Forum: Accuracy of Flu Polymerase

I start this thread to continue the discussion of the accuracy of the flu polymerase. I have not had time to properly analyze all of the available H5N1 sequences. The identity relationships that Dr. Niman points out are accurate. That does *not* prove his point. He does not explain the mutation rate in in vitro experiments. He has cherry-picked a few examples out of thousands.

It will take me some time to read all the comments on the various threads related to this issue, so apologies if I do not respond for a while. I have other responsibilities right now which are occupying most of my time.

Note, I am traveling for the next 10 days and have very limited access to the internet. “Quietness” on my part should not be construed as agreement with Dr. Niman or anyone else.

Whether the examples are cherry-picked or not, they still call for some pretty fancy explanating. But hey, if the virus wants to remain as it is, the worst part of that is we all need new hobbies, right?

‘splainin, explanatizing, whatever.

I would hope this would not devolve into an altercation (left for your friendly neighborhood mods to clean up).

I’m just a practical soul. ;-)

Now, Dem…we would Never be uncivil to one another. That would be unseemly. :)

You know what they say, Dem. Practice makes perfect.

Monotreme (or anyone),

Can you summarize EXACTLY what is the issue here? I’m going to see Taubenberger again tomorrow, so I might ask him.

Dejure practice does make perfect as we have daily proof on the Wicki. Interesting term “cherry picking” the doc I worked for called it coincidence. Proof disputes coincidence especially if you’re examining thousands of pieces of data. Why aren’t there hundreds of examples out of thousands if there is such accuracy as described?

The Bottom Line:

H5N1 is driving the bus…and it doesn’t care one way or another.

Tom DVM – at 23:13

That’s true. It is, however, always helpful to try and understand what we see. We can do that AND not get hung up on the solution.

Tom DVM at 23:13. Good point. But what I’d like to know at this point is, are we dealing with a school bus, a Greyhound, or that crazy bus from the movie, Speed?

anon_22, here is the issue as I understand it. Almost every scientist, except Dr. Niman, believes that flu viruses have a sloppy polymerase and hence a relatively high mutation rate. The mutation rate has actually been measured in vitro and its high. Dr. Niman, OTOH, thinks that flu viruses have extremely accurate polymerases. The reason he thinks this is that he has found examples of relatively long stretches of perfect homology from flu isolates obtained in different years. I admit I don’t know how to explain these. However, his best example was recently erased from GenBank by the submitters. Still, he has a few others including a match from 1977 and 2003, I think.

I think it would be great if you could ask Dr. Taubenberger what he thinks about this issue.

Anon 22:

The dispute was whole gene sequences being removed from Genebank that were depsoited in Feb but made public in May. There is a procedure for removal so if the data comes back you can track it and what exactly was changed, and that procedure wasn’t followed (or looked that way last night). The reason at the site for removal was inablility to confirm the data. The data was in the bank four months and in public domain for around six weeks. To get the full idea you should read the thread. Our concern was this is the data that is being used for future vaccine development and it was in error for 4 months?

DeJure. It’s driving a bus with a neutron bomb for a hood ornament.

DemFromCt. These are important issues but only for us humans.

It seems we have been playing two parallel games for the past few years. The first is the one we think we’re playing, where the humans are way smarter than the bug…and way smarter than the humans were in 1918. We have all this technology and our advanced healthcare system and our magic technological pills. We have made prediction after prediction and as time has gone on, we have discovered we know less and less and now have reached a point where we really don’t know much of anything about viral evolution and adaption, virulence and transmissibility.

Game # 2 is the one the virus is playing. We have a few days of no new reports of illness and all of a sudden maybe we were over-reacting and H5N1 is going away etc. We find ourselves over-reacting first on one end and then on the other…

…and the virus keeps on its determined and continual evolution, mostly beyond our comprehension and our five senses.

I believe our greatest failing is even after a nine year time period and the score is nine to nothing for H5N1, we still deep in our hearts, feel that this is not going to happen…and that mislaced optimism will prove to be man’s greatest folly!!

OK. Here’s my crack at it, please anyone, jump right in.

Conventional wisdom is that AI, like most RNA viruses, is a sloppy replicator. Random mutations creep into viral progeny, which drives the dispersion of different strains of the virus. THese strains inevitably drift apart, mostly at random. A mutation rate has been proposed, approximately 1 mutation per replication cycle per virus. If a replication cycle is a day, then there should be 365 new mutations a year, scattered at random. Over a period of ten years, it is highly unlikely that a string of 1000 bases could be preserved intact, with not a single base out of place.

Niman has combed through the available sequences, and has produced sequences a thousand bases long that do not have a single error. Not just once, but over and over, in birds and swine (the human sequences are locked up). He believes it is a real thing that the virus can do, and that it proves that mutations are not random, and that this is explainable by homologous recombination, and conservation of the really good bits of code. It’s not clear how to bulletproof the code, though.

This would imply some way for the RNA to detect and correct errors, or the polymerases don’t make as many errors in the wild as they do in the lab when people are watching, or the genome is so tightly wound that almost any error is lethal, and so is rapidly weeded out. Recombination would allow the viruse to reassemble itself out of the best bits of code it could find. Since most of the mutation rate experiements have been done with single strains, it is not clear to me that this method is precluded.

Monotreme has proposed that these regions of identity, derived from H1N1, are relatively recent lab escapes. Niman says no, they go back to 1933 or so, and can demonstrate that they show up in H1N5, H9N3, and I think H5N2, I’d have to rummage out the quote.

Is there evidence in the sequestered sequences that there have been repeated lab escapes, either from China, the old Soviet Union, or our lily-white hands? There is a vague correspondence in dates to alleged times of escape.

OK, what did I miss or misrepresent?

Tom DVM: I am asking this question just of you, and I don’t want to divert this thread from its main purpose. And also it would be ridiculous to start another one of “these” threads. But briefly, tell me as of this moment, “Will we even get to October?”.

Tom DVM – at 23:35

Who is we, brother? ;-)

Hop over to the thank you thread and help out. Or the Masks thread, or any one of the myriad activities that happen simultaneously.

DemFromCt. These are important issues but only for us humans.

When Ii stop acting human, please say something. ;-)

Tom DVM at 23:35: You mention “man’s greatest folly”… reminds me of lyrics from a song: “History shows again and again how nature points out the folly of man…Godzilla!” I suppose H5N1 is nature’s version of Godzilla….

Medical Maven. I honestly believe the writing is on the wall…but I am also a human who feels deep in my heart that a pandemic is still not going to happen…or at least not going to happen until I’m dead and gone (haven’t started prepping yet).

The answer to your question I think is in the history with H5N1 over the last nine years. I believe it has followed a pattern of appearing to disappear in the summer-time to return in the fall. That’s not to say that it won’t turn left when we think it is turning right, but I think we are safe until the end of September.

Having said that, its smouldering is more obvious now then it was last summer after Quingyi Lake when there was pretty much silence until the middle of October…hard to believe how far we have come or slipped in the past twelve months!!

If that is not clear enough ask another question.

wetDirt – at 23:36

Good summary. The mechanism for ‘bulletproofing’ the sequences (so they don’t change) is not one I’ve seen proposed (niman proposes how the rest of it changes, not how it stays stable), unless I’ve missed it. But in my limited understanding, this would represent a radical departure from conventional wisdom, no?

Oh, and the fact that the sequences that were listed in GenBank have been pulled is a separate side-issue, maybe coincidence, so Monotreme started a completely separate thread for that so as not to distract from this coversation.

Tom DVM: Your conscious thoughts have been running through my subconscious. Thanks.

WetDirt:

Please show more respect For RNA virus—HIV is a retrovirus which is genetic info in a retovirus particle encoded by RNA. Need I remind you no vaccine in 25 years? Yeah—I know we’re talking influenza but how can you call it sloppy?

The error-check-correct is mine, I’m afraid. THe way DNA creatures check errors is to two strands, one being used and one being inactivated. During reproduction, the two strands unzip, and are rebuilt from parts using one or the other as a template. But DNA had to have learned that trick somewhere. When there are two RNAs in the same cell, the situation is pretty much like the DNA situation with two separate strands. Has anyone ever looked at what really happens during dual RNA infections? The silly thing is, after all, hijacking the cell’s DNA machinery. And DNA has orders-of-magnitude better error checking.

My 2 cents.

btw, this is worth repeating, as it represents conventional wisdom:

 looks like we may not have to wait till winter

 H5N1 virus may have mutated

By Teddy Ng and Chen Hong (China Daily) Updated: 2006–06–17 09:01

The latest human bird flu infection on the Chinese mainland is worrying as it shows the H5N1 virus may have mutated and become as infectious in warm months as in cooler ones, Hong Kong’s health chief said on Friday.

The virus thrives in lower temperatures and is usually most infectious in the cooler months between October and March.

But confirmation on Thursday that a 31-year-old truck driver in the southern city of Shenzhen has been infected has caused uneasiness.

“Is this because the virus has changed, so that it is highly infectious all year round? Or, if it is happening in summer, winter would be even worse?” said the Secretary for Health, Welfare and Food, York Chow.

He said the virus might have become “more virulent and spread wider than we’ve expected,” though its mutation was not confirmed.

“If that is the case, the risk for humans to be infected in the future is higher,” he added.

The truck driver was admitted to hospital and was critically ill on Friday. He had visited a market where live poultry was sold and eaten chicken before he fell ill. But he is not known to have had any other close contact with poultry.

University of Hong Kong microbiology head Yuen Kwok-yung said the Shenzhen case was abnormal and worried the disease would spread in winter.

“If there are human infections from June to August, it means the virus is extremely active. I am worried that a major outbreak will happen in winter,” he said.

In neighbouring Shenzhen, authorities have stepped up virus prevention and surveillance efforts.

The local government said it will now report the situation relating to human bird flu cases every day.

The Shenzhen Centre for Disease Control and Prevention has been asked to enhance its surveillance of any pneumonia-like cases.

So far the city has not reported any poultry infections.

But vendors said they are required to disinfect shelves twice a day and stop on-the-spot slaughtering. Some supermarkets have stopped selling live chickens.

“Business is really bad. I didn’t even sell one chicken today,” said a vendor at a Xiangmei Road market.

Have a question-

If the virus is not a sloppy replicator, and can replicate without error in several different species, why is it that when it infect humans that we see obvious signs of changes? Lower lethality, etc? Is the assertion that these are recombinations, not mutations? If so, why do we not see change during decades of dormancy?

>Leo7 – at 23:50 >WetDirt:

>Please show more respect For RNA virus—HIV is a retrovirus >which is genetic info in a retovirus particle encoded by RNA. >Need I remind you no vaccine in 25 years? Yeah—I know we’re >talking influenza but how can you call it sloppy?

My most humble and fervent apologies. I thought it was clear that the sloppy replication was Conventional Wisdom, not my personal opinion. My opinion is that any explanation better account for the facts, and the facts are in the sequences.

wetDirt:

Yes, though not everyone agrees about what happens. Niman says:

“The major driver of influenza evolution is HOMOLOGOUS recombination, which involves using stretches of sequence identity to copy part of one gene and then switch to another. Homologous recombination happens when the polymerase is using one gene as a template. The newly created RNA the hops off the template 1 and lands on template 2 (the gene from the other virus in a dual infection).”

The details of this “hopping” business is what I’m most interested in (unfortunately, in most of the discussions around here that go anywhere near it, someone invariably ends up getting “hopping mad”).

