In blog posts and comments I’ve seen a few people arguing that the Mexico H1N1 influenza virus must have been around for a while — 10 years? 11 years? — because it doesn’t have any really close neighbours based on sequence. I’m not sure where this is coming from, but the precision of the “11 years” seems strange. Has anyone knowledgeable actually said this, or is it based on guesswork?
In any case, the notion that this H1N1 has been circulating for a decade has struck me as a great underestimate of influenza variation. This virus is more or less 96% identical — depending on which segment you choose — to other ciruclating North American or Eurasian swine flu viruses. That seemed like a really close agreement to me and seemed like the sort of thing that could happen really fast.
But I’m not a flu guy. So I looked for some way to compare. What I wanted was a series of influenza sequences with known dates and phylogeny — that is, preferably from a single outbreak or series of outbreaks with a more or less common cause. I found that in this paper1 from Jian et al. looking at influenza outbreaks in Taiwan in the 2003-2006. I focused on the Influenza A, H1N1 type — and this represented a single flu season.
Looking at the 169 amino acid region they sequenced:
The average numbers of differences at antigenic and nonantigenic sites of H1 in 2005 and 2006 were 0.90 and 2.34, respectively. … Moreover, we have observed that the overall average amino acid difference has steadily increased since February 2006. The average differences at antigenic and nonantigenic sites in 2005 were 0.39 and 0.67, respectively, while in the first 7 months of 2006 they were 0.98 and 2.93, respectively. 1
I’m having a little trouble exactly mapping the sequences they deposted in GenBank with the order they were isolated in and so on. But the bottom line is that it looks as if in a single flu season, this human strain showed up to 5% divergence at the nucleic acid level.
That’s pretty much the same divergence as we’re seeing between separate outbreaks here — that is, between known North American swine flu viruses and the new Mexican H1N1.
In other words, I believe the divergence we’re seeing between the Mexican strain and known strains is pretty much what you’d expect to see in any influenza outbreak.
- Jian, J., Chen, G., Lai, C., Hsu, L., Chen, P., Kuo, S., Wu, H., & Shih, S. (2008). Genetic and Epidemiological Analysis of Influenza Virus Epidemics in Taiwan during 2003 to 2006 Journal of Clinical Microbiology, 46 (4), 1426-1434 DOI: 10.1128/JCM.01560-07[↩][↩]
Cool. We’ll have to take a look at the sequences :-)
11 years because it closely matches a version of swine flu that appears among pigs in 1998 and apparently has become dominant in piggeries in the US ever since.
Source? God know how many articles off researchblogging.org
:)
11 years because it closely matches a version of swine flu that appears among pigs in 1998 and apparently has become dominant in piggeries in the US ever since.
That makes sense, and it’s almost certainly where the meme comes from. But I’ve seen people interpreting this exactly opposite – that the new Mexican H1N1 has been hiding somewhere for 11 years (implying dark secrets held by governments, fill in your conspiracy blank) in its present form but unknown to science. According to your interpretation, the parent virus has been around and studied for 11 years, and we don’t know how long the new virus has been around but it’s very likely a very short time. That fits with the observations. The conspiracy/incompetence theories don’t.
ahh, well. Sorry. Not 11 years hidden. The best matches are from the 1998 outbreak
where this strain first appeared and reassorted, but when recent virus have same distance
as one of them to a direct 1998 ancestor, then their estimated most recent common anchestor
is only 5.5years back, since the accumulated differences add up in both directions.
So, what’s the latest calculated most recent common ancestor date ?
I get 2002 for A/sw/Korea/CY04/2007(H3N2) in PB1 , still 7 years which is unusual.
Other viruses usually have closer relatives in genbank.
Looking at changes and assuming that the new H1N1 is going to drift/adapt at the same rate as previous viruses is not a safe assumption. This virus has clearly jumped species fairly recently. We know that viruses undergo rapid, intense selection when they jump species. So we would expect this new virus to show significant changes from other viruses, that were already adapted to their host and didn’t undergo the same selection pressure. We can’t tell which changes are recent and adaptive, and which are representative of long-term drift.
If you’re trying to use sequences to look at time of change, you really need the original swine version of the virus, not a version that’s cycled through humans for an unknown period. Unfortunately, for this virus we still don’t have any swine-adapted sequences.
yes, it often evolves faster after species switches, i.e. in 1918, but how much ?
I expect 40-50 mutations in the genome
per year.
Human H3N2 acquires ~35 mutations per year.
Sometimes(rarely) we have very few mutations
over years in Swine also recombinations,
I’m not sure about the reasons.
This was i.e. observed in Korean swine, so I’m a bit careful about comparing the Korean sequences with the actual ones
[...] i am on record as being against the flu. i have an attitude problem with segmented RNA genomes (40-50 mutations yearly?!), i’ll admit. i do not like their unpredictability, sudden virulence appearing as if out of [...]