Why SARS-CoV-2 variants had so many names in January 2021
Andrew Page talks with Niamh Tumelty, Head of Research Services and STEM Libraries at the University of Cambridge, about the confusing names given to SARS-CoV-2 variants. They discuss the overlapping meanings of lineage, variant and strain, and why several naming schemes can describe the same virus population. Recorded on 16 January 2021, the conversation also examines incomplete data, hurried reporting and the evidence then available about transmission and immune responses.
Listen to Episode 39: Explaining the naming of SARS-CoV-2 new variants on SoundCloud
In this episode
Routine changes and the meaning of a variant
Niamh starts by asking why changes that previously produced different lineages are now being described as new variants. Andrew says SARS-CoV-2 had been accumulating roughly two genome changes a month, providing markers for tracking its spread. He recalls a lineage first identified in Spanish agricultural workers, spreading through communities and tourists, and accounting for virtually all UK cases during September, October and November 2020.
Andrew describes lineage, variant and strain as overlapping, inconsistently used terms, even among specialists. His work uses lineages to group similar genomes for genomic epidemiology. He does not offer a fixed threshold at which a lineage becomes a new variant; instead, he connects the recent terminology to public-health investigations and concern about particular changes.
Several names for the same UK variant
Andrew traces the UK naming confusion to Public Health England’s designation VUI 2020-12-01: a variant under investigation, identified by year, month and its order within that month. Further investigation led to a variant-of-concern designation, while reporting used labels such as the Kent variant, UK variant and simply the new variant.
Other names included 20I/501Y.V1 and the lineage designation B.1.1.7. Andrew breaks the former into a year, a letter he compares with hurricane naming, the 501Y change thought to be linked to increased transmission, and V1 for the first such variant identified. He says he often includes three or four identifiers when communicating because he cannot know which scheme the recipient uses.
A similar problem applies to the rapidly spreading lineage identified in South Africa: the discussion gives both 501Y.V2 and B.1.351, and notes that a similar mutation arose independently.
Metadata, rapid naming and geographical labels
From a librarian’s perspective, Niamh questions why several names are needed for the same thing. Andrew explains that groups are working rapidly, with some concentrating on phylogenetic relationships and others on changes that raise public-health concerns. Names may be assigned within hours so that researchers and the press have a shared reference.
The WHO is described as trying to bring people together around naming. Meanwhile, discussion of constellations of mutations can make technical descriptions longer and harder to communicate. The conversation acknowledges why the media reach for geographical shorthand. Andrew nevertheless warns against country-based names because they can stigmatise places and persist, citing Spanish flu and Trump’s use of China flu.
Distinguishing P.1 and P.2
Andrew stresses that the recording date, 16 January 2021, matters because the situation changes daily. A Japanese group had identified concerning changes in travellers from Brazil, but the initial announcement was followed by limited data and then further Brazilian reports. Two variants being discussed were labelled P.1 and P.2.
He describes P.1 as the greater concern, with many changes, including one he refers to as E484, and connects it with Amazonia and Manaus. As of the previous night, he says no UK cases of that type had been identified. A similar variant with a similar history but lacking most of those mutations had been found, however. Andrew reports finding it in East of England data and manually verifying mutations, emphasising the need for careful checks before alerting clinicians. Confusion over naming had fuelled misleading reporting.
Answering questions before the data arrive
Niamh highlights the difficulty of responding to global interest while research is still emerging. Andrew recalls a journalist asking about the newly reported lineage from Brazil when the available information amounted to a screenshot of a tweet translated from Japanese using Google Translate. A Japanese-speaking member of his group helped interpret it.
At that point, he says, the data were not yet available in GISAID. Depositing, releasing and analysing them took another day or two. Subsequent posts by Brazilian groups on virological helped clarify the findings. The example shows how public requests for explanations can arrive before researchers have enough information to distinguish variants reliably.
Transmission, immunity and unresolved effects
Andrew separates rapidly available observations from effects requiring laboratory investigation. For B.1.1.7, he points to its rise during lockdown while other lineages declined. In Norfolk, he says it had stabilised at about 80% of new cases after the first local case around mid-November. Niamh notes that increased transmission matters even if the range of symptoms remains similar, because more infections mean more people experiencing those outcomes.
For other variants, Andrew discusses concerns about altered antibody responses, but says their effects still need testing. A computer prediction may not match reality, and combinations of mutations may matter more than individual changes. He also raises the possibility that some changes could reduce disease severity. Vaccines were being deployed, and he reports no vaccine escape observed at that stage; infections in vaccinated people were a future issue to investigate, not an established finding in this discussion.
Highlights
- 00:01:25 — Routine genome changes and the spread of a lineage through European summer travel.
- 00:02:42 — Why lineage, variant and strain have overlapping, fuzzy meanings.
- 00:03:45 — Public Health England’s variant-under-investigation designation and VUI 2020-12-01.
- 00:06:03 — Multiple identifiers for the UK variant, including 20I/501Y.V1 and B.1.1.7.
- 00:08:53 — Niamh questions why the same variant needs so many different labels.
- 00:10:12 — Names assigned within hours and efforts to coordinate naming.
- 00:11:30 — Japanese findings in travellers from Brazil and the distinction between P.1 and P.2.
- 00:14:02 — Responding to journalists using translated tweets before data were available in GISAID.
- 00:15:41 — Why geographical names can stigmatise countries and persist.
- 00:17:06 — Limited time to study variant effects, transmission evidence and unresolved immune questions.
In their own words
I don’t understand why they don’t pick one and then call it that. Why have so many different ones?
— Niamh Tumelty, 00:08:53
Who is talking
- Andrew Page (host)
- Niamh Tumelty (guest, University of Cambridge)
- Lee Katz (host)
Also mentioned: Nabil-Fareed Alikhan.
Tools and resources mentioned
GISAID, Google Translate.
Questions this episode answers
What is the difference between a SARS-CoV-2 lineage and a variant?
Andrew describes the terms as overlapping rather than sharply defined. He uses lineages to group genetically similar genomes for epidemiological analysis, while the prominent use of variant in this discussion is tied to investigations of potentially concerning changes.
Why did B.1.1.7 have several different names?
Different groups were naming viruses for different purposes, including phylogenetic classification and public-health investigation. The episode discusses B.1.1.7, 20I/501Y.V1 and successive investigation and concern labels, alongside geographical names used in reporting.
What caused confusion over P.1 and P.2 in January 2021?
Early reports from Japan and subsequent Brazilian data concerned two distinct variants, but incomplete information and geographical shorthand blurred the distinction. Andrew says the more concerning P.1 had not been identified in the UK as of the previous night, whereas he had verified a similar variant, lacking most of its mutations, in East of England data.
What evidence for increased B.1.1.7 transmission was discussed?
Andrew reports that B.1.1.7 increased during lockdown while other lineages declined, reaching about 80% of new cases in Norfolk. He distinguishes those observations from questions about antibody responses and other biological effects that still required investigation.
This page was written from a machine transcript of the episode and checked against it. Listen to the episode for the whole conversation.
Show notes
Andrew talks to Niamh Tumelty from the University of Cambridge about SARS-CoV-2 ‘new variants’ and tries to clear up some of the confusion around all the names flying around. Hopefully this helps to give some insights into the various names you hear, but probably by the time you listen the whole thing will have changed again since this field moves so rapidly. Andrew apologises in advance for all the errors that will be found in this podcast! If you want to read a bit more theres an interesting news item on Nature: https://www.nature.com/articles/d41586-021-00097-w