HA sequence from Niger available at GenBank
A/duck/Niger/914/2006(H5N1)
Sequence is easily distinguished from Nigeria H5N1 and has a number of polymorphisms not found in other Qinghai sequences, but readily detected in other H5N1’s from Asia.
Another advertisement for recombination.
So this means…. ? That this is yet another “type” or “strain” of H5N1 floating around? Last I understood, there were at least four different distinguishable strains.
(my apologies, I haven’t been keeping up on the strict science lately)
My take on what it means:
There are two lines of thought on how H5N1 spread from Qinghai: One group says it took the train—that is, (commercial) birds were smuggled across the border of china into tibet, then along the route of the Trans-siberia railway into the edge of europe, then across Turkey/Iraq by tramp steamer to northern Africa, where they were smuggled deeper into Africa. If this is to be true, then the sequences found in Africa would be pure Qinghai—that is, they would not show evidence of mixing with local, non H5N1 low-path influenza. They would be relatively pure and unchanged.
Another school of thought says that the principal vector is by migrating wild birds. If this is true, and because birds migrate often in a burst-pause pattern, that is, fly a while, rest up awhile, fly awhile, etc, they have the chance to mingle with local birds carring local avian influenzas. They swap a few infections, and the strains mingle, and now you have a Qinghqi base with small bits that match up with local, non H5N1 avian influenzas.
So which idea is supported by the new sequences? Migratory birds.
Why is this important? Because you only have enough money to fix the problem once, and you better pick the right problem to fix. If the problem is a smuggling problem, you stick your money in education and in border enforcement, and poultry inspections. If the problem is migratory waterfowl, you put the money in surveillance and make sure chicken ranches have good biosecurity and birds are kept away from migrating waterfowl.
And, it gives you information on the makeup of the strains should you need to know for vaccine purposes.
Ah, ok. That’s a great explanation, wetDirt! Well said.
wetDirt,
You are once again spot on with the analysis. This is definitely the Qinhai strain, but each has it’s own travel schedule and can be readily distinguished. If the H5N1 in Nigeria came via trade, then the H5N1 in adjacent Niger would match. If the arrival was by bird, then there would be several varieties.
This sequence points squarly at the migratory birds.
The constellation of shared polymorphisms is like a fingerprint.
This can be seen in the small branches on the tree of Qinghai sequences in Europe in the top and bottom on the tree (tree split for readability purposes).
wetDirt – at 13:01
Oh … my … goodness. Now even I can understand this bit of science. <grin> I even have a college education, but you guys normally leave me in the dust holding my aching head.
Seriously though, while I read most of the “high science” threads I usually only understand a particle of what is being said.
Thank you wetdirt. Putting it into a visual context really helped.
The Niger sequence has 6 polymorphisms that are present in a small subset of the Qinghai strain. The 6 polymorphisms are G253A, A524G, C1198T, G1240A, G1447A, T1603C.
Notice that all changes are transitions (replacing a purine with a purine A=G, or a pyrimadine with a pyrimadine C=T). Its really a binomial system.
These are all transitions and all have a clear H5N1 rap sheet.
These changes are NOT random mutations.
They were all acquired from kissin cousins.
