1. Patients launch $1.27 million crowdfunding campaign for ME/CFS gut microbiome study.
    Check out the website, Facebook and Twitter. Join in donate and spread the word!
Nitric oxide and its possible implication in ME/CFS (Part 2 of 2)
Andrew Gladman explores the current and historic hypotheses relating to nitric oxide problems in ME/CFS. This second article in a 2-Part series puts nitric oxide under the microscope and explores what it is, what it does and why it is so frequently discussed in the world of ME/CFS....
Discuss the article on the Forums.

New Paper by Dr Ruscetti

Discussion in 'XMRV Research and Replication Studies' started by VillageLife, Jan 4, 2012.

  1. alex3619

    alex3619 Senior Member

    Messages:
    6,848
    Likes:
    10,435
    Logan, Queensland, Australia
    This paper requires three (iirc) recombination events. Each and every one would have to produce viable virii for it to succeed. So two older recombinants should still exist in the samples. Where are they? That is how I interpret it currently anyway. Sure, they might have reduced viability and so be there in low quantity - but they should still be there. The only interpretation that permits them to be absent, as I see it, is they were swamped by XMRV as it is more viable. However, some cells should contain the intermediate steps, though in very low number.

    Bye, Alex
  2. RRM

    RRM

    Messages:
    94
    Likes:
    104
    Asleep: because I am afraid my post will get too long if I address every word, I will stick to what I really see as one repetitive error in your argument.
    You commit a very basic probablilistic error. Instead of explaining again (but see the second part of this reply to Alex, I'll adapt your own example somewhat:

    Suppose we assume that Alice didn't buy a lottery ticket after all and Bob wins the lottery. One day after he returns the ticket, a policeman visits Bob's house. He says:

    "Sorry Sir, but I have to arrest you. The chance of you having won the lottery in a fair way was exactly 1 in 10,000,000. Therefore it's almost certain that one of the other 9,999,999 persons who bought lottery tickets actually won. You apparently stole the lottery ticket or manipulated the draw."

    Yes, I know this is silly. But this policeman could be you.

    Is also similar to certain people arguing that LHO killing JFK must have been a conspiracy because the chances of JFK getting within 100 yards of LHO during JFK's presidency would be incredibly low. Irrespective of what you think of the assassination, it's a silly way of calculating the odds.

    Point is that it was no 'singular possibility' that a certain recombination event of many possible events occured in a certain cell line at a certain moment in time (or that a certain lottery ticket was going to be drawn or that JFK had to drive past this madman's workplace). It only became certain to you after one of many possibilities happened to occur.

    Another aspect (that's not really an independent aspect but is really derived from the above) is that, from all of those many "imaginable pathways, there would also have to be one solution, one event that happened to be the source. For that one source (that is surely equally imaginable), the same would really apply: it would have required the same singular, mutually exclusive (and equallly improbable) event to get in that source and then it still had to infect people and 22Rv1 many years later.

    Therefore, even at face value, without doing any other experiments, the hypothesis that XMRV arose in 22Rv1 is already more likely than the alternative. After all, the alternative would have needed exactly the same kind of "luck" XMRV had in its creation, and then some (e.g. infecting people, staying undetected, infect 22Rv1 many years later and without having evolved over the many years).

    I thought I illustrated how the above is nonsensical with my earlier example about getting identical kids from independent conceptions (and therefore, aside from twins).

    Point is that we know that it (XMRV) happened (at least) once, because it is infectious in 22Rv1. Contamino ergo sum. ;)

    Again, the evidence that it occured in 22Rv1 has little to do with the calculation of it occuring twice and the low odds that it happened twice have little to do with it having happened in 22Rv1.

    It happened in 22Rv1 because there is evidence that it did. For instance, 22Rv1 is the best known environment on the planet for XMRV to exist and multiply which is a pretty solid evolutionary indication, its ancestors happened to "live" in the same environment, the phylogenetic evidence supports the notion, etcetera.
    Firestormm likes this.
  3. RRM

    RRM

    Messages:
    94
    Likes:
    104
    Sorry, apparently missed this earlier.