Well, not exactly. Retroviruses do that, but influenza isn’t a retrovirus. Remember the “central dogma of molecular biology”? The cell’s machinery doesn’t work with DNA directly; it gets transcribed into RNA first. Influenza capitalizes on the fact that the cellular machinery can’t distinguish authorship of an RNA strand, and therefore dutifully copies any viral RNA strand delivered to the right place.

“Sloppy” replication is the conventional wisdom for a reason, and making a case for the opposite means challenging a PILE of evidence. And anyway, it is the so-called “sloppy” replication that makes the influenza virus such a serious threat in the first place; if it didn’t change, it wouldn’t keep coming back in a new and improved form each year, and we wouldn’t be worrying about a pandemic resulting from avian strains jumping species boundaries.

could it be, that the polymerse isn’t so accurate, but that mutations in the genetic neighborhood are not stable and thus die ? That would include that the 3rd-base-mutations, which don’t change the amino-sequence still have a function. Do they contribute to the 3d-structure ?

“could it be, that the polymerse isn’t so accurate, but that mutations in the genetic neighborhood are not stable and thus die ?”

That was my understanding, that is until I started reading you guys, now I don’t know what I know.

:-)

clearly those 3rd-base mutations are not as random as they should be. You can find some places, where the 3rd-bases just won’t mutate

Racter at 02:32

Have we considered that our circular logic has now described a line?

If a virus continues to be an infective threat season after season for 100 years like seasonal influenza at 30–35% population infectivity, wouldn’t we say that some very important parts of that viral RNA are being conserved?

Yes or No?

Perhaps the polymerase question is not so important as is the question of transposons, recombination rules and combinational successes in the wild?

I am well aware of the replicative failures even using PFU, a very accurate polymerase. That’s measured in a lab, where all good measurements begin and are later verified.

However, if replication is driven by external pressures and the virus responds by revising or acquiring genetic material, then we begin to understand rules-based recombination.

Consider that, in the wild, dual infections are the norm and that a myriad of other potential second-class, genetic donors exist in a host organism. A wide and fluid gene pool and a promiscuous virus, in the wild, may institute rules of discipline that allow the virus to only take one dessert at the buffet.

Another set of rules may be instituted in vitro where a relatively nominal donor pool exists and practically none of the pressures or natural in-vivo, fertilizers are available.

Take it a step further and consider that in the lab, the virus may do almost nothing that it does in nature because the lab cannot begin to approximate nature.

We do our best and have driven microbiology deep into intellect in the past 10 years, but we still don’t approximate that cozy real estate that a virus most appreciates.

Therefore, our body of knowledge on viral behaviour that is based on laboratory conditions may not be a solid standard against which to hold the identities via recombination that have occured in nature.

The sequences are there and are trackable. Some of the them are still there, anyway. Niman has produced them, somewhat obtusely at times, but he has lead you to the water.

Taste, drink.

If you look under the covers here, you may soon be the one leading others to the pond.

The sequences tell the story.

Seems that at this point, Niman is the only one who can get his thumb on the volume control.

A22,

Please confer with JKT and team on this question if the opportunity emerges.

Are extensive homologies (more than 25% identity within an influenza gene segment) between two or more wild strains separated by time necessarily contra-indicated by our lab observations of sloppy polymerase?

The previous post carries some minor supporting info to my meandering thought.

I am confident that JKT will arrive at a thoughtful answer that may shed substantial light here. We may be encasing our new data in some outdated packaging and I think that he will see the problem immediately.

Finding the identities is easier in swine, because genes in swine evolve more slowly. This can be easily seen in the PB1 genes in the Canadian swine isolates. The isolates were in 2003 and 2004 and the PB1 genes are clearly much more closley related to human PB1 than swine PB1. When I looked in March, there were 1111 human PB1 sequences at Los Alamos, which went through 2005. A BLAST of each of the PB1 sequences from the swine indicated the PB1 was most closley related to the human sequences from the mid-90s. Thus, the PB1 evolved more rapidly in humans than swine.

The swine examples I have are far more compelling than the Chinese H5N1 sequences that were pulled, because for the H5N1 conseervation there was only one H5 sequence from 2003 that exactly matched a sequence from 2001. However, the time gap is just two years, and although two years blows away the in vitro data, almost all in vivo data blows away the in vitro data.

The swine sequences are EXACT matches for large seqments of TWO 1977 isolates. One 1977 siolate matchs PB2 sequences and the other matches PA sequences. Moroever, for PA the matches are staggered across the gene, eliminating arguments that the conservation was because the region was “essential” and couldn’t be chanaged.

Thus, the identity of sequence isolated 26 years apart, effectively puts the “random mutation” pillar of influnce genetics soemhwere between nonsense and utter nonsense.

I don’t know if the data backs this up, but it seems to me that viral sequences from a strain of H5N1 that is adapted to a particular species doesn’t change as much in its sequencing (birds) than strains that make the jump to different species (people). In an embryotic cell, certain strands of DNA can be “turned on” to manufacture whatever particular protein is needed to differentiate the cell. What about viruses? Can an RNA virus “sense” it is in a different environment and initiate some kind of random sequence generator (or permit exchanges of sequences as in recombination)?

>Can an RNA virus “sense” it is in a different environment and initiate some kind of random sequence generator (or permit exchanges of sequences as in recombination)?<

Heredity and environment. Is there something else?

Dr. Niman is correctly identifying anomalies that require explanation. No question there. The anomalies are exact matches bewtween long stretches of nucleotides from isolates that are from different years. The issue is how to explain these anomalies. His explanation is that these examples are due to a polymerase that functions nearly perfectly in vivo. The anomalies represent a tiny fraction of the total number of available flu sequences. However, if you accept that the polymerase complex for flu is nearly perfect than it is much easier to argue that smaller regions of identity between different strains must be due to recombination. These smaller regions of identity are much, much more common than long stretches of homology. Thus the few examples of long stretches of homology are crucial to the idea that recombination is common in the wild.

Studies have been done to measure the error rate flu viruses in vitro. The measured rate is quite high. This empirical data is inconsistent with a super accurate flu polymerase, quite the reverse in fact. Given a high mutation rate, small areas of identity are quite likely to emerge, by chance. Other areas of identity may be expected to be conserved due to negative selection. This is the conventional view.

Dr. Niman has no explanation for the in vitro mutation rate, conventional science has no explanation for his examples of long stetches of homology. Hence the impasse.

The thread on lab generated H5N1 was an attempt to break this impasse. However, the sequence I was basing that analysis on has now been withdrawn by the (Chinese) submitters. Dr. Niman has other examples of relatively long stretches of homology which he believes invalidates the escapee hypothesis since some of these sequences are from North America.

Although it is possible that Dr. Niman is correct and all other flu scientists are wrong and flu polymerases work nearly perfectly in vivo but become very sloppy in vitro, I will advance several other hypotheses.

• Errors in submission of sequences
• Contamination in the lab during PCR
• Escapees from the laboratory - this could happen in North America, not just China

If there were many examples of long stretches of homology between different flu viruses, then none of my hypotheses would be credible. However, my understanding is that these represent less than .001% of all flu sequences. Experimental errors do occur. To determine whether they play a role in the long homology anomalies one would need to know how the samples were collected and what strains different investigators had in their lab when they were sequenced. For example, when was the 1977 strain sequenced? I don’t think it was 1977. Whose lab was it? Did they have other strains of flu in their lab from more recent isolates at the same time?

I give Dr. Niman full credit for identifying the anomalies. I agree they should not be brushed aside. However, I think we should not be limited to one narrow view of how they could be explained.

I would also recomend re-reading DemFromCT – at 23:56. Something to think about.

vis-a-vis in vitro vs. in vivo:

Has recombination in vitro been observed?

Racter – at 02:32 You are right, and I realised this sometime in the middle of the night. Brain was elsewhere while I was typing, finally caught up with it, dusted it off, and put it back in.

LMWatBullRun.

Yes, recombination in vitro has been observed. In fact, the virulence of different strains is studied by deliberately creating recombinants. However, it is accomplished via specific protocols that increase the odds of it happening and then it is (artificially) selected for. In fact, certain types of recombination might be considered evidence of lab strains.

By analogy, a dachsund is clear evidence of artificial selection. This animal is basically a highly modified wolf. But a wolf that looks like a dachsund would never have evolved in nature.

NS1:

A familiar lament, as this caveat may be rather broadly applied to many of our best attempts to understand the world around us. If the rules really do change in the instant that the refrigerator door is closed, we may never know for sure if the little light really goes out. The laboratory may not be perfect, but it’s still the best we’ve got; if it could not even begin to approximate nature, explaining the causes of disease and treating the victims would still be the exclusive domain of the shaman. The ability to predict what’s coming up in the next chapters seems to require a deeper understanding of the author’s methods than what may be immediately available from simply following “the story” (or maybe not; but it would still be nice).

wetDirt,

No worries. If my own brain featured perfect error-correction, I surely would have said: “…therefore dutifully translates any viral RNA strand…”

Monotreme, you wrote at 11:25:

“Although it is possible that Dr. Niman is correct and all other flu scientists are wrong and flu polymerases work nearly perfectly in vivo but become very sloppy in vitro, I will advance several other hypotheses.

Errors in submission of sequences Contamination in the lab during PCR Escapees from the laboratory - this could happen in North America, not just China”

Is it also possible that when an influenza virus sample taken from nature is grown on a different cell type or medium (let’s say for sake of discussion - a virus from human is grown on egg), that the virus sometimes must readapt its RNA sequence to replicate on the new medium, and thus the newly grown in-vitro lab sample (taken in this case from egg) might be a little bit different from the original (human sample)?

I am not completely familiar with the exact steps on how viruses are sequenced, but doesn’t the lab process involve replication on some type of medium? If so, and if the medium or environment for the virus is not exactly the same as the in-vivo conditions for that particular strain of virus, the virus may go through adaptation to the new medium. And thus the new viral sequence might be just a little bit different?

Here I distinctly remember when they were tussling with which influenza strains to include in a vaccine, and Nancy Cox describing how they could not get one of the strains to grow in egg.

I also recall some other articles describing research in which certain changes in the neuraminidase seemed to allow much better virus survival and replication in lower than usual pH levels (very high acid content).

So, just how fast or easily does the influenza virus adapt to changes in its immediate environment? Is the molecular-level mechanism which allows such changes known? Even when (or especially when) there are no other infleunza isolates present to make contribution to such changes.

Would a virus that is endemic in short-lived species have to adopt different strategies in long-lived species, or would there be systematic differences at the cellular level in the replication machinery? Are there different turnover rates for cells in long-lived species versus short-lived species? How about a parrot, versus a duck, for example? Does a parrot’s immune sysem look more like a duck’s or a human’s, considering it needs to recognize virus attacks over decades?

Well, I did have a series of conversations with JKT today, but of course not all on this problem. Here’s the bottom line: he says homologous recombination is theoretically possible but you just don’t see them with flu viruses, and certainly there is no evidence anywhere to suggest it as an important mechanism.

Anyway there was another presentation on treponema bacteria (syphilis) that touched on homologous recombination (exactly what niman has been saying about how you gotta look at the whole long sequence and not just snips etc) so I asked him is that what Niman is talking about. He says yes, but its far more common with bacteria and you just don’t see them with flu viruses. Now the guy is a molecular geneticist which means that he reads sequences for breakfast lunch and dinner so you gotta put that in that context. :-)

He also said (without prompting) that he knows Webster’s lab has got something that says recombination can happen rarely but he hasn’t seen the data yet. The way that he said it, and he said that every now and then about other things, says to me that is just good scientific habit, that you don’t conclude anything unless you’ve examined the data yourself, even if it is webster’s data and even if it is just one sentence.