Stamped passport of A524G (also in Nigeria isolate)
DQ447199 A/chicken/Egypt/960N3–004/2006 2006 H5N1
DQ406728 A/chicken/Nigeria/641/2006 2006 H5N1
DQ515984 A/Cygnus olor/Czech Republic/5170/2006 2006 H5N1
DQ435200 A/domestic cat/Iraq/820/2006 2006 H5N1
DQ435201 A/domestic goose/Iraq/812/2006 2006 H5N1
DQ659113 A/duck/Niger/914/2006 2006 H5N1
DQ464377 A/Egypt/2782-NAMRU3/2006 2006 H5N1
DQ435202 A/Iraq/207-NAMRU3/2006 2006 H5N1
DQ458992 A/mallard/Bavaria/1/2006 2006 H5N1
DQ449031 A/mallard/Italy/332/2006 2006 H5N1
AB233319 A/bar-headed goose/Mongolia/1/05 2005 H5N1
DQ320898 A/chicken/Guangxi/604/2005 2005 H5N1
DQ449632 A/chicken/Kurgan/05/2005 2005 H5N1
DQ323672 A/chicken/Kurgan/3/2005 2005 H5N1
DQ389158 A/Cygnus olor/Astrakhan/Ast05–2−1/2005 2005 H5N1
DQ434889 A/Cygnus olor/Astrakhan/Ast05–2−10/2005 2005 H5N1
DQ343502 A/Cygnus olor/Astrakhan/Ast05–2−2/2005 2005 H5N1
DQ358746 A/Cygnus olor/Astrakhan/Ast05–2−3/2005 2005 H5N1
DQ363918 A/Cygnus olor/Astrakhan/Ast05–2−4/2005 2005 H5N1
DQ365004 A/Cygnus olor/Astrakhan/Ast05–2−5/2005 2005 H5N1
DQ364996 A/Cygnus olor/Astrakhan/Ast05–2−6/2005 2005 H5N1
DQ363923 A/Cygnus olor/Astrakhan/Ast05–2−7/2005 2005 H5N1
DQ399540 A/Cygnus olor/Astrakhan/Ast05–2−8/2005 2005 H5N1
DQ399547 A/Cygnus olor/Astrakhan/Ast05–2−9/2005 2005 H5N1
DQ320899 A/duck/Guangxi/793/2005 2005 H5N1
DQ449640 A/duck/Kurgan/08/2005 2005 H5N1
DQ320897 A/quail/Guangxi/575/2005 2005 H5N1
DQ320137 A/swan/Astrakhan/1/2005 2005 H5N1
DQ407519 A/turkey/Turkey/1/2005 2005 H5N1
AB233320 A/whooper swan/Mongolia/3/05 2005 H5N1
AB233321 A/whooper swan/Mongolia/4/05 2005 H5N1
AB233322 A/whooper swan/Mongolia/6/05 2005 H5N1
ISDN140936 A/Whooping swan/Mongolia/244/2005 2005 H5N1
AY651346 A/Ck/Hong Kong/31.2/2002 2002 H5N1
AY651350 A/Ck/Hong Kong/3176.3/2002 2002 H5N1
AY651347 A/Ck/Hong Kong/37.4/2002 2002 H5N1
AY651345 A/Gf/Hong Kong/38/2002 2002 H5N1
AY575880 A/pheasant/Hong Kong/675.14/02 2002 H5N1
AY651348 A/SCk/Hong Kong/YU100/2002 2002 H5N1
M18450 A/Duck/Ireland/113/83 1983 H5N8
OK. My take on what this means:
There are two schools of thought about how changes arise in the H5N1 sequence. One school of thought is that these represent random mutations in a virus notorious for random mutations. When you see a change in the sequence, it happened by Cosmic Rays from Outer Space™, and will probably not be repeated elsewhere. Mutations, however, are not expected to be coordinated, that is, it would be wierd for two mutations a certain distance apart to happen the same way at the same time.
Another school of thought is that, while mutations do happen, the bad ones disappear fast, and the good ones are saved, like halloween candy, to be pulled out later. These little packages show up as a ‘chunk’, with specific changes, that remain stable with time, some of which stay geographically isolated, and some of which are widespread.
So what Niman has noticed is that all these changes have a certain pattern: The A purine is swapped for a G purine, or a C pyrimadine with a T pyramidine. Say the purines are bumps and the pyramidines are valleys. Now if you imagine the key-lock thing, what this means is that the little teeth on the key change shape only a tiny bit, because you are swapping one hill for another hill, and a valley for another valley. Similar shapes. And he is noting that it is a consistent pattern of transitions, it’s like the virus saying, hey, look, guys, cool, I can pick the lock with either version of this key, as long as I keep the same general shape.
And he is saying these are no longer random changes, that he sees this pattern over and over and it fits into a larger pattern.
