Tuesday, April 9, 2013

influenza genetical evolution basics


genetic evolution of influenza
(some of these things are being simplified for an easier understanding)

Influenza A has 8 segments with a total of 13300 nucleotides from {A,C,G,T}
groups of 3 nucleotides (64 combinations) form one of 21 amino acids,
so changes in nucleotides not necessarily change the amino acid.
Changes in amino acids ("nonsynonymous") are much more important for the
properties of the virus than changes in nucleotides that keep the same
amino acids ("synonymous").

In this simplified model each nucleotide changes (mutates)
with probability 0.0025 per year and 80% of mutations are synonymous.
So from the number X of nucleotide differences in 2 influenza-A viruses
with Y commonly available nucleotides in the alignment
you can calculate an estimate of the date of the most recent common ancestor
(MRCAD) by the formular 
where D_i is the isolation date of virus i  (for evolution over small periods
of ~<20 years, else use the formula:
The diversity in the 8 segments is different.
Segments 1,2,3,5,7 all go back to a common ancester
~200 years ago. Segment 4 is usually subdivided into 16
subtypes, segment 6 into 9 and segment 8 into 2.
(ignoring flu in bats). Let's assume for simplicity that
each of these subtypes in segments 4,6,8
has also a common ancestor ~200 years ago like.
And that the common ancestor of 2 different subtypes (16,9,or 2)
in the same segment is ~400 years ago.
The usual name of the subtypes is by the enumeration of
segment 4 (HA) and segment 6 (NA) while segment 8 is being
ignored, in mammals and poultry it's almost always the same
(81), while in wild birds 82 occurs with frequency ~20%
and 81 with ~80%
The nucleotide mutation-rate in segments 4 and 6 is ~50% higher and
the amino-acid mutation rate (nonsynonymous) is ~100% higher
as in the other segments.

In addition to the mutations at nucleotides there is the
possibility that entire segments are shuffled ("reassortment"),
when two different influenza-A viruses enter the same cell.
Of course you need one of each of the segments for a working
new virus, so there are 256 possibilities to form a new combination.
Some of these are more likely than others, e.g. segments 2,4,6
should "fit" to one another and are less likely separated
in the shuffling. Let's say ~half as likely.

The typical hosts for influenza A are birds,humans,swine,horses.
Let's consider birds and humans here.
The base reservoir are waterfowl : mallards,pintails,teals,
shorebirds. Not geese,domestic ducks,gulls,
(for simplicity called waterfowl below)

In these we see reassortments of different viruses more often than
in other species, i.e. switching of segments 4 and 6
is more common. In addition to this in the inner segments
(1,2,3,5,7,8) nonsynonymous mutations are rarer and in fact
at the amino-acid level mallardflu inner segments all cluster
around one index-strain (or avian consensus) from which they
usually only move away 5-10 amino-acid differences.
This is not the case in segments 4,6 nor in the inner segments
of poultry or mammalean flu, where they move steadily away from
the index with rate ~0.005 per year.
So, that's how you can usually see whether an inner segment
is mallardflu or poultry or mammalean.
There is one other distinction and that is the proportion
of A or T nucleotides, which grows after prolonged time of
evolution (decades) in mammals at rate ~

There is (currently) no separate evolution of influenza-A in poultry
or mammals. They occasionally catch a new flu-A virus from waterfowl
this evolves for a while outside waterfowl until a new strain
is introduced and co-circulates or replaces the old one.
Thus the oldest current segments outside waterfowl are
the human segments that go back to the pandemic 1918.
horse-H7N7 probably extinct since the 1980s
South America

While human influenza goes around the world each year,
avian and swine influenza usually stay in their
continental block (NA,SA,EuAsOzAf) for years and decades
without widespread occurrance in the other blocks.
For WFF that means that in the inner segments we
have NA and EA variants that differ much in the
nucleotides but few in anino acids

to do:


Nonhuman influenza


list of substrains
events of avian-->mammal,poultry

distances between subtypes , big 180*180 table

additions, improvement suggestions are welcome


next:  H7N9(2013)


No comments:

Post a Comment