In case anybody is wondering, I didn’t get that it was in any way meant to be evasive as he was very forthcoming about everything that I asked.

So Taubenberger certainly does not discount recombination completely, but just that it is so rare that ‘he doesn’t understand’ how anyone can conclude that as an important mechanism.

I asked about sequences being exactly the same. He says well some sequences, especially avian ones and internal genes, can be very well conserved. So I said what if it was HA, and he says (again) he’ll have to see the data but the first thing that comes up would be could it be contamination.

I’m thinking THAT might be the reason why the sequence was pulled. Cos the chinese got embarrassed when they realized the mistake, maybe?

With my non-existent laboratory background, I think contamination may be a very common problem because there was a presentation by a French guy on how they isolated the sequences from the 14th century to demonstrate that the Black Death was plague. They take the issue of contamination so seriously that after carefully creating the primer for a particular sequence, they only run it ONCE in the lab, cos otherwise if you get the same thing the second time, you don’t know if it was contamination from before. He called it ‘Suicide PCR’ which was way cool.

Black Death was plague? that’s gonna disappoint clark ;-)

Anon 22 at 15.18

Thanks for the update and I hope you will continue with them. I suppose I beleived that the contamination issue would have been examined before submitting to GeneBank as part of quality control. Now who said the French don’t have a sense of humor?

“Black Death was plague? that’s gonna disappoint clark ;-)”

Actually somebodies did raise a question, but this was a study on Marseilles, and I’m not sure whether it was supposed to be representative of the rest of Europe. I just didn’t get that part.

“I suppose I beleived that the contamination issue would have been examined before submitting to GeneBank as part of quality control. Now who said the French don’t have a sense of humor?”

They only run a database. They are not a research facility, how can they possibly verify all the data that is sent to them?

Anon:

Meant it to read the scientists would have examined for contamination prior to submitting to a data bank international scietnists have access to. I thought only data came from labs that follow international standards. Otherwise I guess I could submit sequences if I had some?

“Otherwise I guess I could submit sequences if I had some?”

Yeah, I got some of mine to submit, they run like this:

?????? ///// ????? XXXXX

Sorry, that was a bad joke :-)

“They only run a database. They are not a research facility, how can they possibly verify all the data that is sent to them? “


checking by computer for identical sequences from different years ?

Anon_22, during your series of conversations, did Dr. Taubenberger express any of his own concerns over the recent outbreaks in Indonesia and China?

“I’m thinking THAT might be the reason why the seauence was pulled. Cos the chinese got embarrassed when they realized the mistake, maybe?”

                                annon 22 15:18

So what we are saying is that by sheer coincidence, at the same moment that Monotreme made his claim, the Chinese realized that they had a contamination problem and had made a mistake after the sequences were in Genbank for three months being examined by a plethora of other virologists, geneticists etc.

I think its time to step back and give credit where credit is due. I assume that DemFromCt, Melanie, Pogge and Revere put flu wiki on the internet a year ago to have some influence in the world…if they didn’t then it is a lot of work to go through for an interesting conversation…and there are easier ways to do that.

Well, twelve months later, here is your proof that you are having an effect. As noted by others on flu wiki, there have also been other recent instances of items on flu wiki showing up later in other official venues…

…not bad for our small band of merrymen and women…

…Congratulations on twelve months of flu wiki, you deserve it!!

Thanks, but this one is likely coincidence. Others, I’m not so sure. ;-)

Well, if it was inaccurate, pulling the data back is the right thing to do.

I have great respect for your opinion but on this one…not a chance.

De jure,

“Anon_22, during your series of conversations, did Dr. Taubenberger express any of his own concerns over the recent outbreaks in Indonesia and China?”

No, it was really mostly pure science, like how does the flu virus jump species, and when we talk about receptors and stuff, do we really know enough to know what else we are missing that might have nothing to do with receptors? Interesting paradigm stuff like that.

I wanted to post a few more things, on his presentations as well as some of the ‘maybe’s, but I forgot to ask for his slides and I don’t know that I can give you accurate information from memory :-(

Y’all going to have to wait for a bit till he responds to my email. Sorry.

You think they reacted that fast?

Let’s assume for the moment that there are stable sections of sequence that, for whatever reason, behave differently than other regions of instability. Why? What’s the mechanism?

Anyway, this has been a fascinating discussion, once again illuminating how much we don’t know about this virus. Well, let me speak for myself. What I don’t know about this virus could fill several libraries. That’s why every time something else is brought up that I don’t know, I’m never shocked. ;-)

“What I don’t know about this virus could fill several libraries. That’s why every time something else is brought up that I don’t know, I’m never shocked. ;-) “

Well, I DO remember on the slides that Taubenberger has more questions about what we don’t know than answers about what we do know.

As I said the other day, the more I read Webster and JKT’s stuff, the more I’m struck by how frequently they talk about what they don’t know.

annon and Dem. Too bad that the WHO didn’t follow the example of Dr’s Webster, Taubenburger and many others overthe last two years. If they had things would be quite different today.

Using software to exclude sequences leaves much to be desired. We could code only for the rules and patterns that we recognise today. What if we failed to see a pattern before we encoded the rules.

As a first check, software is great. But not to exclude.

A potential error report is likely produced?

Anyone know the IT guys at GenBank?

wetDirt:

The way these questions are framed makes it hard to even attempt to answer them. As convenient as it might be to treat properties like “longevity” and “pathogenicity” as if they were explicit features of an organism’s genome, in reality they are emergent properties arising out of complex interactions: between various parts within an organism, between the organism and its environment, etc. If we identified some specific aspect of (say) cellular structure as critically affecting longevity among potential hosts, we might speculate on how that might influence (say) pathogenicity of a specific virus for a specific host — but we could also do the same with many other properties besides longevity. It makes sense to start with receptor specificities, not only because it’s obviously important, but because it’s something we have at least some kind of a handle on; we can even identify specific genes as being instrumental. “Longevity”, by comparison, is hopelessly fuzzy; where exactly are the genes for that?

Racter— (and I have both the book, “the policeman’s beard is half constructed, and the original program on 5″ disk),

The exact thing in my mind was centromere length—I was thinking, I wonder if cells designed to reproduce many many times have tighter specs on the polymerases than cells that will only reproduce a few times, like the diff between stem cells and neurons, say. Do ducks have have shorter centromeres than parrots? Does that help?

“As I said the other day, the more I read Webster and JKT’s stuff, the more I’m struck by how frequently they talk about what they don’t know. “


Webster became this only after they pulled him back when he gave his ABC-interview.I didn’t observe this before. There seems to be a government directive…

“Webster became this only after they pulled him back when he gave his ABC-interview.I didn’t observe this before. There seems to be a government directive…”

Not true. I am talking about journal articles and scientific papers, not interviews.

wetDirt:

Since copying of viral RNA is performed with polymerase coded for in the viral genome rather than the host genome, it wouldn’t matter, at least not directly, or in any way we’re presently ready to understand. Interaction of the viral genome with the host nuclear structure is not well understood, but I don’t see the virus picking up any tricks from the host DNA.

I’m glad you like my nick, btw. I like yours too.

anon_22, thanks from the report from Taubenberger. Pretty much what I expected. Contamination is a possibility. As far as the timing of the withdrawal, that’s for another thread.

How fast a virus can adapt to a new host depends on it’s mutation rate. If that’s very low, then adaption is going to be extremely slow. The faster the mutation rate, the more raw material for selection to act on.

I was away yesterad and will be away today, but swine data do away with hand waving arguments about wrong sequnce, contamination and otehr trivial explanations.

The conventional wisdom would say the conservation is due to a lack of selection. The flu in swine is in equilibium with the swine so there is no need to cnage to survive, so most errors don’t see the light of day (they don’t offer an advantage and are overwhelmed by the sequences that are copied without error.

There is some truth to those comments. Swine flu does evolve more slowly, because there is less selction and fewer cases of recombination. However, this slow motion allows a view of what is really happening. Sone of teh swine have nested segments with a 1998 isolates from North Carolina. Therefor the center of teh gene goes 1998–1977–1998. This is because the 1998 portion has the 1997 portion added in the middle of the 1998 sequences. This addition essentailly goes “backwards” in time (1977 is AFTER 1998) because 1977 sequences are in 2003 swine, so 1998 region can acquire 1977 after 1998.

The swine series involve all of the isoaltes in recombination and teh exampke shows why it isn’t a lab articfact because it would require many different sources present the same time for sequencing of one gene (like PB2) and then another set for another gene (PA). Thus, the lab wouldhave to be VERY sloppy again and again.

Those upholding teh “conventional wisdom” will acknowledge that recombination happens sometimes, but it is hard too find and isn’t important.

Conceptually, that really dopesn’t make sense, because if flu can recombine in swine as indicated, then it can happen MUCH more frequently in dual infections involving closely related sequences because there are VERY long stretche of identity, which provides MANY more chnaces for the newly transcribed sequences to hop off one templat and onto another.

In fact it is the frequntly chnages over time that reduce the long stretches on identity. In the example above the long 1998 region was reduced to two sorted regions by sticking 1977 sequences in the middle (creating 1998–1977–1998 from 1998–1998–1998.

However, flu likes to make minor changes (drift) so there is more selction from tweeking the gene. Thus, the single base changes are more common. However, these chnages can pair up on sourcing.

A good example of that can been seen in the Indonesian sequences just released, which can be seen here.

These are partial sequences from HA and teh Indonesian sequences are clearly most closley related to other Indonesian sequences. However, one has a Qinghai cleavage site (GERRRKKR), which is overlayed on an Indonesian background. However, it also has anpther polymorphism, that is shared with Qingai sequences. This polymphism is also found in another new Indonesian sequences, and that sequences has a third Qinghai sequence.

These Indonesian sequneces have a VERY limited number of these ‘foreign” polymorphisms. However, the acquisition is NOT random. Each of the two isolates above (in bold in the linked list) have TWO Qinghai polymorphisms, whch pribably came from the same parental sequences. However, a longer stretch could have beebn interupted by subsequent acquisitions from Indonesia, so not the two Qingahi polymorphisms are seperated by the more common Indonesian sequences.

Thos upholding the convention wisom really don’t look for these relationships. They assume these are all “random mutations”. However, the “travel history” of the polymorphisms is quite clear in the linked list.

THe convention wisdom is evolving. There are three stages:

1. It can’t be true

2. Even if it is true, it can’t be important

3. We knew it all the time

Most are at stage 2. They will acknowledge that homologous recombination happens, but it is said to be rare and unimportment. However, it happens all of the time, bit most examples are called “random mutations” because the acquired information is only a single nucleotide. However, sometimes it is a small cluster, like the European swine example. Other times it is like the Canadian swine listed above.

Influenza A Virus Replication http://tinyurl.com/m7fg5

Dr. Niman, there are over 11,000 H3N2 sequences in GenBank. What proportion of those have long stretches of identity? Most of your examples seem to involve H5N1 or H1N1. Is that because you are focused on this viruses or do you really see more signs of recombination in these subtypes than in H3N2?

Does isothermal DNA amplification apply?

PCR Troubleshooting: The Essential Guide http://www.horizonpress.com/hsp/books/pcr2.html

>A unique PCR troubleshooting guide that is an essential companion for anyone who uses the polymerase chain reaction technique. Aimed at a reader with some experience in PCR the book discusses the many and varied problems encountered with PCR together with tips, advice and procedures to obviate rather than overcome the PCR problems. Written in the language of the laboratory with a little humour and a down-to-earth approach, the book is easy to read and understand. If you fail at PCR, consult this book! The advice in these pages is invaluable and is the sort of advice that is not usually found elsewhere.