C1198T shared by H5N1 but none are Qinghai sequences
ISDN138756 A/chicken/Malaysia/935/2006 2006 H5N1
DQ659113 A/duck/Niger/914/2006 2006 H5N1
DQ095624 A/Chicken/Yunnan/447/05 2005 H5N1
DQ095625 A/Chicken/Yunnan/493/05 2005 H5N1
AY651322 A/Dk/Indonesia/MS/2004 2004 H5N1
DQ017300 A/duck/Nakornsawan-2–02/2004 2004 H5N1
AY651323 A/Ck/Indonesia/2A/2003 2003 H5N1
AF509026 A/Chicken/Hong Kong/822.1/01 2001 H5N1
AF509028 A/Chicken/Hong Kong/830.2/01 2001 H5N1
AF509029 A/Chicken/Hong Kong/858.3/01 2001 H5N1
AF509030 A/Chicken/Hong Kong/867.1/01 2001 H5N1
AF509033 A/Chicken/Hong Kong/876.1/01 2001 H5N1
AF509031 A/Chicken/Hong Kong/879.1/01 2001 H5N1
AF509034 A/Chicken/Hong Kong/891.1/01 2001 H5N1
AF509035 A/Chicken/Hong Kong/893.2/01 2001 H5N1
AF509024 A/Chicken/Hong Kong/SF219/01 2001 H5N1
AF509017 A/Chicken/Hong Kong/YU562/01 2001 H5N1
AY221529 A/Chicken/Hong Kong/YU562/01 2001 H5N1
AF509018 A/Chicken/Hong Kong/YU563/01 2001 H5N1
AY221528 A/Chicken/Hong Kong/YU822.2/01 2001 H5N1
AY221527 A/Chicken/Hong Kong/YU822.2/01-MB 2001 H5N1
AF509023 A/Pigeon/Hong Kong/SF215/01 2001 H5N1
AF509022 A/Quail/Hong Kong/SF203/01 2001 H5N1
AF509021 A/Silky Chicken/Hong Kong/SF189/01 2001 H5N1
it would be wierd for two mutations a certain distance apart to happen the same way at the same time
I’m an ignorant looking for enlightment on this one.
Could it be that there are many many many mutations inside the cell, and only those that work get selected, so that it’s not that difficult to see the same wheel reinvented all over the place? I mean, maybe random mutation mechanisms produce a million varieties, and only a hundred are able to make a viable virion, so that most mutations are hidden from view (they never make it on their way out of the cell). With enough random mutations, you might have “twins” that have emerged at random from different branches of the tree: they look like they are relatives, but they are really reinventions of the same thing without family relationship.
I really don’t know.
Thanks.
“niman – at 13:28 The Niger sequence has 6 polymorphisms that are present in a small subset of the Qinghai strain. The 6 polymorphisms are G253A, A524G, C1198T, G1240A, G1447A, T1603C. “
So Henry, Taking any of these polymorphisms at random, how stable are they over time? Are there, say, 100 base pairs that have remained identical for over 5 years? I would like to calculate the probability of that happening using everyone’s quoted mutation rate. I need a nice even number of base pairs because I’m that lazy.
G1240A matches only one Qinghai sequence, but matches low pat from Europe and North America
DQ659113 A/duck/Niger/914/2006 2006 H5N1
AB233319 A/bar-headed goose/Mongolia/1/05 2005 H5N1
DQ309440 A/duck/BritishColumbia/CN26−6/05 2005 H5N2
DQ095629 A/Duck/Fujian/1734/05 2005 H5N1
DQ320897 A/quail/Guangxi/575/2005 2005 H5N1
DQ320893 A/chicken/Guangxi/2439/2004 2004 H5N1
DQ320883 A/duck/Guangxi/1311/2004 2004 H5N1
DQ320885 A/duck/Guangxi/1586/2004 2004 H5N1
DQ320886 A/duck/Guangxi/1681/2004 2004 H5N1
DQ320887 A/duck/Guangxi/1793/2004 2004 H5N1
