Ever since the novel coronavirus, SARS-CoV-2, began jumping from human to human, it’s been mutating. The molecular machinery the virus uses to read and make copies of its genetic code isn’t great at proofreading; minor typos made in the copying process can go uncorrected. Each time the virus lands in a new human victim, it infects a cell and makes an army of clones, some carrying genetic errors. Those error-bearing clones then continue on, infecting more cells, more people. Each cycle, each infection offers more opportunity for errors. And, over time, those errors, those mutations, accumulate.

Some of these changes are meaningless. Some are lost in the frenetic viral manufacturing. But some become permanent fixtures, passed on from virus to virus, human to human. Maybe it happens by chance; maybe it’s because the change helps the virus survive in some small way. But in aggregate, viral strains carrying one notable mutation can start carrying others. Collections of notable mutations start popping up in viral lineages, and sometimes they seem to have an edge over their relatives. That’s when these distinct viruses—these variants—get concerning.

Scientists around the world have been closely tracking mutations and variants since the pandemic began, watching some rise and fall without much ado. But in recent months, they have become disquieted by at least three variants. These variants of concern, or VOCs, have raised critical questions—and alarm—over whether they can spread more easily than previous viral varieties, whether they can evade therapies and vaccines, or even whether they’re deadlier.

Here, we’ll run down what we know and what we don’t know about these variants. With much research yet to be done, there’s a lot of unanswered questions. But researchers are working quickly to address the most important unknowns. High on the list is whether the vaccines we already have will be effective against the variants. So far, it seems likely that they will be. Still, the virus is sending a clear message: with rampant transmission accelerating viral evolution, more variants will arise and we need to be prepared.

With more data becoming available by the day, we’ll update this story with significant findings as they come along. Before we get to the data we have, a quick note on names: it’s problematic to identify diseases or infectious agents—in this case, virus variants—based on where they were identified. Such geographic associations risk creating stigma and may discourage reporting, so there is an active discussion in the scientific community about how best to name the current variants. In the interim, it has become all too common to refer to these by their country of origin. We’ll try to avoid that as much as possible while making clear which variants we’re talking about.


Alternate names: 501Y.V1 and VOC 202012/01
Geographic association: United Kingdom
Number of countries reporting cases: 70
Increased transmissibility: Yes
Increased disease severity/mortality: A “realistic possibility”
Vaccine efficacy: Still effective

In early December 2020, researchers and officials in the UK began warning of a new variant that seemed to be spreading abnormally fast while carrying an unusually large number of mutations—23. The first record of the variant in the UK stretched back to two samples taken from infected people on September 20 and September 21. In a matter of weeks, the variant began making up a larger and larger proportion of total cases there. Researchers quickly suspected the variant had evolved to become more transmissible—that is, it’s able to spread more easily from person to person.


Data analyses since December have supported that hypothesis, but researchers are still working out exactly how much more transmissible it is compared to earlier versions. In early January, UK researchers released preliminary results from a series of models that estimated the variant tacks on an additional 0.36 to 0.68 onto SARS-CoV-2’s observed reproduction number. That means, on average, people infected with B.1.1.7 will go on to infect an additional 0.36 to 0.68 people on top of how many they would have infected if they were carrying an earlier version of the virus. More recent estimates have been roughly in this range, suggesting B.1.1.7 has around a 47 percent or 56 percent increase in transmission.

B.1.1.7 has now been detected in more than 60 countries beyond the UK, including the United States, where it has been found in at least two dozen states. A modeling study published by the US Centers for Disease Control and Prevention on January 15 estimated that it will become the predominant strain in the US in March.


Some of the mutations B.1.1.7 carries seem to help explain the virus’s newfound ability. The variant carries 23 mutations in all: 13 mutations that change the virus’s protein sequences (non-synonymous), four deletions, and six synonymous mutations. Of B.1.1.7’s mutations, eight occur in the virus’s spike protein, the now notorious club-like protein that juts out from the virus’s spherical particle. That spike is what the virus uses to latch onto and infect cells, which the protein accomplishes by binding a receptor on the outside of human cells called ACE2.

So far, we know that at least three of B.1.1.7’s eight spike mutations may be relevant to the variant’s boosted transmission. Chief among them is a mutation that changes one of the spike proteins’ critical amino acids—the amino acid at position 501 of spike’s protein sequence. Specifically, the mutation changes the amino acid at 501 from an asparagine (N) to a tyrosine (Y), so the mutation is written as N501Y. The 501 amino acid is critical because it lies within the area of spike that directly binds to ACE2—called the receptor binding domain (RBD)—and it is one of just six key contact residues in the RBD. Lab experiments have suggested that changing from an N to a Y at 501 increases spike’s ability to bind ACE2, and experiments in mice linked the mutation to increased infectiousness and disease.