    Anyway, no other recombination events are 'required' to have happened in 22Rv1, just a single one. Perhaps you could point me to the part from the paper which you think suggests this?
  4. Jemal

    Jemal Senior Member

    Messages:
    1,031
    Likes:
    60
  5. alex3619

    alex3619 Senior Member

    Messages:
    6,848
    Likes:
    10,435
    Logan, Queensland, Australia
    Hi RRM, this is not from the paper but the original conference presentation which was on the net - a retrovirus conference in May ? last year?. The explanation for six template shifts in the virus during replication (along with additional changes required including four substitutions and an insertion) was multiple recombinations - my memory said they were discussing three in the example given at the time, but then again my memory is fubar. That is how it was presented as I recall anyway. Now the paper itself describes it as a single event at a probability of 1 in 10^12. Taken as a single event then intermediaries are unlikely and unnecessary. However, six template shifts in replication, four substitutions, and one insertion, in a single event? I think their original explanation made more sense, but it wouldn't have been as improbable as their final argument I dont think. It also would mean they would have to find the intermediaries, as I said before, to fully justify their claim. Maybe they went with the less likely event because they couldnt find the intermdiaries? I don't know, just suggesting it.

    For those who are interested, the Paprotka paper is here:
    http://www.sciencemag.org/content/333/6038/97.full.pdf

    Bye
    Alex

    PS Does anyone have a handy link to the conference video?
  6. Sam Carter

    Sam Carter Guest

    Messages:
    296
    Likes:
    184
    I think this is a new article, Jemal. Here's the full citation:

    Nat Rev Urol. 2012 Jan 10. doi: 10.1038/nrurol.2011.225. [Epub ahead of print]

    XMRV and prostate cancer-a 'final' perspective.

    Sfanos KS, Aloia AL, De Marzo AM, Rein A.
  7. Firestormm

    Firestormm Senior Member

    Messages:
    5,822
    Likes:
    5,951
    Cornwall England
    Agree. Pubmed have it: http://www.ncbi.nlm.nih.gov/pubmed/22231291

    XMRV and prostate cancer-a 'final' perspective.
    Sfanos KS, Aloia AL, De Marzo AM, Rein A.
    Source

    Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD 21231, USA.
    Abstract

    XMRV was first described in 2006, when it was identified in samples isolated from prostate cancer tissues. However, studies have since shown that XMRV arose in the laboratory and was formed by genetic recombination between two viral genomes carried in the germline DNA of mice used during serial transplantation of the CWR22 prostate cancer xenograft. These new findings strongly imply that XMRV does not circulate in humans, but is only present in the laboratory. Thus, there is no reason to believe that it has any role in the etiology of prostate cancer or other diseases.

    January 10th 2012 apparently.

    I would suggest a new thread. Will try to get hold of the paper but it may be a perspective and not an actual study I suppose. Even so worth a look.
  8. RRM

    RRM

    Messages:
    94
    Likes:
    104
    The video link can be found here. If this direct link doesn't work, you can first go here, then scroll down to about the middle, and play the presentation that says "Oral Abstract: VirusCell Interaction and Co-factors".

    I am afraid your assertion is based on a misinterpretation of the authors findings, however. A single recombination uses a (variable) number of template shifts. The hypothesized recombinant uses six, which falls within the expected range of single-event recombinations. Thus, this finding is really independent evidence in favor of a single recombination event. Also, the "substitutions" you are talking about refer to random mutations that happened after the recombination event. I don't understand how you can see 4 nucleotide differences (and a single insertion) as evidence against the generation of XMRV in 22Rv1 - it would have happened in any other hypothesized recombination event, and the fact that it's only 4 nucleotides is consistent with the idea that the XMRV that's "setttled" in XMRV is extremely close to its source.

    Your "calculation" against this single recombination event happening is wrong anyway, however. Even though my examples don't have the desired effect, I'm giving one more. ;)

    Suppose a lottery draw of 10 numbers between 1-100. Does it make any difference if you take out 10 balls at once, or if you take the balls one by one, waiting one minute (or one day or one year) after each drawn ball? Of course it doesn't - every outcome of 10 disctinct nummers between 1-100 is equally likely to happen either way, and you are equally (un)likely to win this lottery either way.