In the words of the author, remember that there are many other things in life than PCR … for example, isothermal DNA amplification.>

http://www.biohelix.com/pdf/EMBOr_HDA.pdf

For those of us that enjoy learning from diagrams and pictures, there is an article at http://tinyurl.com/zh3qy which on page 4 has a color diagram which clearly shows the difference between mutation - recombination - and reassortment events when influenza virus replicates.

This review article is one of the most lucid and carefully written pieces I have come across, and might have value as a general education article for those wishing to get up to speed. However, it does not seem to sacrifice content for lucidity. Its title is “Molecular constraints to interspecies transmission of viral pathogens”, by Richard Webby, Erich Hoffman & Robert Webster.

The article also has a nice color picture of ciliated vs. non-ciliated cells from human trachea, showing (respectively) positive vs. negative staining for the avian alpha-2,3 type of virus antigen on those two types of cells. This shows that the 2,3 type of receptor does exist in the human trachea.

Here are a few sentences from the article which address RNA viruses in general, these concepts would apply to influenza virus:

“Many of the emerging diseases that threaten humans are caused by RNA viruses and this is not by chance…RNA viruses are extremely mutable and use very efficient strategies for generating viral diversity…RNA viruses have few or no proofreading mechanisms and many mutations are introduced during replication. As such, RNA viruses exist as a quasi-species comprising viruses of slightly different genetic composition…variants with an advantage are quickly amplified.”

The article goes on to explain the three types of processes (mentioned above) that lead to evolutionary change in the influenza virus.

Thank you, B-hiver.

Thanks, beehiver!

“mutation - recombination”, which I prefer as well, has become the cw way of avoiding a fight over which is the more important process. it’s a different process entirely than reassortment. It’s why I sometimes use the phrase “it doesn’t matter”, because “mutation - recombination” does not require figuring out the proper ratio of one to the other.

Everyone agrees all three happen. When the topic is germaine, such as this one, then the ratio matters. For most of what we do on the wiki, it does not. Now, that is really not a profound statement, it’s rather banal. But one doesn’t really need to get all worked up about the banal.

You’re very welcome, MW42 and DemFromCT.

I posted this article because I have sensed many would enjoy a better understanding what is going on with all this discussion about mutation vs. recombination vs. reassortment. The diagrams help to bridge that gap. The ratio of incidence of these three processes is indeed another entire matter. Monotreme raised the question of whether that ratio could vary among different H or N types, and Dr. Niman has raised the issue of the incidence ratio being higher or lower among one species type (in this case, swine). And perhaps both of those phenomenon are occurring in nature at the same time. But Dem, you are right, none of the finer points of this discussion change the fact that at this point in time, we have a potentially great big problem looming in “Dodge”.

beehiver, this is my graphic for that. Idle hands is the devil’s tool. ;-)

Now, I’m still interested in learning what’s in the black box, mind you.

Oh, and someday I’ll learn how to spell germane.

DemFromCt. I agree…now that should go on the door of the experts.

However, shouldn’t the A) part of the picture be relatively smaller than the B) part…just to be numerically and porportionaly correct?

There are some who claim they KNOW how A works, and therefore it should be bigger ;-)

Actually, “current status” and “potential pandemic” should be singular and shared.

Yep, but that would confuse things and dare I say it, cause arguments.

I love all of you’s sense of humor. Tom you seem to have noticed how dicey it is to post anything without causing an argument, lol. Kudos to you all and to the mods who do a fabulous job. And sorry for the off-topic post.

To get back on track! A question has been bugging the heck out of me.

You are in a lab setting, and have a single strain of influenza virus and intend to passage it several times through a specific cell line or healthy live animals, to test for increased virulence as the number of passages increases, and the virus adapts to the live material.

If / when the virulence begins to increase, wouldn’t that by definition be due to mutation as opposed to recombination or reassortment, because you are using a supposedly pure strain of virus, and there is no other strain to make a contribution? This would assume there is no contamination from any other influenza virus strain in your culture or animal experiment.

yes, but why not just sequence the virus, once it becomes more virulent ?

Beehiver at 12.39

Thanks for the article but also for pulling out the paragraph on RNA—which stated so much better than me, on my own thoughts. Virulence has to be in the mutations, where else can it be explained.

Anoymous at 15:21, yes that is a good point. I guess my question becomes how could anyone really decide in any one situation out in nature, if a change in a viral sequence is due to mutation or recombination (or even reassortment), without having access to the sequences of every other strain that might have been circulating in the area, and knowing all the important environmental circumstances (especially presence of other animals, etc).

It’s hard to believe that the submission of influenza viral sequences to the databanks is anywhere near complete, certainly not now and especially from earlier years. We all know how these viruses can smolder unseen for periods of time. And, even if there were more submissions, it would be difficult to pull up relevant sequences from any one particular locality. What might be present at one end of a state or province, might be entirely different at the other end. So can relatively spotty sequence data (as it currently exists) be reliably used to prove in what way the virus strains are evolving, except in the most obvious large changes such as reassortment…or perhaps when there is intense widespread surveillance including many submissions to the databanks? It just seems that quite often, we might be lacking the necessary data to settle on which mode or modes of viral evolution are occurring at any one point in time. Just my thoughts.

Beehiver-

Excellent investigation in these recent posts!

You’ve said it well . . .

We lack the necessary data to know anything with certainty.

No data = No answers = No predictions

Niman may have found a way to illucidate patterns from the small amount of data that is available. If he has, I hope that he will share more of the rules with us soon so that some of us can track his efforts more closely.

Monotreme, I have focused more on H5 and H1 than other serotypes, but swine H1 sequences have been in human H3N2 and the acquisition is in the dominant haplotype, although the region acquired is small.

beehiver – at 15:12 ‘’To get back on track! A question has been bugging the heck out of me. You are in a lab setting, and have a single strain of influenza virus and intend to passage it several times through a specific cell line or healthy live animals, to test for increased virulence as the number of passages increases, and the virus adapts to the live material. If / when the virulence begins to increase, wouldn’t that by definition be due to mutation as opposed to recombination or reassortment, because you are using a supposedly pure strain of virus, and there is no other strain to make a contribution?’‘

You are correct. However, I think Dr. Niman argues that the polymerase complex operates more efficiently in vivo than in vitro. Hence, for him, in vitro experiments are irrelevant. I’m sure he’ll correct me if I’m misinterpreting his position.

Dr. Niman, I’d very much like to know if you can find proportionately the same number of long stretches of identity in H3N2 sequences as you have found for H5 and H1. Given the much larger database of H3 sequences, I assume it would be easier to find convincing examples of obvious recombination, even if these are rare. I don’t know what kind of software or server capacity you have available to you, but if you need some help, there are tech oriented people who would probably be willing to help.

Additional examples of extensive identity would strengthen your case. If you cannot find proportionately as many in the available H3N2 sequences, I will need some tinfoil.

DemFromCT, the mechanism for evolution of H5N1 might matter.

treme,

Why the emphasis on H3N2 . . . is it just the larger range of available sequences or are you chasing something phenotypical?

Wouldn’t you expect to find some variation in the numbers between subtypes?

Maybe we should compare all known High Path influenza v. all know Low Path influenza?

NS1 – at 23:58 “Maybe we should compare all known High Path influenza v. all know Low Path influenza?”

Maybe we should compare all influenza v each other and within subtypes. That’s the basis of the Influenza Genome Project, which is v ambitious and being headed by Taubenberger who is moving to the NIH next week, and Scott Layne of UCLA. The idea is to get everyone to give them either sequences (I think) or samples of any flu virus from anywhere anytime and build a database. That will answer a lot of the current ‘don;t know’ hopefully.

Also jkt and john oxford are teeming up to sequence autopsy samples of suspected flu cases from 1905 −17 from the Royal London Hospital. Apparently they used to do autopsies on every death at the hospital, so there’s a huge archive sitting in basements with clinical notes and tissue samples etc. They’ve identified 200 patients plus 200 controls, and will do the painstaking process of trying to find fragments to piece together, like what JKT and his team did with the 1918 virus. They hope to find what was in circulation prior to 1918.

I’m probably going to start another thread later today for this.

Monotreme, I have already explained that human flu evolves faster than swine, so I am not sure why you are expecting a large number of long stretches if identity in humans.

Most change is via small stretches and as indicated, Human H3N2 is acquiring swine H1N1 sequences. The swine sequences are in the dominant human haplotype.

1968–1969: In 1968, a new flu strain, H3N2, originated in Guangdong Provence in Southern China. The first indication of the problem came when a rise in respiratory illness was noted in Hong Kong and then reported in the press in mid-July. The outbreak had a significant impact on Hong Kong, infecting fifteen percent of the city. Despite the city’s smaller size and more limited contact with outside regions, the disease quickly spread around the world. Overall, the spread followed a similar pattern to that of the 1957–1958 flu pandemic. Beyond Asia, the epidemics were more circumscribed as worldwide dissemination of the virus occurred over the year. High excess mortality was really observed only in the US. The excess death figure for the US was about 28,000 which put it lower than the flu toll in 1957–1958. This lower death rate probably occurred because the Hong Kong strain had only a surface antigenic shift from pre-existing A-strains. As a result, the epidemic may have been somewhat blunted by partial pre-existing immunity.

…may have been somewhat blunted by partial pre-existing immunity.

If it cannot be stopped, can it be blunted?

Co-circulation is a distinct possibility.

Frequency of Amantadine-Resistant Influenza A Viruses during Two Seasons Featuring Cocirculation of H1N1 and H3N2

http://jcm.asm.org/cgi/reprint/41/5/2164

All -

Here is my question. We are searching for a ‘universal’ flu vaccine, one based on highly conserved features of the virus. The matrix protein comes to mind. If the matrix protein is so consistent that we can hope to derive a ‘universal’ vaccine based on it, where is the variability that random mutation would seem to predict? It appears to me that the variability of the H and N surface proteins (that cause us to have to reformulate the seasonal trivalent vaccines every year) is not reflected in the matrix proteins and other internal virus proteins. The variability of the surface proteins allows the virus to dodge mechanisms of immunity, enhancing its survivability, no?

This is where I start to beome a Nimanite - I don’t agree with all of his conclusions, but he has hit on some troubling points - one of which is that there seems to be an as-yet unexplained mechanism that yields genetic diversity in the bug that is ultimately good for the bug (and bad for us) - that it does indeed seem that the ‘virus knows what it is doing’. ‘Sloppy’ polymerase would seem to me prone to errors in the replication and hence wide variability of all viral proteins, not just the surface proteins.

Please help me out here.

Dr. Niman, Even if H3N2 is evolving faster than H5N1 or H1N1 there are so many samples that the large areas of indentity seen in your examples should also be seen in some of the H3N2 sequences.

NS1, I do think it’s important to see if there is something different going between H5N1 and H3N2.

anon_22, great news about the flu database and Dr. Taubenberger’s involvement. We should get answers to many of our questions from this work.

The Sarge

This is where I start to beome a Nimanite - I don’t agree with all of his conclusions, but he has hit on some troubling points - one of which is that there seems to be an as-yet unexplained mechanism that yields genetic diversity in the bug that is ultimately good for the bug (and bad for us) - that it does indeed seem that the virus knows what it is doing.’

I am also a Nimanite to the extent that I think something strange is goin on with H5N1. Viruses don’t know what they are doing - but people do.

