Mystery Rays from Outer Space

Here are the results of a quick BLAST lineup of some of the segments from the new H1N1 influenza viruses.

gsgs said in comments that “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.”

But as far as I can see that’s simply not true.  Homology lineups — even just looking at H1N1 strains — pull out recent viruses long before the 1998. The closest H1N1 I find matching the HA is A/swine/OH/511445/2007(H1N1); for polymerase (which is probably better for lineups) it’s A/Iowa/CEID23/2005, just three years before the new H1N1. Only a NP search finds a 98 H1N1 as a best match, and that’s only a fraction of a percent better match than a 2005 Korean strain — ten more residues out of 1500.

What’s more, it’s wrong-headed.  The differences in homology between different viruses here represent only a tiny handful of nucleic acid changes.  And that kind of difference represents a week or two in the replication of a single virus.  Look at the HA from isolates from the current outbreak!  There are half a dozen changes between two different California isolates!  That’s just about the difference that makes gsgs say the parent must be a 98 virus and not a 2005 virus, and yet that’s the same virus. There are a bunch of H3N2 viruses that are closer than any H1N1– are these therefore the parental viruses?

I’m sorry, it’s craziness to look at these kinds of changes and make the claim that it’s somehow informative of parental lines.  The virus changes too fast to put any parent/progeny relationship on a handful of residues, because that takes like a week in the virus cycle.