Note: I will be updating this post through the day, as I play with the sequences in my spare time. (Right now I have to get the kids up and get them ready for school.) Don’t expect to learn anything particularly new from this, since I’m no influenza expert; but since I’m going to be doing this for my own curiosity I may as well save others the time.
Update: Also see my Wednesday post for more info.
For those playing along at home, some of the swine flu genomes are now available. Only one of them has a complete genome, and these are the California and Texas cases — now we really want to compare them to the Mexico strain to see if the mortality is associated with strain variation or not.
I haven’t looked at the sequences much yet. I did run some quick alignments of the HA gene; nothing unexpected. These are all the same strain, but not identical (as you expect from influenza). It’s significantly different from the HA of the H1N1 strain that was circulating this year (A/Brisbane/59/2007) (I think that was the circulating strain, it’s close anyway). If you’re interested you can check:
Alignment compared to A/Brisbane/59/2007
Alignment of the 7 swine flu HA proteins
Alignment of the 6 complete swine flu HA proteins
The closest matches to the HA in GenBank are from A/swine/Minnesota/1192/2001(H1N2) and several other swine flu strains, which are ~ 94% identical. I haven’t compared the other genomic fragments yet to see what other strains got into the mix.
Update 1. I’ve run the segments of the most complete new virus (A/California/04/2009(H1N1)) to see what viruses this looks most like. The only remarkable thing I see is how unremarkable it is. It’s pretty much a generic swine flu top to bottom, with close matches only to other strains of swine flu. Since those other strains didn’t jump into people, that’s just another indicator of how much we have to learn about molecular determinants of virulence in viruses.
Quick results, showing only the first few matches (there are many!):
| PB1 matches | ||
| Accession | Strain | Pct identity |
| AF342823.1 | A/Wisconsin/10/98 (H1N1) | 96% |
| AF250130.1 | A/Swine/Indiana/9K035/99 (H1N2) | 96% |
| CY033790.1 | A/mallard duck/South Dakota/Sg-00125/2007(H3N2) | 96% |
| PB2 matches | ||
| Accession | Strain | Pct identity |
| EU301177.2 | A/swine/Korea/JNS06/2004(H3N2) | 96% |
| AF455734.1 | A/Swine/Minnesota/55551/00 (H1N2) | 96% |
| CY033794.1 | A/mallard duck/South Dakota/Sg-00128/2007 | 96% |
| PA matches | ||
| Accession | Strain | Pct identity |
| AF455722.1 | A/Swine/Illinois/100084/01 (H1N2) | 96% |
| AF251433.1 | A/Swine/Minnesota/593/99 (H3N2) | 96% |
| F251425.1 | A/Swine/Iowa/569/99 (H3N2) | 96% |
| HA matches | ||
| Accession | Strain | Pct identity |
| AF455680.1 | A/Swine/Indiana/P12439/00 (H1N2) | 95% |
| AF250124.1 | A/Swine/Indiana/9K035/99 (H1N2) | 95% |
| AY038014.1 | A/Turkey/MO/24093/99(H1N2) | 95% |
| NP matches | ||
| Accession | Strain | Pct identity |
| AF251415.2 | A/Swine/Iowa/533/99 (H3N2) | 96% |
| EU798854.1 | A/swine/Korea/CY05/2007(H3N2) | 96% |
| EU798853.1 | A/swine/Korea/CY04/2007(H3N2) | 96% |
| NA matches | ||
| Accession | Strain | Pct identity |
| AF250366.2 | A/Swine/England/195852/92 (H1N1) | 94% |
| CY038009.1 | A/swine/England/WVL7/1992(H1N1) | 94% |
| Y038001.1 | A/swine/Spain/WVL6/1991(H1N1) | 94% |
| M matches | ||
| Accession | Strain | Pct identity |
| AY363575.1 | A/swine/Hong Kong/5212/99(H3N2) | 97% |
| AY363574.1 | A/swine/Hong Kong/5200/99(H3N2) | 97% |
| AY363573.1 | A/swine/Hong Kong/5190/99(H3N2) | 97% |
| NS1/NS2 matches | ||
| Accession | Strain | Pct identity |
| AF153262.1 | A/Swine/Minnesota/9088-2/98 (H3N2) | 96% |
| AF153261.1 | A/Swine/Texas/4199-2/98 (H3N2) | 96% |
| AF153263.1 | A/Swine/Iowa/8548-1/98 | 96% |
[...] or more proteins (illustrated). Each RNA segment has a name – PB2, PB1, PA, HA, NP, NA, MP, and NS. Mystery Rays has done a quick analysis of the sequences. The isolates are all the same strain, but they are not [...]
[...] Swine flu genome sequences [...]
Keep having fun. It seems those sequences are far more divergent than expected.