Monotreme -

A virus shouldn’t know what it is doing - being insensate and not really ‘alive’ in the usualy accepted sense. But I do perceive a certain teleological inference in Dr. Niman’s assertions. (The ‘virus knows what it is doing’ remark was attributed to him in another thread.)

However, if I grasp your implication - has, or is, the evolution of the virus being externally and consciously directed - by man? This I don’t know. I am disinclined to think so at this point. Evil bugs have arisen unaided by human will. However, the technology now exists to create them from scratch - and talk about plausible deniability! I just think that influenza would be way down on the list of bugs that someone of ill intent would create, precisely because there are so many unknowns about it. But, until science catches up with answers - who knows?

I going out to the store now for a fresh roll of tin foil ;o)

A/swine/Iowa/15/30/(H1N1) and A/swine/St-Hyacinthe/148/1990(H1N1) also look surprisingly similar. Maybe someone can access:
http://www.sciencemag.org/cgi/content/full/293/5536/1842

here you go- A-NON. Access granted.

http://tinyurl.com/ohcpj

Granted, of course, as lomg as you don’t ask for the whole text in a pdf file…

…And then, require access sign-on. sorry, you already tried this.

Swine from 1931 also related to 5′ end of 2003Canadian swine.

Niman:

Highly discontinuous distribution of conserved areas within genes does seem persuasive as evidence for recombination, but where the genes involved are those directly affecting copying fidelity, I get lost in the recursion. I would have a hard time accepting that the polymerase acidic protein is NOT essential (or at least important), and I don’t believe you are suggesting that. I also don’t see you saying that selection is not at work on every aspect of this. You seem to be saying that because the third bases are invisible to selection at the level of protein, they are available for exploitation by the virus to permit it to perform strange tricks during replication — tricks which offer a selective advantage. This seems to involve a paradox I cannot resolve; my simple mind demands that the third bases must either be invisible, or not. Strange replicative tricks require specific mechanisms, mechanisms which involve enzymes and other proteins, (some of) which must be coded for in the viral genome, and which therefore become visible to selection as the associated codons are expressed (unless part of what I am missing is that you are saying that portions of the genome may be visible to selection before they are expressed).

shouldn’t these 3rd base changes occur statistically balanced ? When there is no selection beyond the amono-acid level, then I can’t see why some encodings are obviously preferred by nature.

bump

at 139 from 542 HA-positions there were no changes in the 3rd base. Another example for 100% accuracy ? Also the mutations at some places seem to be linked to each other since many of the counts below are larger than 1

sequence pairs out of 304 full-length HA-H5N1-sequences from 2004–2006, which differ at a fixed 3rd-base location within HA
,
count of locations with that many pairs differing there

 00000 ,139
 00606 ,87
 01208 ,56
 01210 ,5
 01806 ,35
 01810 ,5
 01812 ,1
 02400 ,17
 02406 ,4
 02408 ,1
 02990 ,7
 02998 ,4
 03002 ,2
 03576 ,16
 03586 ,2
 03592 ,2
 03594 ,1
 04158 ,8
 04170 ,2
 04178 ,1
 04182 ,1
 04736 ,4
 04750 ,2
 05310 ,5
 05326 ,3
 05340 ,1
 05880 ,5
 06446 ,5
 06482 ,1
 07008 ,6
 07078 ,1
 07566 ,2
 07590 ,3
 07630 ,1
 08120 ,7
 08170 ,1
 08186 ,2
 08670 ,8
 08742 ,1
 09216 ,5
 09246 ,1
 09294 ,1
 09758 ,5
 09900 ,1
 10296 ,3
 10830 ,2
 11360 ,3
 11886 ,2
 11926 ,2
 12408 ,3
 12618 ,1
 12926 ,1
 13440 ,3
 13728 ,1
 13950 ,1
 14238 ,1
 14456 ,1
 14958 ,2
 15310 ,1
 15456 ,1
 15606 ,1
 15840 ,1
 16926 ,1
 16986 ,1
 17408 ,1
 17886 ,1
 17902 ,1
 18360 ,2
 18488 ,1
 20290 ,1
 21566 ,1
 22294 ,1
 23310 ,1
 23736 ,2
 23826 ,1
 24158 ,1
 24250 ,2
 24576 ,5
 24990 ,2
 25768 ,1
 25806 ,1
 26208 ,1
 26294 ,1
 27168 ,1
 27602 ,1
 27776 ,1
 30008 ,1
 30366 ,1
 35550 ,1
 41206 ,1
 44640 ,1
 44958 ,1
 45326 ,1
 45408 ,1
 45486 ,1
 45560 ,2
 45670 ,1
 47848 ,1

anon,

Please tell us more about your algorithm that produced the table?

algorithm is too big a word. Just counting differences in the nucleotide-sequences aligned with clustalw. About 64% of mutations occur at the 3rd base, 18% at base 1 and 18% at base 2.

A more detailed statistics would detect whether a mutation results in an amino-acid change…

Please send a spreadsheet of the table in 03:05 post with titles to domain singtomeohmuse dawt communications tool with my posting name as the login.

send china.45
200 Port command successful
550 china.45: Permission denied


table uploaded to here

see here for the list with the
number of pairs (out of the 304*304 combinations which differ at the fixed position. 1626 aligned HA-positions chosen.

Link is unavailable at magictour.

please transmit to the reflection of 1sn am singtomeohmuse punkt com

404 on the second link also

don’t know, what “reflection of 1sn am” means. Above links works fine for me.


for example the 00606,87 entry above means, that there are 87 positions
where exactly one of the 304 sequences has a mutation and the other
303 coincide.

reflection, like in a mirror. print the page and hold it next to a mirror.

the reflection of 1sn is my login name.

am = at punkt = dot

The links don’t work for me.

what does the 00606 mean?

maybe you have to wait some time until your server gets it. The files were just uploaded some minutes ago.
I’ll try the reflection thing now…
00606 means: 606 pairs of sequences

331 Password required for ns1.

email it to me. not ftp. just email

Don’t you have a class to teach?

PairCnt , ?
00000 ,139
00606 ,87
01208 ,56
01210 ,5

Racter, The PA example shows large regions of conservation between 2003/2004 isolates and a 1977 isolate. There are many isolates that have active polymerase genes that have changes in portions of the gene that are unchanges for over 25 years. Thus, there argumanet that the conservation is due to a essential sequence fails because other isoaltes have succesfully changed that region. The PA examples, with active links are in this commentary.

Here are the six recent isolates which have identity with the listed 1977 isolate. One example of what I said above is the region between 150 and 541. The region is exactly the same in a 2003 isolate (53518) a 2004 isolate (48235) and the 1977 isolate (Tennessee/24). Thus, even though this region was unchnaged for 26 years of copying, it wasn’t because it was too essential to chnage because the other four recent isolates listed below changed it (as did virtually all other flu isolates in the database). The same argument can be made of the back end of the gene where 23866 and 48235 match the 1977 isolate between 992–1745, but all other isolates chnage this region.

48235 is the same across most of the gene. Differences flank the region of identity. The clustering of the chnages at the two ends of the gene is more evidence for recombination.

The same arguments can be made for the PB2 gene, where 5 of the 6 isolates below have large regions of identity with another 1977 isolate from Tennessee (Tennessee/26/1977).

Identity in PA with Tennessee/24/1977:

A/swine/Ontario/11112/04(H1N1) identity between 721–1319

A/swine/Ontario/23866/04(H1N1) identity between 992–1745

A/swine/Ontario/48235/04(H1N2) identity between 150–2016

A/swine/Ontario/53518/03(H1N1) identity between 25–1469

A/swine/Ontario/55383/04(H1N2) identity between 589–1787

A/swine/Ontario/57561/03(H1N1) identity between 541–1817

A/Swine/Tennessee/24/77 (H1N1)

why is it evidence for recombination ? recombination can’t explain this curiosity either.

I mean, recombination can’t explain why sequences don’t change over decades. It does explain the Tennessee-Ontario combining.

Henry,

Thanks for the nicely formatted example of the PA identities to the 1977 TN Swine.

Can you use that exact format with 5 more origin strains across some of the other gene segments?

We’ll then have a concise catalog in one spot to show your ideas.

anon,

I don’t think that recombination is so concerned about the time spans.

Niman’s recombination describes the transformations and their ancestries. The back-tracking is what makes it remarkable because what can be tracked also becomes more predictable.

He-who-must-not-be-named:

There are three alternatives. Reassortment can’t explain it, due to a problem with what connoisseurs of programming languages refer to as “granularity”; reassortment deals only in whole segments. Simple mutation can’t explain it for nearly the same reason, but in reverse: we would expect to see wide distribution of changes rather than clusters. The conventional view of recombination as too rare to be significant can’t (or won’t) explain it. Some combination of mutation and reassortment (or mutation, reassortment, and recombination) might explain it, as might deliberate or accidental tinkering by humans (or some combination of deliberate and accidental) — but next to that sort of thing, I’d have to regard Nimanesque homologous recombination as the default by Occam’s razor.

As I have indicated to Dr. Niman on several occasions, the examples he gives do look like recombination. I don’t have a problem with this mechansim. What I do have a problem with is a super accurate polymerase. I need to check on the 1977 sequence, but I don’t think it was sequenced in 1977.

Issues to be resolved:

• How to explain the in vitro mutation rate?
• Why do most of the examples involve H5N1 or H1N1? There are many more H3N2 sequences available.

Niman:

This is where the logic throws me (and Monotreme as well, looks like). In my experience, this is usually the result of flawed assumptions. For recombination to be as significant as you claim and yet to have gone unappreciated by so many seems to suggest that the mechanics must be fairly subtle and sophisticated, and therefore rather sensitive themselves to small changes in the areas where the “rules” reside — which I further assume must be in the polymerase complex. The unnamed one also makes a good point: is not near-perfect conservation over decades a problem for recombination as well? I mean, it’s supposed to be a mechanism for change, isn’t it? You can have “mutable rules”, and you can have “immutable rules”, but when you start talking about the rules determining which rules are mutable and which are immutable, things start getting pretty hairy.

Swine evolves more slowly because there are fewer dual infections involving diverse isolates. However, flu does evolve in swine. It is just easier to find examples of recombination or sequence conservation that are more obvious.

The most common recombination involves two closley related genes. In those circumstances the acquired changes look like mutations.

Even when more diverse sequences are involved, such as Qinghai sequences on an Indonesian background, the Qingahi sequemces are limited to 1 or 2 poymorphsims.

That can be seen in the recent Bali sequences.

One isolate has a Qingahi cleavage site and one additional Qinghai polymorphsim, while the other has two Qinghai polymorphisms.

The number of infprmative polymorphsims in these isoaltes is limited. Usually there are about 1/2 dozen that have a limited distribution witin the closely related sequences (such as the other Indonesian sequences in this case). Most of these informative chnages will still be from H5N1, but sometimes they are paired, like the two examples above. Each isolate has TWO Qinghai polymorphisms, which are NOT due to random mutations, even with selection thrown in.

A quick and simple comment.

In the industrialized countries, artificial insemination of swine (and other farm animals) is very common. If my memory serves me correctly, semen or eggs (or accompanying body fluid) from an infected animal could preserve a virus when they are frozen in liquid nitrogen. These are sometimes stored for years. What are the chances that the donor animals were infected with influenza and then it was passed on to the recipient animals years later? I do not know. But this is one mechanism for preservation of a particular strain of influenza over a long time span.

Sorry about muddying the water, but it needed to be brought up I think.

Racter, you’re correct about my concern. For perfect conservation at the wobble position over 30 years, you have to assume not only a perfect polymerase, you would also have to assume a perfect mechanism for error correction. I find it very hard to believe that there would no trace of this other than a handful of examples.