DQ320890 A/duck/Guangxi/2291/2004 2004 H5N1
DQ320892 A/duck/Guangxi/2396/2004 2004 H5N1
DQ320881 A/goose/Guangxi/1097/2004 2004 H5N1
DQ320882 A/goose/Guangxi/1198/2004 2004 H5N1
DQ320888 A/goose/Guangxi/1832/2004 2004 H5N1
DQ320889 A/goose/Guangxi/2112/2004 2004 H5N1
DQ320891 A/goose/Guangxi/2383/2004 2004 H5N1
AY590577 A/openbill/Thailand/CU-2/2004 2004 H5N1
CY005927 A/shorebird/Delaware/101/2004 2004 H5N7
CY005969 A/shorebird/Delaware/75/2004 2004 H5N7
AY651336 A/Viet Nam/3062/2004 2004 H5N1
AY573917 A/chicken/Taiwan/1209/03 2003 H5N2
AY747609 A/swine/Fujian/1/2003 2003 H5N1
AY296079 A/duck/ME/151895–7A/02 2002 H5N2
AY296080 A/duck/NY/185502/02 2002 H5N2
AY296081 A/duck/NY/186875/02 2002 H5N2
AY296086 A/duck/NY/191255–59/02 2002 H5N8
AY296082 A/duck/NY/191255–79/02 2002 H5N2
AY995886 A/mallard/Sweden/22/02 2002 H5N9
AY995887 A/mallard/Sweden/27/02 2002 H5N9
AY995888 A/mallard/Sweden/28/02 2002 H5N9
AY995889 A/mallard/Sweden/31/02 2002 H5N2
AY995890 A/mallard/Sweden/37/02 2002 H5N9
AY995892 A/mallard/Sweden/40/02 2002 H5N6
AY995893 A/mallard/Sweden/49/02 2002 H5N9
AY995894 A/mallard/Sweden/52/02 2002 H5N9
AY995895 A/mallard/Sweden/58/02 2002 H5N3
AY995896 A/mallard/Sweden/64/02 2002 H5N2
AY995897 A/mallard/Sweden/79/02 2002 H5N9
AY995898 A/mallard/Sweden/80/02 2002 H5N9
AY296083 A/turkey/CA/D0208651-C/02 2002 H5N2
AY075033 A/Duck/Hong Kong/380.5/2001 2001 H5N1
AY296077 A/duck/NJ/117228–7/01 2001 H5N2
AY747617 A/swine/Fujian/F1/2001 2001 H5N1
AY296076 A/unknown/NY/101250–18/01 2001 H5N2
AY296075 A/unknown/NY/9899–6/01 2001 H5N2
AY296073 A/chukkar/NY/51375/00 2000 H5N2
AY296072 A/duck/NY/44018–2/00 2000 H5N2
AY585377 A/duck/Zhejiang/52/2000 2000 H5N1
AJ305306 A/chicken/Italy/8/98 1998 H5N2
AF194169 A/Chicken/Italy/312/97 1997 H5N2
AF194990 A/Chicken/Italy/367/97 1997 H5N2
AF194991 A/Guinea Fowl/Italy/330/97 1997 H5N2
U79448 A/Chicken/Queretaro/7653–20/951659 1995 H5N2
CY006040 A/chicken/Hidalgo/28159–232/1994 1994 H5N2
U28920 A/Emu/Texas/39442/93 (HP progeny) 1993 H5N2
U28919 A/Emu/Texas/39442/93 (non-HP parent) 1993 H5N2
U67783 A/Mallard/Ohio/556/1987 1987 H5N9
U69277 A/Mallard/Wisconsin/944/82 1982 H5
AF082040 A/Duck/Minnesota/1525/81 1981 H5N1
U79454 A/Turkey/Minnesota/3689–1551/81 1981 H5N2
CY005918 A/mallard duck/Alberta/57/1976 1976 H5N2
U79452 A/Mallard/Wisconsin/169/75 1975 H5N3
U79456 A/Turkey/Wisconsin/68 1968 H5N9
AB241613 A/turkey/Ontario/7732/66 1966 H5
M30122 A/turkey/Ontario/7732/66 1966 H5N9
U20460 A/tern/South Africa/61 1961 H5N3
wetDirt,
There is not much of a database on the Qinghai strain (it was just identifed a year ago) but for this HA sequence there are over 1600 bases and only 6 are not common (and all 6 can be easily found in other H5N1s).
Are we talking about an exact match, that is , 100%? If so, then your last entry, a tern from 1961, says that the this polymorphism has remined unchanged for, umm, let’s see, 45 years, hanging around in low-path land.
Some polymorphisms go back to 1918, but that is just a match at that position. The earlier eact matches I talked about are long stretches that including a long string of polymorphisms.
