Rhiza Labs FluTracker Forum

amino-acid differences apparently do increase 1975-2005. Why?
(I suppose you are using the >A/Index/birds on page 3 as index.)

- Evolutionary pressures changed during this period. OR
- Index choice needs correction to minimize distances, taking on account 1975-2005 changes.

The price of index change is to increase distance to other isolates, including your
your best central virus (A/Dk/Altai/1285/1991/08/15(H5N3)). So I think it would be premature to modify the index.

Maybe the use of averages on 1975-2005 masks important evolutionary details. Perhaps the index still holds for many contemporary strains, and this increasing divergence is concentrated on a small number of strains. I feel it’s better hold present index, even if it doesn’t fit to all avian influenza strains and all avian hosts!

I think it doesn't really increase or increase much.
It depends on which species and locations are being sampled
preferrably in some period.
Even with the reduced evolution of 2^10^-3 nucleotides per nucleotide-position per year
and 3*10^-4 amino acids per amino-acid position per year
(see post 13723
posting.php?mode=quote&f=26&p=13723
(BTW. enumerate posts in a thread as v-bulletin does it ! Makes referrence and linking easier) )
it still should drive away from the index with 0.6 amino acids per year in each of the 3 big genes
(PB2,PB1,PA) - 3 changes per 5 year, ~10-15 in 30 years


Looks like flu is essentially the same as it was e.g. 1000 years ago.
As for the amino-acids in inner genes of avian-flu, that is.
The encoding of these by nucleotides may change, though.
Only smaller changes of ~30 amino acids per segment do/did occur. Due to
differences in species, habits, environment, immunity escapes...

Of course, then the birds still have 16 entirely different HAs and 9 NAs, which reassort a lot
thus increasing avian virus diversity.
But even then, it seems that not every variant reassorts with every other, else the
Altai-duck would not match the index so well in all segments.


These observations may help us to predict which reassortments are possible,
what can be expected from future birds, how many other H5N1-like events are still lurking
in the virus genome...

_________________
no patents on genes, publish the GISAID sequences !

there was concern that mexflu or swine H1 might go into
the bird population.
There were infections at turkey farms with swinish viruses
and some sequences exist at genbank with swine viruses
in turkeys. Some South Dakota pintails and mallards
even had that triple-reassortant swine virus,
the ancestor of mexflu. And it was strangely preserved
for 9 years, no nucleotide mutations.
Then in birds it could reassort with other bird viruses
and spread and improve and evolve and go back to
swine and humans.

But with this virus being driven away from the avian index
since decades by mammalean evolution, can it be successful
in birds ? Nature didn't favour these viruses in birds
distant from the index


amino-acid differences avian index - mexflu:
17,23,17,28,7 differences in PB2,PB1,PA,NP,M1

other bird viruses distant from the index:
A/stint/Australia: 15,6,19,30,7
A/Ck/Henan/26: 19,18,29,16,6

_________________
no patents on genes, publish the GISAID sequences !

On "Slopes" .-Before I resume the increased mutation rate discussion, here is a deserved explanation of my slopes and clouds concepts:

The primary idea of seeking the slope of a straight line assumed amino acid changes to be proportional to nucleotide changes.
It works to the lower “cluster” (“cloud”) with an extremity at the origin (0,0) and a rough linear fit y=ax+b .

To this “cloud”, we get the slope: a=(100/18)=5.6 [Nuc/AminAc], offset b=0. The coefficient estimates the Nucleotide/Amino-Acid ratio.

Interpretation is less clear to other clusters. Some are so clear that I feel they worth at least a qualitative analysis.
The clearer patterns are clusters around an approximate straight line. However, the “offset” b is not zero! Why!?
(I will add comment on other clusters in a separate post.)

>there was concern that mexflu or swine H1 might go into
>the bird population.
>There were infections at turkey farms with swinish viruses
>and some sequences exist at genbank with swine viruses
>in turkeys.

A human influenza pandemic is an uncommon event on influenza ecology.
Pandemic can also have a very limited relevance tho avian influenza.
Trouble is: WE are the hosts! It is legitimate to be a bit … anthropocentric in this case.
So turkeys H1N1 infection may or may not be important to avian influenza as a whole.
It could be, however, relevant to US!

