Omicron itself is "mysterious". It's believed BA.1 through BA.5 are sibling lineages, having recombined with each other prior to spreading. Possibly in 1 chronic host?
Since it emerged in the same region, it's possible this BA.3 could be from the same source, be it chronic person or animals or w/e

And the November sequences have ~7 nt not in the January sequence, so they should have diverged in August or September I think.

But that January sequence has K356T. Maybe it is picking up advantageous mutations still.

Mostly I just thought it was interesting. The major takeaway is that the drop in sequencing will mean that new variants become dominant more quickly after they're discovered (possibly by months). The lack of severity in recent variants probably shifted the justification for sequencing costs though.

Looking at the point at which we have a 50% chance of identifying the variant, we are currently losing this much amount of weeks compared to previous times:
When BA.2.86 was identified - 2.5 weeks.
XBB discovered - 7 weeks.
Omicron discovered - 8.5 weeks.
Delta discovered - 4.5 weeks.

How long does it take to identify a variant that was growing at the rate that BA.2.86 was, under these different sequencing levels?
For it to be identified, I’ll only consider it necessary for it to show up once.

In the 2 month period prior to Delta being discovered, the world submitted 435k sequences.
2 months prior to Omicron - 1.7 million.
2 months prior to XBB (another major one) - 1.1 million.
2 months prior to BA.2.86 (current major one) - 221k.
2 months prior to today - 96k.

Okay, so we are possibly saving a month on identification over what we used to. But as I said, the world isn't taking as many sequences as it used to. How much does that slow down detection of new variants?
Well that's actually the only part I was interested in sharing in this thread!

However, after Delta, more folks learned about the identification process and have started volunteering time, often without any actual affiliation or compensation.
Since then, interesting variants are identified and shared on the forum the same day/day after they appear on the database.

The first two Delta sequences were submitted/available on the global database on March 23 and March 24 of 2021.
The first mention of this variant on the forum is on April 21, nearly a full month later.

If we see something interesting, we can go and share it on a forum which is reviewed for which variants get a designation.
The time it takes for something interesting to be found and shared is much faster than it used to be.

Once enough people are infected, someone will by pure chance have their infection's virus taken back to a lab for sequencing. The world doesn't sequence as much as it used to.
After this, the lab will upload the genome to a global database, where me and similar folks can see it.

Many of the really interesting COVID variants that have taken over are believed to have come from a chronically infected host.
For some reason, the virus becomes heavily mutated inside the host before going and infecting many others.

There were mixed reads at the residue in question, indicating that possibly the mutation happened after infection, which would mean it is only circulating that way if the patient went on to infect others.