G1447A Again no match with other Qinghai strains, but easily found in other H5N1s
DQ659113 A/duck/Niger/914/2006 2006 H5N1
DQ320876 A/chicken/Fujian/1042/2005 2005 H5N1
DQ320875 A/duck/Fujian/897/2005 2005 H5N1
DQ320900 A/duck/Guangxi/951/2005 2005 H5N1
DQ320901 A/duck/Guangzhou/20/2005 2005 H5N1
DQ320911 A/duck/Hunan/1265/2005 2005 H5N1
DQ320912 A/duck/Hunan/1608/2005 2005 H5N1
DQ320913 A/duck/Hunan/1652/2005 2005 H5N1
DQ320939 A/duck/Vietnam/568/2005 2005 H5N1
DQ320896 A/goose/Guangxi/345/2005 2005 H5N1
DQ095628 A/Goose/Shantou/1621/05 2005 H5N1
DQ320915 A/goose/Shantou/2216/2005 2005 H5N1
DQ320916 A/migratory duck/Jiangxi/1653/2005 2005 H5N1
DQ320917 A/migratory duck/Jiangxi/1657/2005 2005 H5N1
DQ320918 A/migratory duck/Jiangxi/1701/2005 2005 H5N1
ISDN117777 A/Viet Nam/JP4207/2005 2005 H5N1
DQ320894 A/chicken/Guangxi/2448/2004 2004 H5N1
DQ320895 A/chicken/Guangxi/2461/2004 2004 H5N1
ISDN48972 A/Chicken/Hu bei/14/2004 2004 H5N1
AY684706 A/chicken/Hubei/327/2004 2004 H5N1
AY770079 A/chicken/Hubei/489/2004 2004 H5N1
AY830774 A/chicken/Macheng/2004 2004 H5N1
DQ083567 A/Ostrich/Samut Prakan/Thailand/CU-19/04 2004 H5N1
DQ083574 A/ostrich/Samut Prakan/Thailand/CU-31/04 2004 H5N1
AY651362 A/peregrine falcon/Hong Kong/D0028/2004 2004 H5N1
AY950236 A/swan/Guangxi/307/2004 2004 H5N1
AY651354 A/Ck/Hong Kong/NT93/2003 2003 H5N1
AY651356 A/Ck/Hong Kong/SSP141/2003 2003 H5N1
AY651355 A/Ck/Hong Kong/WF157/2003 2003 H5N1
T1603 matches the low path ducks from Japan and a series of Qinghai isolates from Astrakhan but no match with Nigeria.
G253A has too many matches to list (several hundred H5N1’s) but no sharing wth any other Qinghai strains.
DQ659113 A/duck/Niger/914/2006 2006 H5N1
DQ389158 A/Cygnus olor/Astrakhan/Ast05–2−1/2005 2005 H5N1
DQ434889 A/Cygnus olor/Astrakhan/Ast05–2−10/2005 2005 H5N1
DQ343502 A/Cygnus olor/Astrakhan/Ast05–2−2/2005 2005 H5N1
DQ358746 A/Cygnus olor/Astrakhan/Ast05–2−3/2005 2005 H5N1
DQ363918 A/Cygnus olor/Astrakhan/Ast05–2−4/2005 2005 H5N1
DQ365004 A/Cygnus olor/Astrakhan/Ast05–2−5/2005 2005 H5N1
DQ364996 A/Cygnus olor/Astrakhan/Ast05–2−6/2005 2005 H5N1
DQ363923 A/Cygnus olor/Astrakhan/Ast05–2−7/2005 2005 H5N1
DQ399540 A/Cygnus olor/Astrakhan/Ast05–2−8/2005 2005 H5N1
DQ399547 A/Cygnus olor/Astrakhan/Ast05–2−9/2005 2005 H5N1
DQ449640 A/duck/Kurgan/08/2005 2005 H5N1
DQ320137 A/swan/Astrakhan/1/2005 2005 H5N1
AB241624 A/duck/Hokkaido/101/04 2004 H5N3
AB241625 A/duck/Hokkaido/193/04 2004 H5N3
AB241626 A/duck/Hokkaido/299/04 2004 H5N3
Here is earlier commentary of the Nigerian sequence. It also had 6 “informative” polymorphisms, but they all matched other Qinghai isolates - 4 with Kurgan and 2 with Mongolia.
Dr Niman,
Your argument is based on the idea that these polymorphisms are not random, and that they are persistent across time and space. It seems to me that your data on regions of identity has gotten a lot of notice (admittedly, some tin-hat notice) because it directly supports your case that random mutations are rare, not common, in currently circulating strains of influenza A, both low and high path. Perhaps you could dig out some other examples of identical sequences, perhaps encompassing a few polymorphims, in low-path as well as hi-path strains. A region of identity in a low-path north American strain that persists for five or more years would be a powerful argument against the tin-foil-hatters, and strenghthen your argument that these regions are highly conserved.
And also. If you get Science, what do you think of the article from mid-February, vol 311 page 960, using the argument that recombination is an error-correcting mechanism. I though immediately of you when I read it.
The recent sequence from the Ivory Coast represents yet another introduction of H5N1 in western Africa (and more recombination).
The filed PCT has some comments on cached sequences correcting errors.
yer a tricky devil mr. DemFromCt :-)
spam is spam and must be assimilated.
Closed to maintain Forum speed.