Organization (WHO) declared it a variant of interest (VOI). Mu carries a mutation, E484K, that likely helps SARS-CoV-2 blunt vaccine and infection-induced immunity.
But Delta seems to readily outcompete Mu. In the Miami hospitals “by early August we were at approximately 97% Delta” and remain there, says pathologist David Andrews of the University of Miami. In Colombia, as of late August, Mu was down to 63% whereas Delta had surged from 6% to 19%. In Andrews’s samples, Delta “has displaced virtually all other lineages,” he says.
“The virus could surprise us but I would bet that Delta is going to fully sweep and [become] the common ancestor of all viruses in the next few months,” Bedford says. That’s likely “where the next variant will emerge from.”
Some states might still be slow to spot a new variant. Nineteen, all but one in the Midwest, South, and Southeast, have sequenced fewer than 2% of their cumulative cases. Oklahoma, where hospitals are overwhelmed with COVID-19 cases, is last among them: It has uploaded 1617 sequences, or 0.3% of cases, to public repositories.
Still, in most states, the percentage of samples sequenced is orders of magnitude higher than it was in February. Wyoming, the leader, has sequenced 20% of its cases. “Our lab is essentially sequencing every positive sample they receive,” says Alexia Harrist, the state epidemiologist. Maine, where cumulative sequencing is at 7.34%, sequenced 21% of new positive cases in the 3 weeks ending 14 August.
Those sequences “are a true random assortment,” from around the state, says Nick Matluk of the Maine State Health and Environmental Testing Laboratory. The lab’s staff has grown from 15 to 50 during the pandemic, including National Guard members deployed to do data entry and dispose of biowaste, and its facilities include a contracted 18-wheeler containing an enhanced biosafety level 2 lab.
In Mississippi, surveillance now reaches 75% of counties, and more than 99% of the samples are Delta. But, “That 1% is interesting to keep an eye on because that might be what seeds the next wave,” says Ashley Robinson, a microbiologist at the University of Mississippi Medical Center genomics core facility. For example, in late August a WHO-designated VOI named Lambda popped up in a single sample from a small town in southwestern Mississippi. Enabled by CDC funding, “I threw in a couple of extra samples from that little town” for last week’s analysis, Robinson says.
Yet problems remain in the U.S. scale-up. They include the fragmented health care system, in which states have different rules about data sharing. For example, the public CDC dashboard shows data for California only as a whole, not county by county, restricting data analysis by outside scientists.
Bedford says connecting the country’s plethora of sequences to patient data remains a key problem, one that countries like the United Kingdom and Israel, with interconnected patient data, do not face. “What proportion of breakthrough cases are Delta versus Alpha? What proportion of hospitalized cases are different variants?” he asks. Answering such questions in the United States “is much harder to do.”