The bottom-line is that most researchers assume random mutations are common and recombination is rare, while Dr. Niman assumes random mutations are rare and recombination is common. More sequences and better databases should help resolve this issue.

Okieman, I like your hypothesis. I don’t know how likely it is that flu viruses could be found in semen, but the idea of transfer of an infectious agent from frozen material used in animal husbandry is intriguing.

RE: Monotreme – at 12:21 …transfer of an infectious agent from frozen material used in animal husbandry is intriguing.

Would this then, be an example of deliberate mutation? Or deliberate recombination? Or - and/or, hence, random error?

Or “C” - all of the above.

Tom, Chime in

Mr. White42. I am here. Not sure what you wanted me to ‘Chime in’ on…

…I am not a pig expert but I believe that most artifical insemination in pigs is done with fresh not frozen semen as occurs with dairy and beef cattle.

I do not believe that semen would be a vector for influenza virus.

The only explanation for what Monotreme observed other than what has already been stated is that a current virus reassorted with another virus to regain missing segments it had lost years earlier…still seems like a long-shot.

Weaponizing influenza virus would be like using the world’s largest atomic etc. bomb to turn an asteroid into the world’s path…kind of an oxymoron.

I was referring mor to :”frozen material used in animal husbandry”

when we see a sequence like Tennessee/77 in the database, then this isn’t sequenced in 1977, since they could not do this in 1977. So, this is from some conserved tissue-sample which then was sequenced only in ???? (since when can they sequence viruses?)

Can someone do a summary of this thread and then re-build a new thread for tomorrow? Thanks.

there are several papers mentioning these sequences, but nobody comments on the exact match over 20 years ! Did they miss it ? Or didn’t they consider this remarkable enough ?

http://www.flutrackers.com/forum/showthread.php?t=2268
http://jvi.asm.org/cgi/content/full/73/10/8851

as for sequencing the 1977 sample, Webster refers here to a method from
Chen,E.Y., and P.H.Seeburg,1985. Supercoli sequencing:a fast and simple method for sequencing plasmid DNA

when virus replication can be so exact over 26 years, why then isn’t this reproducable and they can’t verify this in a laboratory setting ? Where are the papers about exact virus replications over years ? And why are exact matches so rare ?

The sequences for all of teh isoaltes are linked in the commentary for PB2 or PA.

The links have the sequence, which includes submission date. The Tennessee/24 sequences was done around 1989 and the Tennessee/26 sequences was around 1990. However, it should be noted that although the matches with the 2003 and 2004 sequences are exact for long stretches of the genes, none of the genes are exact for the entire sequence.

As I said earlier, the sequencing was done over a decade apart and in different labs.

In reference to Tom’s post:

Tom DVM – at 00:30

“…I am not a pig expert but I believe that most artifical insemination in pigs is done with fresh not frozen semen as occurs with dairy and beef cattle.”

Tom is probably correct. I am only familiar with dairy and beef cattle. I was in the dairy business for a number of years and used artificial insemination for about three years. Use of frozen semen is the norm in the dairy business.

The issue of frozen embryos may remain. I just don’t know. Anyway, this in all likelyhood is not the reason for certain portions of influenza sequences being maintained over long periods of time.

Here is the link for the revision history of Influenza A/swine/Tennessee/26/77 (H1N1) RNA polymerase 2.

Accession M26077 was first seen at NCBI on Aug 2 1993 7:39 PM

I BLASTed the original sequence and can confirm the identity relationships Dr. Niman has listed with the Ontario sequences from 2003/2004. Unless there were many strange clerical or contamination errors, it would appear that recombinant H1N1 viruses from 2003/2004 contain segments from a virus isolated by Dr. Webster’s group in 1977 sequenced in 1989 and deposited in GenBank in 1993.

This needs to be explained. anon_22 or anyone else who can reach Dr. Webster and Dr. Olsen (who’s lab sequenced the Ontario viruses) should ask them how they think this happened.

anon_22, You might also ask Dr Olsen why there is no mention of recombination in his peer reviewed published paper on the swine sequences as well as no comments on the identities (which also apply to EXACT matches with other viruses in PB2)

From PB2 Commentray:

A/swine/Alberta/56626/03(H1N1)

A/swine/Ontario/11112/04(H1N1)

A/swine/Ontario/23866/04(H1N1)

A/swine/Ontario/48235/04(H1N2)

A/swine/Ontario/53518/03(H1N1)

A/swine/Ontario/55383/04(H1N2)

A/swine/Ontario/57561/03(H1N1)

A/Swine/Korea/CY02/02(H1N2)

A/Swine/Tennessee/24/77 (H1N1)

A/Swine/North_Carolina/35922/98(H3N2)

As noted earlier, Ontario/11112 was an exact match with North Carolina/35922 for positions 756 through at least 1601 (only a partial sequences from North Carolina was available - sequence numbers adjusted to reflect longer sequences)..

Ontario/53518 similarly was an exact match with Korea/CY02 from position 568 to the end of the sequence.

Ontario/23866 matched Ontario/53518 through position 568 and then matched Tennessee/24 for positions 1008–1326.

These data showed that each of the above recent isolates exactly matched an earlier isolate indicating each was a recombinant. The data also provided evidence for absolute fidelity in copying a large gene segment for 6 years and a shorter segment for 26 years. These data raised serious questions that attributed rapid influenza evolution to random mutation

niman – at 10:55

These data showed that each of the above recent isolates exactly matched an earlier isolate indicating each was a recombinant.

I agree.

The data also provided evidence for absolute fidelity in copying a large gene segment for 6 years and a shorter segment for 26 years. These data raised serious questions that attributed rapid influenza evolution to random mutation

I think this conclusion is premature. I would like more details from Drs. Webster and Olsen on how their samples were handled and whether the 1977 isolate was ever used to infect experimental animals.

Montreme, There are TWO 1977 isolates (Tennessee/24 and Tennessee/26) in addition to the 1998 isoalte from North Carolina (all have extensive matches) and all of these isolates are in the wild (with many sequences of related isolates in the database).

I don’t doubt the Tenessee isolates are from the wild. I’d still like to know whether these strains were ever sent to Dr. Olsen or whether they were ever used to infect experimental animals.

Hi Monotreme. I just wondered, as time has gone on, what is your preliminary gut feeling on how this is happening?

Have you had any reason to change your initial thinking?

Monotreme – at 10:16

The ones that we are chasing now are just pixels in the painting.

Niman has produced and will likely show us again clear examples of what we are calling these phenomena.

Monotreme, I know that you asked me a question in the influenza genome project thread. But I’m not sure I can answer. Right now after the conversations with Taubenberger the other day, what little there was of my virology is all up for re-evaluation. :-) So you guys carry on and I will abstain for now.

<anon goes into cave to meditate on “pixels in the painting”>

Anon22

The pixels are probably vibrating with the Mastiff’s snoring.

bump

Put the fluwikie t-shirt over your head and hope for the best. Good luck, and Good night.

bump

bump

bump2

Hi Tom DVM, No, my thinking hasn’t changed. I don’t understand how a polymerase could possibly so accurate for 30 years. Yet, the identity relationships that Dr. Niman reports are accurate. I don’t think we can go further in resolving this paradox without input from the people who handled the virus isolates and the sequences. We need to find a way to get them to address the issue.

anon_22, this is pretty simple. Dr. Niman has found a long stretch of nucleotides that are identical in isolates that were supposedly obtained 30 years apart. If everything is as the submitters of these sequences say it is, then we would have to conclude that Dr. Niman is correct and all the flu experts are wrong - flu polymerase complex is the most accurate polymerase on earth. I find this extremely hard to swallow for reasons I have given previously. The only other explanations I can think of involve human error - of some sort. Hence, we need more details from Drs. Webster and Olsen. I

This may not be the right place. I apologize ahead of time, if I interrupt the thread. I’m wondering how these fit into this picture. I don’t understand what I’m reading, but the first one (AAB29507) is linked to Scotland 1959.

AAB29507 566 Avian 4 (HA) H5N1 United Kingdom 1991 Influenza A virus (A/turkey/England/50–92/91(H5N1))

 AAC40503 230 Avian 8 (NS) H5N1 United Kingdom 1991 Influenza A virus (A/turkey/England/50–92/91(H5N1))     
 AAC40504 121 Avian 8 (NS) H5N1 United Kingdom 1991 Influenza A virus (A/turkey/England/50–92/91(H5N1)) 

Text of a telex received on 21 January 1992 from Dr K.C. MELDRUM, Chief Veterinary Officer,Ministry of Agriculture, Fisheries and Food, Surbiton:S. R. - 1 Nature of diagnosis: clinical, post mortem, laboratory (virus isolation). Date of initial detection of animal health incident: 15 January 1992. Estimated date of first infection: 18 December 1991. Number of separate outbreaks identified so far: one (1). Geographical identification of the outbreak: Weston Longville, Broadland, Norfolk county.

Details concerning the outbreak:

No. Species No. of animals in the outbreak No. of cases

1 avi 7,753 …

No. of deaths No. of animals destroyed No. of animals slaughtered

7,129 624 0

Comments concerning affected population: fattening turkeys. Comments to date concerning epidemiology of the disease: first deaths occurred on 18 December 1991. Poisoning was initially suspected. No deaths since 24 December 1991. H5N1 virus isolated on 15 January 1992 with intravenous pathogenicity index of 3.0. Investigations are continuing. Control measures taken to date: All poultry houses with 3 km have been placed under movement restrictions and a 10 km infected area has been imposed from 20 January 1992 at midnight.

glo, I did a BLAST search using the 2 available sequences from A/turkey/England/50–92/91. I did not see any evidence of the long stretches of identity that Dr. Niman has found in his examples.

monotreme, how could you have missed the whole issue, when niman first brought it up here in Feb. or March ?

anonymous, I did not miss the issue, but I had not focused on it until he pointed out that an entire gene was identical in 2 sequences isolated 2 years apart. One of those sequences was subsequently retracted from GenBank, which is very unusual. That prompted me to look at some of his other examples.

I don’t really see the relationship of glo’s comment on the homologies?

Monotreme,

Please look at some of the others that Niman has posted and remember that he’s agreed that most of the identities are not as long as the ones retracted.

Your keen mind will identify the pattern if there is one.

I think, perhaps, that we’ve been staring at the pixels and missing the Picasso?

NS1, this is science, not art. I don’t think there is any Picasso involved. Like your speech is sometimes too artificial for me…

glo is supposed to clarify his question. Did H5N1 in 1959 evolve from H2N1 in 1957 , BTW. ?

anonymous, please choose a name and answer NS1′s question directly.

Melanie, please add a page about the privacy-rules of the fluwikie. I couldn’t find one. Very unusual.

To whom it may concern (anonymous): It would simply be another tool that could help to identify your comments and clarify issues with out the confusion of another anon poster who could easily mis-represent your thread of thought.

Choosing a name would also just be a bit more congenial.

it has proven non-suitable. People are being judged just by their posts to other threads rather than by the content of the actual posts. Choosing a name will cause fluwikie to create a cookie on your computer and store there data from your fluwikie visits, and maybe other things, who knows. Choosing a name also claims for some expertise. Your opinions could be taken as illegal medical or juridical advice. This is weakened when you classify your posts as anonymous. We talked about this before and it turned out that some people are only here because they can stay anonymous and don’t have to register as is required on most other boards.

Point taken.

Melanie, can you help?