After N501Y, there’s P681H. The mutation at position 681—changing the amino acid from a proline (P) to a histidine (H)—falls near a unique furin cleavage site on SARS-CoV-2’s spike protein. For SARS-CoV-2 to successfully get into a cell after binding ACE2, the spike protein needs to be cleaved into its two subunits by enzymes. The split changes spike’s conformation and activates it, allowing it to fuse itself to the cell membrane and dump its contents into the now-infected cell. In animal studies, the furin cleavage site seemed to boost the virus’s ability to enter cells. Researchers suspect the new mutation may boost entry further.

A patient prepares to receive an injection of the Oxford/AstraZeneca COVID-19 vaccine by Royal Navy medics at a vaccination center set up at Bath racecourse in Bath, southwest England.
Enlarge / A patient prepares to receive an injection of the Oxford/AstraZeneca COVID-19 vaccine by Royal Navy medics at a vaccination center set up at Bath racecourse in Bath, southwest England.

Adrian DENNIS / AFP / Getty Images

The third spike mutation known to be significant is a deletion of six nucleotides in its genetic code, which leads to the loss of two amino acids at positions 69 and 70 in the spike protein. It’s unclear what this deletion does for the virus exactly, but it has arisen a number of times in different lineages, suggesting it offers an advantage. For now, there is one clear consequence for researchers: the deletion messes up a diagnostic test for SARS-CoV-2. The test is a three-target RT-PCR test, meaning it works by detecting three snippets of the SARS-CoV-2 genome, including one in the gene that codes for spike. When this 69-70 deletion is present, the test will show up negative for the spike gene but positive for the other two SARS-CoV-2 genetic sequences. This result is referred to as “S gene dropout” and is now used to help identify infections caused by B.1.1.7.

These three mutations are the most notable in B.1.1.7 for now. There’s scant data on the other 20, but researchers are working swiftly to assess what each might do on its own or in combination with the others.

Disease severity/mortality

When researchers first raised concerns about B.1.1.7, all of those issues related to increased transmissibility. Preliminary evidence looking at infection outcomes did not suggest that B.1.1.7 was causing more severe disease or more deaths than other virus strains. Still, some saw little comfort in this, given that any increase in the total number of infections still leads to more severe cases and deaths in absolute numbers.

The situation took a darker turn January 21, when a UK government advisory group—NERVTAG—found preliminary evidence that “there is a realistic possibility that infection with VOC B.1.1.7 is associated with an increased risk of death compared to infection with non-VOC viruses.”

So far, some experts are not yet convinced by the preliminary evidence presented, and they’re calling for much more data before any conclusions are drawn. For one thing, the full data sets behind some of the analyses done so far have not been published, and some of them relied on comparing small numbers of deaths in people infected with B.1.1.7 with larger numbers of deaths in people infected with other strains. Some experts also wonder whether the calculated increase in deaths could simply be explained by overburdened hospitals rather than a deadlier variant.

Vaccine efficacy

With increased infectiousness and the possibility of being deadlier, a critical question raised by B.1.1.7 is whether or not the current vaccines we have—mRNA vaccines from Pfizer/BioNTech and Moderna—will work against the variant. So far, the answer appears to be yes.

On January 19, researchers at Pfizer and BioNTech released a non-peer reviewed study where they pitted antibody-laden blood from 16 people given their mRNA vaccine (BNT162b2) against a pseudovirus that carried B.1.1.7’s mutated spike protein. The researchers found that the vaccines’ antibodies were just as good at neutralizing the pseudovirus with B.1.1.7’s mutated spike protein as they were at neutralizing a pseudovirus with the spike protein from a reference SARS-CoV-2 virus.

“These data… make it unlikely that the B.1.1.7 lineage will escape BNT162b2-mediated protection,” the researchers concluded.

Likewise, on January 25, Moderna released its own non-peer reviewed study, which was similar in design. They tested the antibodies from eight people given their mRNA vaccine against a pseudovirus bearing B.1.1.7’s mutated spike protein. Again, the antibodies neutralized the pseudovirus at levels comparable to those seen with a pseudovirus carrying a reference spike protein.

Yet another similar study, led by researchers at Columbia University and released January 26, found the same results. Antibodies from 12 people who received Moderna’s vaccine and 10 people who received Pfizer’s vaccine were able to neutralize a pseudovirus containing B.1.1.7’s mutated spike protein, with only a modest drop in potency compared with neutralization of a pseudovirus carrying a reference spike protein.