Sam Scarpino
@scarpino.bsky.social
Scientist. Director of AI + Life Sciences and Professor of Public Health Sciences at Northeastern University. External Faculty at the Santa Fe Institute and Vermont Complex Systems Institute. scarpino.github.io
So overwhelmingly K (that's J.2.4.1 right?)
November 10, 2025 at 7:21 PM
So overwhelmingly K (that's J.2.4.1 right?)
6/ Normally we wouldn't expect Hawaii to lead the US, but this is consistent with their stronger mobility connections to Japan where flu has been active for over a month.
www.nature.com/articles/d41...
www.nature.com/articles/d41...
Japan declares a flu epidemic — what this means for other nations
Researchers say that the number of infections for this time of year is unusual.
www.nature.com
November 10, 2025 at 6:50 PM
6/ Normally we wouldn't expect Hawaii to lead the US, but this is consistent with their stronger mobility connections to Japan where flu has been active for over a month.
www.nature.com/articles/d41...
www.nature.com/articles/d41...
5/ Also interesting that the other state showing signs of influenza activity is Hawaii.
Similar to the UK, Japan is experiencing an earlier flu season than normal likely being driven the by the H3N2 "K" variant.
japantoday.com/category/nat...
Similar to the UK, Japan is experiencing an earlier flu season than normal likely being driven the by the H3N2 "K" variant.
japantoday.com/category/nat...
Flu cases top alert threshold in Japan as infections spread nationwide
Japan's influenza cases averaged 14.90 per medical institution during the week through last Sunday, surpassing the alert threshold of 10 as infections increased across all 47 prefectures, the health ministry said. The Ministry of Health, Labor and Welfare said Friday that 57,424 influenza cases were reported nationwide for seven days…
japantoday.com
November 10, 2025 at 6:50 PM
5/ Also interesting that the other state showing signs of influenza activity is Hawaii.
Similar to the UK, Japan is experiencing an earlier flu season than normal likely being driven the by the H3N2 "K" variant.
japantoday.com/category/nat...
Similar to the UK, Japan is experiencing an earlier flu season than normal likely being driven the by the H3N2 "K" variant.
japantoday.com/category/nat...
4/ Wastewater SCAN data from Georgia also shows a sharp rise in influenza concentrations, which is being driven by H3 SCAN doesn't subtype the Ns, but is seems like safe bet that this is the H3N2 K variant I discussed in my thread yesterday. www.wastewaterscan.org/en
November 10, 2025 at 6:50 PM
4/ Wastewater SCAN data from Georgia also shows a sharp rise in influenza concentrations, which is being driven by H3 SCAN doesn't subtype the Ns, but is seems like safe bet that this is the H3N2 K variant I discussed in my thread yesterday. www.wastewaterscan.org/en
3/ Interestingly, Walgreens data through Nov 1st, showed no signs of influenza-specific activity. That has changed with the data through Nov 8th. While rates are still quite low, they trend is consistent with us entering the flu season.
November 10, 2025 at 6:50 PM
3/ Interestingly, Walgreens data through Nov 1st, showed no signs of influenza-specific activity. That has changed with the data through Nov 8th. While rates are still quite low, they trend is consistent with us entering the flu season.
2/ Walgreens data through Nov 8th shows much higher levels of generalized respiratory infection and also the first signs of influenza activity.
You can see the dashboard and learn about their methods (not very transparent, but at least we have something) here: www.walgreens.com/seasonal/res...
You can see the dashboard and learn about their methods (not very transparent, but at least we have something) here: www.walgreens.com/seasonal/res...
November 10, 2025 at 6:50 PM
2/ Walgreens data through Nov 8th shows much higher levels of generalized respiratory infection and also the first signs of influenza activity.
You can see the dashboard and learn about their methods (not very transparent, but at least we have something) here: www.walgreens.com/seasonal/res...
You can see the dashboard and learn about their methods (not very transparent, but at least we have something) here: www.walgreens.com/seasonal/res...