Mr White42 Maybe gs is right…Maybe we should all go anonymous.

anonymous – at 09:56

Nonetheless, moving forward, registration is likely to be implemented. It is the only way to manage the forum, as just about every database file program requires registration. it also prevents name hijacking, as well as enhance the ability to search previous comments. Might as well bite the bullet and get used to it.

As far as anonymous’ concern about behavior on other boards, there’s truth to it. Some people are so obnoxious that their screen name careens from one flu board to the next, spoken only in whispers lest they might be invoked. The logical solution, of course, is to pick a different screen name when you register if you are such an individual. There is no database of “all flu boards” that share info.

anonymous – at 05:47

we are working on an FAQ and it will be included there. We do not collect or share personal information. We do read other flu boards from time to time.

“if you are such an individual.”(nice disclaimer.)

Thanx, Dem. you too, Tom.

Each and every “anonymous” poster is corrosive to communication. (Okay now, who are we talking to here? Whose syntax is that? etc.) This irritant has not crashed communication among us because such posters are a rarity. I can see such behavior as legitimate when a new poster is “testing the waters” and is seeing if he wants to engage in an extended conversation. But otherwise, “anonymous” posters are claiming special status. If it stays limited to gs and “fly by” posters, I guess we will get by.

Hi Tom, Personally, I think having a handle gives this site personality. I also believe that is why so many are drawn to this site. It doesn’t hurt to have an extra security. Especially now! gina

Opps. “It doesn’t hurt to have an extra security blanket.”

LOL!

Can you imagine anonymously talking to Tom DVM, monotreme, NS1 and “currently anonymous on this thread” simultaneously about the accuracy of flu plymerase and then having the anonymous poster from recombinomics drop in?

For us veterans, we could tell who is who (we all have our unique style). But for new people, it’s be a blur of information and confusing as hell. Tom DVM was being tongue in cheek, and it was pretty funny.

Anyway, back to science…

DemFromCT, Really, I did know. Yes! The dim blond strikes again. Oh, thank for an extra LOL. gina

but that’s just an assumption and can’t be used in court.
It could be someone mimicking anonymous.
No name just means no name.


And now imagine this anonymous from CT running for Governor in 2016 and some Journalist picking up some silly post of him from 2006 and publishing it in a newspaper…

anonymous 11.55.

“And now imagine this anonymous from CT running for Governor in 2016…”

Anonymous, which anonymous were you talking about?

If anyone isn’t convinced about the wisdom of registration after this exchange, there’s no convincing you.

the DemFromCT. That’s for “democrate from CT”. “Democrate” is one of the two major parties over there in USA. CT is a state in USA.

Mr White42 – at 10:17

And, what, exactly does my name tell anyone about my “expertise?” That’s a specious argument.

Let’s all be anonymous for a while to show what it would be like. Your visit is already tracked on the site meter. You can look under “who’s on” and find your computer server and location listed. Having a handle will make no difference to your anonymity. Pogge could identify the different anonymouses by the IP number, if there was a troll problem. But most of the time, I’m sure he does not care to keep track of everyone’s IP number. But if you are here, and not even posting, you are on the site meter list. I don’t see how calling yourself “VIP” or whatever, will make any difference to your privacy.

Hey. This is fun!!

Who said that?

I said that-

Come on in everybody…the water is fine.

Oh Shoot That wasn’t me because I am now captain anonymous, the caped crusader.

Hey gs, I think you are really on to something here!!

anonymous at 13:03 - good try! LOL anonymous at 12:58 - Have I ever told you how much I admire your mind?

Now I KNOW. Lily, I have experienced hive-mind in realtime!

Hey, who turned the lights off? Someone call Eccles. We got to recharge the windup for this thread. Oops, need some baling wire and duct tape for this broken piece here.

Whir, whir, whir….there we go.

Lights are back on. This thread will now return to its original purpose.

Coming soon: “Deep science That I Do Not Understand Relating To The Accuracy Of Flu Polymerase”

I am one of many anonymous Influenza Polymerase proteins vying for your assistance in replicative matters.

Please remain anonymous so that I may continue to freely, accurately or not, replicate in your anonymous brothers and sisters around the world without hindrence.

Thank you for your anonymous assistance.

I assume that one expression of thank you is enough to cover this tapestry of anonymity.

All anonymous means is that you have to judge the post by its content rather than by who wrote it. We should do this anyway, but I’ve noticed that some people in other threads have difficutly. Because of this, I say we should all be anonymous.

But you are really not anonymous, anonymous.

…sooo funny and rediculous and deathly serious at once! I love this place! {BUMP}

are there maybe some factors to influence the accuracy of flu-polymerase which are subject to mutation and selection themselves ? Would make sense. If I were a virus, I would do it that way.

anonymous. I am going to go out on a limb and say that you are not my friend gs so first, please pick a name and jump in…we could use the help.

Second, your comment makes sense and you should restate it as a new thread. I would do it but I am computer semi-literate.

He-Who-Must-Not-Be-Named:

It’s safe to assume that every locus on every genome is subject to mutation. Even the genome of a hypothetical organism whose replicative machinery featured perfect copying fidelity would be subject to random mutation. The genomes of “higher” organisms typically include large portions which are not subject to selection, by virtue of being untranscribable — but viral genomes in general tend to be compact, and uncluttered by such “junk”.

The polymerase complex portion of the viral genome consists of three subunits: polymerase basic protein 1 (PB1), polymerase basic protein 2 (PB2), and polymerase acidic protein (PA). I think it would be hard to make a case for these being either exempt from mutation or invisible to selection, and theoretically navigating the potential consequences of changes to these proteins seems almost hopelessly ambitious even before we take into account that in order to do its work, the polymerase must interact with the host nuclear structure.

There doesn’t seem to be any obvious reason not to start with the simplest of logic. Say you had a really sloppy polymerase. This would mean lots of errors (“mutations”, by one definition) during transcription and replication, including errors in the polymerase complex itself — leading to virions with even sloppier polymerases, which, when their time came, would produce even sloppier copies, etc. The sloppiest of these would leave no viable progeny at all. But among all those sloppy copies, a few might emerge with polymerase even more accurate than the previous generation — so if the virus is to achieve better accuracy in replication, the way it does that is through poorer accuracy in replication.

For an organism that depends so heavily as does influenza on the ability to adapt quickly to new hosts and to flank their immune defenses, very high fidelity is as much a dead end as very low fidelity. What we would expect is a “sweet spot”; a range within which an optimal balance is stuck between fidelity good enough to produce viable progeny, but not so perfect as to limit opportunities for aquiring new changes.

This is not a problem except for these unexplained instances of apparently near-perfect conservation (and discontinuous distribution) of lengthy sequences over such long periods of time.

racter—There is another way, and that is to think of a virus as an ant, and the H5N1 as an ant colony. While I agree that what you describe is a good mechanism for =generating= diversity, there is another way to acquire changes, and that is to store them offsite. It’s much faster to conserve good bits of code, subroutines if you will, by scattering them far and wide, and using dual infections to swap them in and out. This is an efficient way to recycle the good bits, and shuffle out the worthless ones. Is there a way to decide whether a given change is due to mutation or replacement? High fidelity isn’t necessarily a bad thing if you can trade to get the information you want. Thus the totality of genetic information resides in the colony, not in the ant.

WetDirt:

Yes, I see your point. It’s also easy to see how this whole business tends to be something of a teleological minefield. It’s tempting to view template-switching as essentially a malfunction in the system, but if it does provide the virus with a shortcut for the aquisition of change, it might be “not a bug, but a feature”, and the underlying mechanisms could be positively selected for. We’re clearly dealing with massive complexity here.

More conclusively in vitro than in vivo, as we’ve observed.

wetDirt – at 12:43

“Thus the totality of genetic information resides in the colony, not in the ant.”

In reference to the above statement, I couldn’t help but think of the FluWiki Forum. A diverse, messy bunch of ideas and thoughts put forth by many individuals in the quest for “truth” and a goal of survival. No one individual has all the information (for survival), and the discussion frequently goes off in wild directions (like now), but in the end the group has a greater approximation of the truth about H5N1 and a greater gathering of information for survival.

wetDirt – at 12:43

Promiscuous and fluid gene pool . . .

now widen the mound to a group of interconnected mounds (each subtype) with fingers into smaller mounds representing sharable information from other organisms via secondary acquisition methods.

You’ve placed in analogy what I’ve been trying to communicate for months.

• Autonomous systems are fault-tolerant by design.
• Small genomes must have a backup system, a method of versioning.
• The backup system is the hive.
• Proximity is the method of restoration or acquisition.
• Proximity is assured due to circulation or connectivity.
• A set of hueristics (pressures) determines the acquisition / expression of genetic material.
• Although true random mutation does occur, the primary mode of trade is along the spice trails (established paths).

Now we have to find the rules and the established paths.

Please ask Niman.

Thank God.

Perhaps our taxonomies have been too narrow for these small fellows.

What if we discover that more of the gene pool is accessible by the average organism than we ever realised?

The recent report of a mutation in an H5N1 isolate from one of the members of the Karo cluster would seem to argue against the perfect polymerase hypothesis.

only one example… Sometimes it’s perfect, sometimes not. See the Dutch case with H7N7, lots of mutations.

anonymouse, there are many examples of mutation in vitro, and some additional examples in vivo.

It makes no sense that the polymerase could work perfectly for 30 years with one strain, but start making mistakes within a few days with another strain.

M-treme, It make no sense, but could it be the case? Mistakes co-mingled with another strain?

why are there zero mutations in some areas, at some positions of the genes, 3rd bases, and lots of mutations on other positions ? Apparantly the polymerase is always perfect at some positions, we see no mutations, not even those which don’t change the amino-acid.

Anonymous at 12:22. Would it be possible to rephrase your last sentence. Not sure what you mean, there’s a double negative. Your thought is important.

at the 3rd base in a codon you can exchange A<→G or C<→T without changing the amino-acid and thus there should be little selective pressure against such a mutation. Indeed, about 70% of mutations are at the 3rd base. But codons at some positions just won’t seem to change, not even this way. So, when you compare all the sequences in the database then there are quite some positions where all sequences coincide. Hmm, I hope this is correct, my last sample was a bit small - I’m running this with 700 H5N1-HA-sequences now, wait some minutes for the result…

732 aligned H5N1-HA-sequences, 800 contiguous nucleotide-positions, about 18million differences(mutations) in total out of 800*732*731/2 possible ones. Differences-rate = 8.5%
17%,15%,68% of the differences occur at base 1,2,3
130 of the 267 3rd-base-positions had zero differences, so all 732 sequences did coincide at these 130 positions ! Almost 50% of the 3rd base positions were always accurately copied by the polymerase. Despite the lack of selective pressure.

uhh, sorry. I had an error. The 130 was for base 1. For base 3 it’s 106. Only 7 for base 2, hmmmm. I hope there isn’t another error.

Anonymous at 14:01. Is there a way to tell if the 50% of accurately-copied 3rd base positions are localized to greater or lesser degree (perhaps in more than one locality), or are they spread out all over the HA sequence(s)?

Thanks for doing this time-consuming analysis.

they are spread. I could send the file.I couldn’t detect a pattern. It could depend on the neighboring nucleotides, so that certain nucleotide-patterns are copied with greater accuracy.


I also tried 384 aligned H5N1-PB2-sequences now. differences are with 9% at base 1 ,19% at base 2,72% at base 3.
63% positions are unchanged for base 1,
23% for base 2,
and 70% for base 3.

Rules-based recombination.

It’s not a case of a perfect or imperfect polymerase.

Polymerase makes copies or recombinants as it is instructed to do. Sometimes long strings of indentity, sometimes random mutation, sometimes single nucleotide recombinations.

correction again and one new dataset.