    The same applies here. I note that we are talking recombination here (gradual mutation is a bit more complicated). The odds of getting to the certain "lottery draw" result that we call XMRV out of the lottery of all possible recombination results are equally likely to happen in one, two or 20 recombination steps. There are in fact two factors that make a multiple-recombination event less likely:

    1) The odds are equal assuming that the number of necessary recombinations between preXMRV-1 and preXMRV-2 (and subsequently between any of the partial recombinants) occur. This is no guarantee and in fact not true (for reasons I won't further go into at this point)

    2) We must be aware that it is the environment that selects if a recombinant (or in fact any mutation) "survives" or dies. We know that XMRV survives in 22Rv1. In fact, it's the best known environment for XMRV to "live" in. Now, any hypothesized partial recombination event should also have had to survive the environment it was generated in. Therefore, whether the partial recombination(s) happened in 22Rv1 or somewhere else, a lot of the partial recombinations that could eventually lead to XMRV would "die" before they could recombinate again. And therefore, because of this, if you find the ancestral sequences of any recombinant and the recombinant is (essentially) a perfect complement of the two ancestral sequences, it is extremely likely that the recombinant was generated by a single event.
  9. natasa778

    natasa778 Senior Member

    Messages:
    1,386
    Likes:
    1,155
    London UK
    RRM, are you are patient? Do you or any family members suffer from neuroimmune disease - is there ME, autism, MS etc in your family?

    just curious
  10. RRM

    RRM

    Messages:
    94
    Likes:
    104
    My sister has been diagnosed with ME/CFS. I also had a (now diseased) aunt that was never diagnosed with autism, but I expect that she'd fall within the spectrum if it had ever been looked into (note that this is more or less a guess based on my limited understanding of autism and her 'behaviour' - I really don't have any expertise in the matter).
  11. Ecoclimber

    Ecoclimber Senior Member

    Messages:
    602
    Likes:
    957
    Mercer Island Wa
    This is a abreviated version of an earlier paper by Sfanos et al., who sequenced the whole EKVX retrovirus from the EKVX lung cancer cell line. The Sfanos paper is listed in the Ruscetti paper, ref. 92 and is discussed on another thread. The EXVX virus is more similar to known xenotropic mouse retroviruses than it is to XMRV, but all of the xenotropic retroviruses show similarities

    Multiple recombinant retroviruses have been observed in human cells grown in nude mice starting in the 1960s. But so far, the recombination event that produced XMRV appears to have happened only once, in precursors of the 22Rv1 prostate cancer cells. XMRV has a unique fingerprint (DNA sequence), much like a human individual, and the chance of it happening again is extremely low.

    Eco
    Firestormm likes this.
  12. alex3619

    alex3619 Senior Member

    Messages:
    6,848
    Likes:
    10,435
    Logan, Queensland, Australia
    Going back to the lottery analogy and how it relates to the Paprotka et. al. paper, its more like this:

    Bob has proceeds from a lottery. We don't know where it came from, maybe he inherited it (his parents are died recently) or maybe he was very lucky and won.

    Searching back through the records we find that ten years ago his parents hadn't won the lottery.

    Calculating the odds that his parents won it, in the manner of Paprotka we conclude that Bob must have won it, the odds of anyone winning it are so low his parents could not have won it. They didn't have the money ten years ago, so its almost impossible it was inherited.

    In mine, I recognize that the odds (in the lottery case) are that his parents won it, so I conclude it is more likely, but not certain, that he inherited it from his parents. In that ten year window, anything could have happened. However I concede that he might have won it himself, I just don't find the evidence compelling enough to be anything like certain of this.

    If Paprotka is right it is likely that a continued failing to find XMRV in other sources over time will steadily increase the strength of the explanation. If Paprotka is wrong then it only takes ONE example of XMRV in an earlier culture or source to put Paprotka in serious doubt. However it is possible that XMRV is much older than Paprotka suggests, but that it did appear spontaneously in the culture anyway. It might be unlikely in this case, but its not impossible - and we don't know all the factors. The probability calculations presume equal probability of events, but I have been saying for a long time now that things are not all equal.

    Bye
    Alex
  13. Ecoclimber

    Ecoclimber Senior Member

    Messages:
    602
    Likes:
    957
    Mercer Island Wa
    I don't understand your analogy nor the logic in post 72. My eyes sort of glazed over reading it but that's probably me as I'm not at my sharpest right now. To me, we are comparing apples with oranges here. Money and other factors as stated in calculating probabilities is different in comparison then to say to wit the world of biology or virology.