>Some South Dakota pintails and mallards
>even had that triple-reassortant swine virus,
>the ancestor of mexflu. And it was strangely preserved
>for 9 years, no nucleotide mutations.
>Then in birds it could reassort with other bird viruses
>and spread and improve and evolve and go back to
>swine and humans.

It is not clear to me how important birds reassortments can be on pandemic dynamics.
Apparently swine reservoir does and it was already included on a recent epidemiological modeling of H1N1 pandemics.
Modeling influenza epidemics and pandemics: insights into the future of swine flu (H1N1)
http://www.biomedcentral.com/1741-7015/7/30

Anyway, the evolution times involved are comparable to pandemic duration, 9 months – 5 years.
So avian reassortments role, if any, is a nonstandard avian evolution pattern.

for segment 1 the top cloud was made by (American,Eurasian) pairs.

The 2nd cloud is largely made by (Australian,Eurasian) pairs,
but also (early Eurasian,Eurasian) pairs and others
which are not so easy to specify.

The other segments are probably different !

here is a new picture:

http://magictour.free.fr/panflu/sepp1c0b.JPG

_________________
no patents on genes, publish the GISAID sequences !

human and swine viruses rarely reassort with segments from entirely
different viruses. When they do, it's often a dramatic event.

Birds reassort all the time their HAs and NAs.
16 strains of HA, 9 of NA. Compare that with the 2 or 3 in humans.
Once it's in birds, it has access to a big pool of different
HAs and NAs, including H5N1.

That feared reassortment with H5N1 is IMO more likely to happen
in birds - if they are able to get this mexflu circulating.
The South Dakota mallards are a first warning sign.

However, now we know that such viruses in birds
violate the "close to the index"-rule and thus may not spread
in wild birds

_________________
no patents on genes, publish the GISAID sequences !

>The South Dakota mallards are a first warning sign.
Why?
>However, now we know that such viruses in birds
>violate the "close to the index"-rule and thus may not spread
>in wild birds
This is likely better to wild birds health. Not sure about humans in short term!

>for segment 1 the top cloud was made by (American,Eurasian) pairs.

>The 2nd cloud is largely made by (Australian,Eurasian) pairs,
>but also (early Eurasian,Eurasian) pairs and others
>which are not so easy to specify.

Whe have to bear in mind that each dot is represents a pair of isolates.
The top cloud contains the largest Nuc/Am ratios.
Your previous example with 350 nucleotide differences and just 4 amino acid differences is there,
close to the left border of the upper cloud N/A=87. [A/pintail/Alaska/779/2005/08/20(H3N8) X A/duck/Altai/1285/1991/08/15(H5N3)]

Basic straight line fit:
b=340; a=(390-340)/37=50/37=1.35, and the fitted linear equation reads:

N=340+1.35A (N=Nucleotide differences; A=Amino acid differences)

This is the “easy” part of the job. Explanation of such a behavior harder step.
Wathever the ultimate (biological) explanation could be, the cloud is puzzling by itself.
It represents a large collection of HA sequence pairs sharing the common feature of this odd offset (>340 nucleotides difference)
AND a small slope (Delta_N/Delta_A). This appears solid.

I am not that sure on what individual clouds actually represent. Are they really sampling-dependent (continents, hosts variety)
OR they share some subtle common trait (eg. Different classes of evolution patterns).

About the new pictures:
Picture “Eurasian Lineage” shows sharper details on intermediate clouds.
The upper cloud realy disappears without American lineages (or it is just not shown in the picture)? (EA is HA, OK?)
Picture “Without Australia” appears to attenuate the “second (top to bottom) cloud” density, but the shape is quite preserved.

>> The South Dakota mallards are a first warning sign.
> Why?

first example of swinish flu in wild birds.
And close to mexflu in 5 segments.
And preserved since 1998

>> However, now we know that such viruses in birds
>> violate the "close to the index"-rule and thus may not spread
>> in wild birds
> This is likely better to wild birds health. Not sure about humans in
> short term!

the next pandemic will likely come and likely from birds

_________________
no patents on genes, publish the GISAID sequences !