That's the assumption, but we will need more data from both the UK and US to be confident. Nevertheless, the vaccine should provide some protection against severe disease. Given that's it's often free in the US and has very very minimal side effects risk, it's important to get vaccinated.
November 10, 2025 at 4:04 PM
That's the assumption, but we will need more data from both the UK and US to be confident. Nevertheless, the vaccine should provide some protection against severe disease. Given that's it's often free in the US and has very very minimal side effects risk, it's important to get vaccinated.
14/ My guess is that the K variant is here and that cases will start rising consistently in the US in the coming weeks. We'll want to watch the UK closely as they are effectively a couple weeks ahead of us.
The vaccines should still reduce severity!
I'd get mine now if you haven't already.
The vaccines should still reduce severity!
I'd get mine now if you haven't already.
November 10, 2025 at 2:16 AM
14/ My guess is that the K variant is here and that cases will start rising consistently in the US in the coming weeks. We'll want to watch the UK closely as they are effectively a couple weeks ahead of us.
The vaccines should still reduce severity!
I'd get mine now if you haven't already.
The vaccines should still reduce severity!
I'd get mine now if you haven't already.
13/ Pulling this together, I'd say the evidence is consistent with low, but rising levels of flu in the US. I'm basing this on wastewater data showing some rising levels, but nothing consistent and no signal from Walgreens. Seeing genomes likely coming from imported cases supports this conclusion.
November 10, 2025 at 2:16 AM
13/ Pulling this together, I'd say the evidence is consistent with low, but rising levels of flu in the US. I'm basing this on wastewater data showing some rising levels, but nothing consistent and no signal from Walgreens. Seeing genomes likely coming from imported cases supports this conclusion.
12/ Grain of salt continued...
It also looks like a fair number of the K variant genomes came from individuals in the military or via the traveler surveillance program. I'm making a min. educated guess here based on meta-data, but this may mean we're seeing imported cases not local transmission.
It also looks like a fair number of the K variant genomes came from individuals in the military or via the traveler surveillance program. I'm making a min. educated guess here based on meta-data, but this may mean we're seeing imported cases not local transmission.
November 10, 2025 at 2:16 AM
12/ Grain of salt continued...
It also looks like a fair number of the K variant genomes came from individuals in the military or via the traveler surveillance program. I'm making a min. educated guess here based on meta-data, but this may mean we're seeing imported cases not local transmission.
It also looks like a fair number of the K variant genomes came from individuals in the military or via the traveler surveillance program. I'm making a min. educated guess here based on meta-data, but this may mean we're seeing imported cases not local transmission.
11/ Grain of salt about the genomes:
1. The sample sizes are really small (e.g., 7 genomes in Oct meaning the CIs would range from ~12 - 75% depending on the method).
2. No genomes collected after Oct. 10th (not unusual as it takes awhile to sequence & post) but that means we're a month behind.
1. The sample sizes are really small (e.g., 7 genomes in Oct meaning the CIs would range from ~12 - 75% depending on the method).
2. No genomes collected after Oct. 10th (not unusual as it takes awhile to sequence & post) but that means we're a month behind.
November 10, 2025 at 2:16 AM
11/ Grain of salt about the genomes:
1. The sample sizes are really small (e.g., 7 genomes in Oct meaning the CIs would range from ~12 - 75% depending on the method).
2. No genomes collected after Oct. 10th (not unusual as it takes awhile to sequence & post) but that means we're a month behind.
1. The sample sizes are really small (e.g., 7 genomes in Oct meaning the CIs would range from ~12 - 75% depending on the method).
2. No genomes collected after Oct. 10th (not unusual as it takes awhile to sequence & post) but that means we're a month behind.
10/ I looked at publicly available flu genomes from the US. There are 68 genomes from H3N2 viruses that have been posted online recently.
In the figure, I look at the raw proportion of those genomes that belong to the K variant/clade.
But, you should take these data with a HUGE grain of salt...