384 aligned H5Nx PB2 nucleotide sequences.
10.8% , 4.5% , 84.6% of differences are at base 1,2,3
79.4% , 8.9% , 66.8% of the positions at bases 1,2,3 have the same nucleotide for all the 384 sequences


734 aligned H5N1 HA nucleotide sequences
17%,15%,68% of differences are at base 1,2,3
49%,3%,40% of the positions at bases 1,2,3 have the same nucleotide for all the 734 sequences


1024 aligned H3 HA nucleotide-sequences
20.5%,15.1%,64.5% of differences are at base 1,2,3
65.1%, 3.7% , 50.7% of the positions at bases 1,2,3 have the same nucleotide for all the 1024 sequences

Mr. White, the only other mechanism I can think of is latency, which is not supposed to happen with influenza.

anonyomous, some basics of evolution. Mutations occur randomly (more or less). However, some mutations are deleterious to the organism. This is true of virii and people. If the mutation is very deleterious, the organism is at competetive disadvantage and will not produce progeny. Because the genetic cod is degenerate, mutations at the third position are less likely to be deleterious than mutations in the first 2 positions. Therefore, one expects to see more mutations in the 3rd position over time because mutations in this position are less likely to be punished with genetic death. Thus, it is not that the mutations occur more often at the 3rd position, it is that these mutations do not stop organisms with them from reproducing sucessfully. Hope that is clear.

Monotreme – at 22:23 >>>Mr. White, the only other mechanism I can think of is latency, which is not supposed to happen with influenza. >>>

Monotreme: Can you explain what you mean by “latency”? Do you mean a state of being dormant?

petperson, some viruses have a latent phase when they remain in the host but don’t replicate. This can go on for decades before the virus is reactivated. During the latent period, the mutation clock stops. This could explain paradox, but there is no evidence that I am aware of that influenza has a latent phase - it’s always burning. At least that’s the dogma.

If the dogma is incorrect and influenza can in fact be latent in the human body, would testing reveal that? Would that be an asymptomatic individual that tests positive for antibodies? Or would it not be revealed in testing?

petperson, if the person had an active infection and then the virus went into a latent state, then yes serology should reveal that they were infected at one time. The problem would be to prove that the virus was in a latent state in a person. This would be very difficult, I would think. But this is pretty far out of my area.

Hi guys. There is no such thing as a latent influenza infection. The virus either replicates too kill you, replicates to make you damn sick or is taken out by the immune system through identification or senescence.

Monotreme. It still seems that no reasonable explanation has come forward to explain the pattern you observed. Maybe there is only one explanation…the one you started with?

Tom DVM, I would bet you 13 donuts that if you were to ask the average virologist how could you have a recombinant flu virus that had a long segment of sequence that was identical from 30 years ago, he say “That’s easy. The parental strain was stored in a freezer and a recombinant was made in a lab.”

If you then told him the recombinant was found in an animal outside a lab, he’d say “Oh shit!”. He would *not* say, “Everything we know about flu virus mutation rate is wrong - the flu polymerase turns out to be the most accurate polymerase in all of biology. So, no worries about 30 years of perfect replication”.

I want to make clear that I in no way want to discourage Dr. Niman from presenting his viewpoint. He has done important work by documenting the anomalies. But, I think people should have some idea of what mainstream scientists think about flu polymerases and recombinants.

I really wish we could get Dr. Taubenberger to comment on this. His interpretation would be much more authoritative than mine.

First of all your opinion is good enough for me. Second, there is no chance that anything could be that perfect, nature isn’t perfect ever.

But cold animation is pretty darn perfect!!

Sorry, thirdly It has already been demonstrated that Dr. Taubenberger follows flu wiki…so if there was something you have said that he had significant disagreement with, I expect he would already have commented, as he has done in the recent past.

Previously posted in the News Reports June 24 Thread, if it helps to clarify for petperson:

From Veterinary Virology by Fenner et al.

“The replication of RNA is a phenomenon unique to viruses. Transcription requires…an RNA-dependent RNA polymerase, a virus-coded enzyme not found in uninfected cells.

Where the viral RNA is of minus sense(orthomyxoviruses…), the complementary RNA will be of plus sense.

Retroviruses have a genome consisting of plus sense RNA. Unlike other RNA viruses, they replicate via a DNA intermediate.”

Avian Influenza is an orthomyxovirus and would not be using a DNA intermediate, therefore it would NOT be able to integrate into the host DNA and remain dormant there for years.

Hi Tom, The “sleeping dragon” will be awakening soon. I honestly do not believe birds will be our indicator. At least not the feathered type. Maybe a bird with a metal skin?

Have you noticed any unusal bird patterns in your area? gina

Tom DVM, thanks for the vote of confidence. I do wish a real virologist like Dr. Taubenberger would comment on this issue. I don’t find Dr. Niman’s explanations convincing, but I still fear I may be missing something. And if I’m right…

ManyCats, thanks for the citation. So, what do you think?

• A. Pefect polymerase
• B. Frozen lab strain
• C. Other - please explain

Not to put you on the spot or anything :-)

I would LIKE to vote for C (clerical screw-up), but with the NEJM situation now in the mix, B or something equally (or even more) discomforting appears to be creeping out from the mists….

I wish this were a bad movie… :(

Many Cats. Thanks. important reference…pertinent to this discussion.

gina. I agree completely. Yes, I have noticed that the birds are acting more like they do in Aug. before they migrate then like they should and would normally in June. I don’t know why and am wondering how they will act later in the summer or start the fall migration early…not sure at this point.

gharris mentioned to me that there may be a report of a bird testing positive for H5N1 in our area (Ontario, Canada) but it is not confirmed and I would imagine it would be of the low pathogenic variety as occured with testing last fall in Canada. It should be said, as Name said a few days ago, that the numbers of cases took the Canadian Food Inspection Agency completely off gaurd…I don’t think they were expecting it in the concentrations that they found across Canada…however, this is probably normal background scatter.

Many Cats, thanks for voting, although I was hoping someone would convince me I’m wrong.

Tom,

Thanks for the info. normally, we don’t see blackbirds until Sept. and they like to hang ot at the malls. They started showing up in May. This year they are everywhere but the malls. The Robins, Blue Jays and Wood Peckers are nowhere to be seen. Even the Squirrels are missing. Is Mother Nature trying to warn us? gina

Monotreme. I would be very surprised if you were wrong on this…

…I want to know when we are going to start the

     WORMS AT THE World Health Organization Thread.

I notice none of their public relations people have had time to discuss the issue with the informed public on flu wiki.

I have to wonder why. Have we demonstrated that we cannot debate the issues with open minds, fairly and respectfully?

monotreme and others, is there any explanation, why we see no mutations in lots of 3rd-base nucleotides ? I think, this is significant. The polymerase seems to be very exact there.

why can’t we get some experts or students to join here, like we see it in usenet and mailing lists ? Where are they secretly discussing ? There must be some other forums…

It’s not just the polymerase.

do you understand it, is it examined ? Why can short segemnts be replicated exacly but long segments not ?

anonymous. You raise important questions here that hopefully can be answered by NS1, Monotreme and Dr. Niman.

If not you could probably get information from DemFromCT and Melanie and contact the experts directly…Taununburger or Webster etc.

There are also probably many universities with independent experts you could contact.

When you get the answer, please post it so I and others can read it.

Thanks.

Sorry, should have said Dr. Taunburger and Dr. Webster.

….ok guys, if this accuracy is due to a frozen lab strain (with all of its dark implications) then i’m sure that it would be very imprudent for Dr. Taubenberger to comment…i admire Monotremes bravery concerning this issue and i have a gut feeling that he has spotted something that we were not meant to spot….Dr. Niman seems awfully keen to dismiss this line of reasoning (i understand that recombination is as much a business interest as it is a scientific one for him and that all of his pooh-poohing, nay-saying and tinfoil hat wearing bashing may well reflect this….).

If it is a frozen lab strain (escaped or released) then with a currently 80% fatality rate i guess we can call it weaponised at this point…it is said that flu would be the wrong thing to weaponise…it would continue to evolve…would be impossible to control…would be difficult to vaccinate against with modern methods……which of course would make it the perfect thing to weaponise because it is so darned “natural”……who would suspect such a thing….!…if there were (as yet) unheard of methods of prevention or treatment of virii in general then the “weapon” would be perfect…

Which leaves me with a question.

Does tinfoil really work….? ;-)

TomDVM, why do I have to do so much ? I think, I contributed enough to this thread. Someone else please help to search google and papers to find some answers. When I go to universities… you know what happens. Secrecy, papers, delays and clearly _no_ postings to fluwikie. It’s just not supported by universities to do research freely and openly in forums. No payment, no grants no reputation-increase for such internet-research. We are somehow in _competition_ to universities here. They don’t want private research, they don’t want us to figure out things … before they did.

Amouse,we are not talking about H5N1 here, but H1N1-swine virus from 1977

here is a list of the amino-acids, presented in a way to demonstrate how 1st,2nd,3rd base mutations affect the change of amino-acids:


+:A,G
-:C,T
*:A,G,C,T

sort by : base2,base1,base2


Lysine, Lys,K,2, AA+
Asparagine, Asn,N,2, AA-
GlutamicAcid, Glu,E,2, GA+
AsparticAcid, Asp,D,2, GA-
Glutamine, Gln,Q,2, CA+
Histidine, His,H,2, CA-
STOP, Sto,},3, TA+
Tyrosine, Tyr,Y,2, TA-

Arginine, Arg,R,6, AG+
Serine, Ser,S,6, AG-
Glycine, Gly,G,4, GG*
Arginine, Arg,R,6, CG*
STOP, Sto,},3, TGA
Tryptophan, Trp,W,1, TGG
Cysteine, Cys,C,2, TG-

Threonine, Thr,T,4, AC*
Alanine, Ala,A,4, GC*
Proline, Pro,P,4, CC*
Serine, Ser,S,6, TC*

Isoleucine, Ile,I,3, ATA,AT-
Methionine, Met,M,1, ATG
Valine, Val,V,4, GT*
Leucine, Leu,L,6, CT*
Leucine, Leu,L,6, TT+
Phenylalanine,Phe,F,2, TT-\\

unfortunately pmwiki did cut the blanks :-(

….ooops…thank god for that…..!….best cancel my tinfoil order…..!…sorry for being an idiot…! :)

well, now that you say that, I should mention that we were talking about H5N1 earlier. The Chinese sequences Fujian 2001 and Guangdong 2003, but that’s only 2 years, not 26. See the other thread.

….ooops…thank god for that…..!….best cancel my tinfoil order…..!…sorry for being an idiot…! :)

AMouse,

You’re just inspecting the data that you’ve seen.

These are all very real issues and cogent possibilities to be examined. I’m just not certain that the issue is binary - perfect or imperfect polymerase. I am certain that the issue of lengthy identities is multi-factorial.

…sorry for the strange double post (don’t quite know how it happened)….with regard to the fujian and guangdong sequences isn’t two years still a fairly long time for them to remain unchanged according to the traditionalists out there….?

I read a while ago on the wiki that Dr. Niman had been mistaken with regard to his conclusions on SARS…..can anybody be a little more specific about these conclusions and/or predictions because I don’t seem to be able to find any reference to his work on SARS on the recombinomics website…did he talk of the evolution of SARS in the same way that he talks of recombination in H5N1….?

i guess i’ll just keep hold of the tin foil i already have and leave the bulk order for later… ;)

Closed due to length. Conversation is continued here.