Ryan Hisner
ryanhisner.bsky.social
Ryan Hisner
@ryanhisner.bsky.social
Teacher. Learner. Investigating mysteries of SARS-CoV-2 evolution. LongDesertTrain on another platform.
New Cryptic (North Carolina) derived from a 2020, pre-B.1.1 lineage, meaning from an infection that has lasted nearly six years.

It has one of the classic patterns of Cryptic reversions to Bat-CoV/SARS-1 spike residues: Q498Y-N501T.

Eager to see the rest of the genome.
January 24, 2026 at 11:03 PM
New @dropsitenews.com video from opposite side confirms: man in gray coat & red-billed cap disarmed the victim—who was legally carrying & never reached for or touched his gun—before any shots were fired.

See his mittened, empty hands reaching to the victim's waistband, then emerging with the gun.
January 24, 2026 at 7:59 PM
Hell yes. Thank you for doing what you're doing.
January 24, 2026 at 12:43 AM
Can you imagine how Kepler would have reacted had he known about the geometry of viral capsids? Pretty sure he would've written books about it and expanded it into a theory of all life.
January 24, 2026 at 12:42 AM
Several of the A->G muts here have been seen in previous A->G clusters.

And as always, the entire A->G cluster happens all at once—never stepwise.

Phylogenetic tree courtesy of the @angieshinrichs.bsky.social-curated SARS-CoV-2 Usher tree. 2/2
January 22, 2026 at 11:28 PM
One new A->G cluster showed up today in an XFG from Brazil. 10/13 private nuc mutations here are A->G, and eight of them appear in a 229-nt stretch in the spike NTD.

Still no idea what's behind these, though some sort of unusual ADAR fusillade seems to be the best guess. 1/2
January 22, 2026 at 11:28 PM
Yes, this was a memorable post of his on mRNA vaccines and how their use is tantamount to eugenics.
January 21, 2026 at 2:10 AM
Yes, alongside some other bold claims.
January 21, 2026 at 2:03 AM
Still more new NSP1 mutations in the BA.3.2 uploaded from the Netherlands today. We've never seen anything like this kind of rapid evolution in NSP1 (whose primary role is to shut down host-protein translation and degrade host mRNA).

I've labeled all BA.3.2-specific NSP1 muts here. 1/3
January 20, 2026 at 2:09 PM
No comparison to influenza in the macaque study (due to limited numbers of macaques/resources I imagine), but elsewhere experiments with precision cut human lung slices found that SARS-CoV-2 productively infects alveolar macrophages during acute infection while IAV does not.
January 18, 2026 at 5:11 PM
The near absence of RBD spike muts in chronic-infection BAL-sample sequences fits neatly w/the cell-cell spread they found in alveolar macs—as do some other aspects. Less clear why this might incr E antibodies, but the extraordinarily high rate of E mutations in BAL seqs hints at a connection.
January 18, 2026 at 3:33 PM
There are many other really interesting findings in that paper—including an intriguing counterbalancing act of IFN-γ & MHC-E in alveolar macs that could help explain persistence (IFN-γ associated w/viral control but also increased MHC-E expression, which inhibits NK cytotoxicity).
January 18, 2026 at 3:33 PM
They found that RNA- and antigen-positive BAL samples contained infectious virus that productively replicated in alveolar macrophages in cell culture. Even a few of the RNA- and antigen-negative samples showed evidence of productive replication (84% showed increasing levels of spike over time).
January 18, 2026 at 3:33 PM
...they had chronic infection in the alveolar compartment. A majority of macaques tested + for viral RNA (68%) or antigen (80%). The typical explanation for this is that these are fragments of lingering viral material and that no viable virus remains. But...
January 18, 2026 at 3:33 PM
It's also consistent w/the findings of a non-human primate experiment in which macaques completely cleared all viral RNA from from their nasal tract & trachea within 3 weeks—like the typical human—but who nonetheless had elevated inflammatory markers for >1 yr (duration of study). It turned out...
January 18, 2026 at 3:33 PM
From Scott's "Two Cheers for Anarchism," a short collection of essays & one of my favorite books, which I hand out to anyone that might be interested as often as possible.
January 17, 2026 at 3:31 PM
Good reason to practice James C Scott's "anarchist calisthenics"

"You know, you & especially your grandparents could have used more of a spirit of lawbreaking. One day you will be called on to break a big law in the name of justice & rationality. Everything will depend on it. You have to be ready"
January 17, 2026 at 1:50 PM
I think there's a good chance this is why, in by far the best & largest molnupiravir trial we have, outcomes among immunocompromised patients treated with MOV were worse that for those not treated with MOV.

And this is of course the exact population most likely to receive MOV now.
3/3
January 16, 2026 at 12:26 PM
For people treated with MOV who do not clear the virus shortly after infection—which we know to be a common occurrence—this is a bad.

MOV is supposed to cause so many harmful mutations that the virus dies out quickly. Instead, it's clearly abetting adaptive mutations. 2/3
January 16, 2026 at 12:26 PM
Another day, another sequence displaying undeniable evidence of positive selection of molnupiravir-induced mutations (i.e. adaptive mutations—good for the virus, bad for the infected person).

G->A is the classic MOV mutation—and the vast majority of the positive selection is in G->A muts. 1/3
January 16, 2026 at 12:26 PM
Another weird aspect of BA.3.2 evolution so far: an extraordinary number of NSP1 mutations. NSP1 is only 180 AA, but a new mutation shows up almost every day there, often on top of previous new ones. The @nextstrain.org visual here can't even accommodate all the NSP1 muts—at least 2 are invisible.
January 15, 2026 at 11:54 PM
ORF6 is only 61 AA long, but its primary known function (shutting down nucleus-cytoplasm trafficking) is mediated by its C-terminal domain (i.e. the last few AA). This stop codon slices off those AA. This is the last variant I ever expected to see an ORF6 loss-of-function mutant in. Weird. 3/3
January 15, 2026 at 11:36 PM
...if that's the case, then I expect we'll see a JN.1/BA.3.2 recombinant with BA.3.2's spike but with ORF7a (and 7b and 8) restored.

Will add one extremely weird fact here about BA.3.2: there are two unrelated branches with ORF6:Q56* (Denmark & Australia, 5 sequences total). 2/3
January 15, 2026 at 11:36 PM
January 11, 2026 at 11:45 PM
Another great passage from Yankee:

"There are wise people who talk ever so knowingly and complacently about 'the working classes,' and satisfy themselves that a day’s hard intellectual work is very much harder than a day’s hard manual toil & is righteously entitled to a much bigger pay.... "
January 11, 2026 at 9:05 PM