This is the latest (5/1) on promed:
See the first article on origins.
http://www.promedmail.org/pls/otn/f?p=2400:1001:2088121034750757::NO::F2400_P1001_BACK_PAGE,F2400_P1001_PUB_MAIL_ID:1000,77310
There’s nothing in the ProMed articles that says the sequences are “far more divergent than expected.” You’re probably looking at Donis’s comment that “it’s a lonely branch there. It doesn’t have any close relatives.” What he’s presumably referring to is the mixed Asian/North American origin (“It’s almost equidistant to swine viruses from the United States and Eurasia”) because the sequences are not in fact all that divergent. Each segment is a really very ordinary looking swine flu sequence. The segments are put together in a particular way — but that happens all the time. That’s what influenza does. It reassorts. This is a reassorted virus whose sequences are not particularly divergent.
so 6 segments were presumably evolving in
Mexican pigs since ~10 years, without surveillance and detection and now reassorted
in March (?) with an imported virus from Asia
which also had evolved separately without detection for long time
shouldn’t there be laws to forbid/restrict/control the mixing of such viruses ?
Test imported life pigs for flu
> There similarities of about 94 percent
> in the hemagglutinin [H] to the
> nearest strain we know.
> It’s almost equidistant to swine viruses
> from the United States and Eurasia
it’s classical swine HA (~1915), so how can it be similar to Eurasian viruses ?
Those must be from pigs imported from America
then
No evidence the 6 segments were evolving in Mexico. They look more like standard North American swine flu — Indiana, Ohio, etc viruses are among the best matches for the present sequences.
As for laws, what on Earth do you think you could do? Might as well pass a law making it illegal for wood to float. This isn’t manmade recombination, it’s natural.
On the other hand, this is clearly a wakeup call for more surveillance — swine influenza surveillance has been very scattered and desultory. But you have to remember — influenza, and swine influenza, are universal. Viruses that look like the parent strains of this virus would be everywhere, and would look perfectly ordinary on surveillance. For that matter, this H1N1 would itself look perfectly ordinary on surveillance. You’d have to ban all pig import and export (and much more than that) and it still likely wouldn’t work because some other species’ flu would hop over into pigs and cause an epidemic (remember that the H1N1 swine flus are human-derived!).
The biggest and scariest change in swine influenza that would have been flagged in surveillance was the rise of the triple recombinant in the 1990s — and that turned out to be essentially harmless to humans (unless there turns out to be a 15-year lag in the harm).
We don’t know enough about influenza in general, or swine influenza in particular. Finger-pointing, jumping to conclusions, and prejudice isn’t going to give us any answers.
in USA these 6 segments would have been more likely detected within those 11 years.
Not any Mexican swine sequences at genbank.
You could just demand to hold life imported swine separately from domestic ones
until they are tested and authorities give an OK.
Where are you getting 11 years from?
The Mexican H1N1 shows about 96% identity to its nearest neighbours. If you look at recent US swine flu strains you’ll find they’re often about 97% identical to their nearest neighbours, e.g. A/swine/OH/511445/2007(H1N1), especially if you don’t look at multiple isolates from the same outbreak. What’s more, the guys with the most matches tend to be the older strains for obvious reasons — there you look at their progeny as well as their parents. In other words, the sort of variation we’re seeing for the MExican strain is pretty much exactly what we’d expect to see in a random US outbreak — such as happen every year or two. Or are you arguing that A/swine/OH/511445/2007(H1N1) was hiding out in Mexico? Or is it simply that this sort of variance is fairly normal for swine influenzas?
we had the reassortment event from 1998 (or some years earlier ?) when this type of
virus emerged. The ancestor of modern
American Swine-flu.
Several other progenies from USA of this reassortment are at genbank.
But the new virus (in 123458) is more similar to the 1998 virus than to any of these, suggesting that it evolved separately
(in Mexico ?) since ~1998.
511445 is more similar to several other viruses than to 1998.
Where are you getting this? As far as I can see (see my latest post) it’s simply not true. What’s more, it’s trying to draw conclusions based on a tiny handful of changes, whereas we know the virus can make that many changes in a week or two and still be the same virus.
Hey,
Just wanted to say thanks for doin these alignments,this blog. There is a huge disconnect between major media fluff coverage and the bench-level emerging research. i am no expert in flu either. You have provided what i wanted/needed to know. Saved me the trouble of doing these line-ups myself.
[...] ??Evolution of Swine H3N2 Influenza Viruses in the United States?? ??Swine flu genome sequences [...]
see here for a list of differencs:
http://www.zonegrippeaviaire.com/showpost.php?p=35341
while swineflu sequences from the last years from Minnesota,Ohio,Illinois,Iowa are closer to each
other, the new strain is closest to
A/Wisconsin/10/1998
You can keep repeating that all day, but it still won’t make it any truer. Your link is impossible to interpret — what is it, some home-rolled program? It doesn’t seem to have any meaningful interpretation. As far as I can tell you’re ignoring the obvious fact that this H1N1 is a recombinant and trying to shoehorn everything into some preconceived bias, throwing away a huge amount of data the disagree with you and only accepting the rare pieces that support your theory — but I can’t even tell if that’s the case because you haven’t actually made any argument, you’ve just repeatedly made a claim as if that supports itself.
If anyone is interested in actually seeing data, I have put a (large!) table up here showing flu strains that are most similar to each of the segments of the present H1N1 strain. Also see my earlier posts “Has the new H1N1 been hiding for 11 years? (No.)” and “How fast can influenza change?” for more commentary.
Please note – I am perfectly open to the concept that this virus is most closely related to a 1998 strain and that it’s been hiding — but I need to see actual evidence that fits with biology and observation, not mere repeated assertions and incoherent numerology that’s apparently pulled out of thin air.
Yeah!