In the figure, I look at the raw proportion of those genomes that belong to the K variant/clade.
But, you should take these data with a HUGE grain of salt...
November 10, 2025 at 2:16 AM
10/ I looked at publicly available flu genomes from the US. There are 68 genomes from H3N2 viruses that have been posted online recently.
In the figure, I look at the raw proportion of those genomes that belong to the K variant/clade.
But, you should take these data with a HUGE grain of salt...
In the figure, I look at the raw proportion of those genomes that belong to the K variant/clade.
But, you should take these data with a HUGE grain of salt...
9/ It looks like Walgreens updates data on Sat/Sun, so I'm surprised we don't have more recent information.
I'll post an update once their dashboard is refreshed.
You can read about what they're doing here:
corporate.walgreens.com/newsroom/wal...
I'll post an update once their dashboard is refreshed.
You can read about what they're doing here:
corporate.walgreens.com/newsroom/wal...
Walgreens Launches Enhanced Respiratory Index to Track Flu and COVID-19 Hotspots - Walgreens Corporate Site
corporate.walgreens.com
November 10, 2025 at 2:16 AM
9/ It looks like Walgreens updates data on Sat/Sun, so I'm surprised we don't have more recent information.
I'll post an update once their dashboard is refreshed.
You can read about what they're doing here:
corporate.walgreens.com/newsroom/wal...
I'll post an update once their dashboard is refreshed.
You can read about what they're doing here:
corporate.walgreens.com/newsroom/wal...
8/ Walgreens reports weekly on COVID, flu, and respiratory activity.
They use data collected from their stores (like sales and positive tests).
As of Nov 1st, they are reporting essentially ZERO flu activity.
It looks like they update on Sat or Sun, so I'll add last week's data to this thread.
They use data collected from their stores (like sales and positive tests).
As of Nov 1st, they are reporting essentially ZERO flu activity.
It looks like they update on Sat or Sun, so I'll add last week's data to this thread.
November 10, 2025 at 2:16 AM
8/ Walgreens reports weekly on COVID, flu, and respiratory activity.
They use data collected from their stores (like sales and positive tests).
As of Nov 1st, they are reporting essentially ZERO flu activity.
It looks like they update on Sat or Sun, so I'll add last week's data to this thread.
They use data collected from their stores (like sales and positive tests).
As of Nov 1st, they are reporting essentially ZERO flu activity.
It looks like they update on Sat or Sun, so I'll add last week's data to this thread.
7/ Here are the data from Newark.
One thing I don't understand about the wastewaterscan data is that they don't always seem to subtype. My guess is that it's because the sample quality wasn't sufficiently high, so I tend to pay attention to the positives and take the negatives with a grain of salt.
One thing I don't understand about the wastewaterscan data is that they don't always seem to subtype. My guess is that it's because the sample quality wasn't sufficiently high, so I tend to pay attention to the positives and take the negatives with a grain of salt.
November 10, 2025 at 2:16 AM
7/ Here are the data from Newark.
One thing I don't understand about the wastewaterscan data is that they don't always seem to subtype. My guess is that it's because the sample quality wasn't sufficiently high, so I tend to pay attention to the positives and take the negatives with a grain of salt.
One thing I don't understand about the wastewaterscan data is that they don't always seem to subtype. My guess is that it's because the sample quality wasn't sufficiently high, so I tend to pay attention to the positives and take the negatives with a grain of salt.
6/ But....
We are seeing some concerning signs in Boston and Newark where flu activity in the wastewater has been steadily increasing for the past two weeks.
Unfortunately, we can't tell what variant/clade is causing the signal (although in Newark we know it's H3 and not H1, no data from Boston)
We are seeing some concerning signs in Boston and Newark where flu activity in the wastewater has been steadily increasing for the past two weeks.
Unfortunately, we can't tell what variant/clade is causing the signal (although in Newark we know it's H3 and not H1, no data from Boston)
November 10, 2025 at 2:16 AM
6/ But....
