The U.S. now stands at 47 new COVID-19 cases per 100K population per day, down 81% from the peak. Only two states remain above 100 per day (KY, ID). The majority of states are below 50; five states are below 25 new cases per 100K per day (NY-NJ-CT, MD-DC).
Cases are down 39% in the past seven days, a bit slower than in the recent past.
As I said in recent posts, I think the decline in the official new case count is being exaggerated, probably as a result of the growing use of home testing using over-the-counter rapid tests. As an alternative measure that should be unaffected by that, new COVID-19 hospitalizations are down 55% from the Omicron wave peak.
Data source for this and other graphs of new case counts: Calculated from The New York Times. (2021). Coronavirus (Covid-19) Data in the United States. Retrieved 2/15/2022, from https://github.com/nytimes/covid-19-data.” The NY Times U.S. tracking page may be found at https://www.nytimes.com/interactive/2020/us/coronavirus-us-cases.html
Source: Calculated from CDC COVID data tracker, accessed 2/15/2022.
Son-of-Omicron still not a factor in the U.S.
The CDC came out with its weekly update on COVID-19 “variants of concern”, and I’m not quite sure what to make of it.
First, for BA.2 (a.k.a. son-of-Omicron), the CDC revised it’s estimated incidence numbers downward. Last week’s estimate was 3.9%, now that’s been revised downward to about 1.8%. As I wrote at the time, a downward revision was expected, given the growth rate observed in other countries. This week’s estimate is that BA.2 accounts for 3.6% of U.S. new Omicron cases. This percentage is now estimated to be doubling ever week, in line with observed rates in Europe.
Source: CDC COVID data tracker.
Second, the CDC is now breaking out two sub-strains of what was formerly packaged as Omicron. That’s BA.1.1, and B.1.1.529. But they offer no explanation as to why they have now chosen to break those out separately.
Up to now, nearly every source I can find has considered those to be equivalent, up to and including them as being just different names for the same variant. So I have no idea why CDC is now breaking them out separately.
The upshot is that BA.1.1 is squeezing out B.1.529, but I have no clue why this now seems to matter to the CDC.
A sidebar on nomenclature: How B.1.1.529 begat BA.1.1
Those number-and-letter COVID-19 variant names have a well-defined structure, derived from a single software product, Pangolin (Phylogenetic Assignment of Named Global Outbreak LINeages).
For COVID-19, there’s a drag-‘n’-drop website to get any particular COVID-19 genetic sequence identified. Sequence your COVID-19 sample, drag the resulting list of DNA fragments to the website, and it’ll tell you which variant you’re looking at.
Somehow, that seems a whole lot less like rocket science than I had imagined it to be.
In any case, you can read a brief description of the naming rules at this reference.
What, exactly, constitutes a new lineage of COVID-19 — a variant that deserves its own designation — is, near as I can tell, determined by an international committee appointed to oversee this whole process.
That said, they identified two different strains in the original outbreak, given the names A and B. These are the ones that just showed up out of nature, with no known ancestors. So those, and those alone, are designated by a single letter.
Then, after that, from the reference given above:
"Each dot in the numerical suffix means “descendent of” and is applied when one ancestor can be clearly identified. So lineage B.1.1.7 is the seventh named descendent of lineage B.1.1 and C.1 is the first named descendent of lineage C."
Hence, B.1.1.529 is a great-grandchild of the ancestral COVID-19 lineage give the name “B”. The first-ever variant of B that was identified was B.1. The first sub-variant of B.1 to be identified was B.1.1. And the 529th variant of B.1.1 to be identified was B.1.1.529.
They only allow three suffixes (e.g., original generation and three levels of descendants) in a name. After that, they assign a new letter and keep going. Once you get to some significant great-great-grandchild variant, you assign a new letter to its parent, designate the parent as that letter, and now the great-great-grandchild is that new letter-dot-1.
"For example, C is an alias of B.1.1.1 hence the descendant of B.1.1.1 is called C.1 (rather than B.1.1.1.1"
Weirdly enough, whereas all the other parts of this are hierarchical, the assignment of new letters just walks through the alphabet. So you have to have some explicit table that tells you what new letter is derived from what old strain. You can find such a table — what appears to be a comprehensive list of COVID-19 lineages — at this reference.
When they run out of letters, they start using multiple letters, just like the column designations in Excel. After Z comes AA, after AZ comes BA, and so on. And so the “BA” designation was just the next available pair of letters, once they ran past three generations of some prior strain. It has nothing to do with the fact that BA is derived from the original strain B. That’s purely by chance.
In this case, from the same reference above, we find that BA.1 is “Alias of B.1.1.529.1” So, BA = B.1.1.519, and BA.1 is the first significant sub-variant of BA.1.1.529 to be discovered.
There is also the potential for recombinant lineages, which I believe means that they contain parts of multiple different ancestors, not just one. I don’t think that matters, yet, for discussion of variants of concern, but those are all given an X prefix.
So the answer to the puzzle is that BA.1.1 is a grandchild of the strain B.1.1.529. It’s the same as B.1.1.529.1.1.
I still have no idea why CDC has decided to break that out separately. But at least I have some idea what it is.
As a final consequence, BA.2 isn’t really son-of-Omicron. More properly, it’s “second son of Omicron”. It’s B.1.1.529.2.