    Where I believe your logic is flawed is applying your laws of statistical probability to those fields without considering other factors. Living organisms, molecules, viruses etc are governed by variations and constraints based on various factors such as a set of rules, laws, theories and hypothesis etc that were developed overtime and became observable.

    However, an organism or molecule could variate or be constrained over time and have a variation or a constraint within a variation based on a set of principles. So you would need to calculate those set of principles in the statistical analysis to achieve an accurate probability of your outcome. Now, my eyes are glazing over.

    As stated in post 71, XMRV has a unique fingerprint (DNA sequence), much like a human individual. That fingerprint can be traced backwards in time to present to determine whether XMRV is relevant to any virus variant found. So, the bottom line is that I don't believe it is necessary to get into the law of statistical probability if a virus can be identified by its DNA fingerprint which is easily traceable. I understand your thought processes and welcome them but I think everyone understands the issues and you shouldn't feel the need to justify your position as I believe both positions are quite explainable to the reader.

    Eco
  14. RRM

    RRM

    Messages:
    94
    Likes:
    104
    I didn't use the example to perfectly illustrate the Paprotka study, but to show how the 'hindsight' probability argument used by some of the study's detractors really has no merit.

    The bolded part is where you've made a critical error. Paprotka never calculate the odds of "it" having happened in the 22Rv1 cell line. After they conclude it must have happened in the 22Rv1 cell line, the authors calculate the odds of it happening again independently.

    To explain this entirely in the context of your own example:

    We could conclude that it was Bob who won the lottery on the basis of evidence and inference. Suppose for instance that we find creases in the winning lottery ticket that exactly match a compartment in Bob's wallet, and/or that we have CCTV footage of him buying the winning lottery ticket, and/or a few other independent lines of evidence.

    In any case, after we've concluded that it is Bob who won the lottery, a critic could argue that the fact that I've shown that Bob won the lottery does not exclude the possibilty that his parents won it AS WELL.* It is important to see that we can use a probability analysis in this context, and that Paprotka did exactly do the analysis for this purpose.

    Although it is of course impossible to absolutely disprove that he and his parents both won the lottery, I could use a probability analysis to show that this is extremely unlikely to have happened when the lottery odds are as low as Paprotka's odds. After all, it would be a miracle that we've accidentally stumbled into a family that won the lottery (that has a 1 in 770 billion chance of winning) twice.

    Now, the important thing is this: you cannot use this probability analysis for some kind of "hindsight probability analysis" on Bob's probability itself. You (and others) are/were basically arguing that the calculated odds of winning twice affect Bob's chance of winning it once too, but that is simply not true. We've not stumbled upon Bob by chance, but because he is the owner of lottery proceedings.

    The thing is that the evidence seems extremely compelling, in my opinion. The fact that both preXMRV1 and preXMRV2 were "found" near the (hypothesized) source, the fact that XMRV is extremely well adapted to the hypothesized source, the fact that no strain of mice tested positive for XMRV and the fact that no strain of wild-derived mice tested positive for both preXMRV1 and preXMRV2, is already very convincing evidence, in my opinion.

    In fact, I haven't heard Mikovits or anyone else really argue against 22Rv1 being the origin of XMRV. Even Alter has said that he found the Paprotka findings "very convincing". The only thing I've heard Mikovits arguing, is that it might have happened twice independently, not that the evidence for it happening in 22Rv1 is not convincing.

    (* note that in this regard the example is unable to exactly follow reality because you'd then also have twice the money. To be exact, you'd have to hypothesize that somebody else won the lottery with the exact same numbers, but that would make even less sense for the sake of the example, and the odds are really the same)
    Firestormm likes this.
  15. Firestormm

    Firestormm Senior Member

    Messages:
    5,822
    Likes:
    5,951
    Cornwall England
    And that - obviously - it wasn't XMRV that she totally found of course i.e. she also found something else or at least has reportedly speculated this to be the case.

    Whether or not the 'Lipkin Study' will be enough to show this speculation has merit is another source of speculation.

See more popular forum discussions.

Share This Page