We are seeing some concerning signs in Boston and Newark where flu activity in the wastewater has been steadily increasing for the past two weeks.
Unfortunately, we can't tell what variant/clade is causing the signal (although in Newark we know it's H3 and not H1, no data from Boston)
We are seeing some concerning signs in Boston and Newark where flu activity in the wastewater has been steadily increasing for the past two weeks.
Unfortunately, we can't tell what variant/clade is causing the signal (although in Newark we know it's H3 and not H1, no data from Boston)
5/ We can see from the WastewaterScan dashboard the across-the-US, influence activity is quite low (although note higher levels in a sampling sites in a handful states).
Data from: www.wastewaterscan.org/en/about
Data from: www.wastewaterscan.org/en/about
November 10, 2025 at 2:16 AM
5/ We can see from the WastewaterScan dashboard the across-the-US, influence activity is quite low (although note higher levels in a sampling sites in a handful states).
Data from: www.wastewaterscan.org/en/about
Data from: www.wastewaterscan.org/en/about
4/ In the US, we're not yet seeing much flu activity.
How can I feel so confident given that the government is shutdown and we don't have CDC data?
Answer: wastewater surveillance and Walgreens testing data.
How can I feel so confident given that the government is shutdown and we don't have CDC data?
Answer: wastewater surveillance and Walgreens testing data.
November 10, 2025 at 2:16 AM
4/ In the US, we're not yet seeing much flu activity.
How can I feel so confident given that the government is shutdown and we don't have CDC data?
Answer: wastewater surveillance and Walgreens testing data.
How can I feel so confident given that the government is shutdown and we don't have CDC data?
Answer: wastewater surveillance and Walgreens testing data.
3/ As the BBC article above reported, the UK is seeing an earlier and dramatic rise in flu activity.
I haven't seen any genome sequencing data from the UK*, but the article suggests that the rise is due to this "K" variant/clade.
*I haven't looked, but I suspect there are plenty of sequences.
I haven't seen any genome sequencing data from the UK*, but the article suggests that the rise is due to this "K" variant/clade.
*I haven't looked, but I suspect there are plenty of sequences.
November 10, 2025 at 2:16 AM
3/ As the BBC article above reported, the UK is seeing an earlier and dramatic rise in flu activity.
I haven't seen any genome sequencing data from the UK*, but the article suggests that the rise is due to this "K" variant/clade.
*I haven't looked, but I suspect there are plenty of sequences.
I haven't seen any genome sequencing data from the UK*, but the article suggests that the rise is due to this "K" variant/clade.
*I haven't looked, but I suspect there are plenty of sequences.
2/ Sometime this summer, an H3N2 "variant" (normally we call them [sub]clades) started spreading.
Scientists are worried about this variant, now given the designation "K", because its genome suggests it may evade prior immunity and vaccine-derived immunity.
utppublishing.com/doi/abs/10.3...
Scientists are worried about this variant, now given the designation "K", because its genome suggests it may evade prior immunity and vaccine-derived immunity.
utppublishing.com/doi/abs/10.3...
Emergence of seasonal influenza A(H3N2) variants with immune escape potential warrants enhanced molecular and epidemiological surveillance for the 2025–2026 season | Journal of the Association of Medi...
Background: All of the major antigenic changes in influenza A(H3N2) viruses since 1968 have involved mutations at just nine amino acid positions, called cluster transition sites, surrounding the recep...
utppublishing.com
November 10, 2025 at 2:16 AM
2/ Sometime this summer, an H3N2 "variant" (normally we call them [sub]clades) started spreading.
Scientists are worried about this variant, now given the designation "K", because its genome suggests it may evade prior immunity and vaccine-derived immunity.
utppublishing.com/doi/abs/10.3...
Scientists are worried about this variant, now given the designation "K", because its genome suggests it may evade prior immunity and vaccine-derived immunity.
utppublishing.com/doi/abs/10.3...
