Why does 5AZA matter?

Discussion in 'XMRV Research and Replication Studies' started by Lee, Oct 6, 2011.

Page 2 of 7

  1. Legolas

    Messages:
    51
    Likes:
    32
    Belgium
    I've got some questions. I don't know if all of them are as intelligent, but i give it a try anyway.

    Did VipDx use 5-AZA?

    Is not mentioning 5-AZA one of the reasons why others didn't find the results Lombardi et al did?
    Mikovits said it was because they didn't used the exact same procedure as the WPI lab did. Could 5-AZA shed another light on the negative studies?

    What are the (official) reactions of other (highly respected) researchers in this field ( retrovirology or CFS)?

    Why did Mikovits only recently mentionned 5-AZA? Was there a specific occasion for it?

    Concerning the slides, did she change "2905 PBMC" (without 5-AZA) into a "normal", or was it the other way around? Because the Ottawa slide was more authentic etc..
    Legolas, Oct 7, 2011
    #21

  2. Bob Senior Member

    Messages:
    4,368
    Likes:
    2,069
    South of England, UK
    I'm not sure if we know the answer to that. The details of their testing procedure would be commercially sensitive, and so it might not be public knowledge. I don't know the answer though.

    It could be a factor that explains why the negative studies got different results to Lombardi et al. But until we know the actual details surrounding the 5AZA omission, then we don't know how significant it is. If all of the patient samples were treated with 5AZA, but none of the control samples were, then like Lee says, I think it makes the original study quite meaningless, and it could explain the negative studies. However, it might be the case that 5AZA didn't feature in the study at all, apart from a small number of exploratory samples that weren't significant to the bulk of the study - in which case the new info doesn't change anything.

    We can't know until we know the full details. The new info regarding 5AZA would only make an obvious important difference to the negative studies if both the patient samples and the controls, in Lombardi et al, were treated with 5AZA. If only the patient samples were treated with 5aza, in Lombardi et al, then that info might make a difference to the negative studies, but then they might get equal positive results for the normal controls as patient samples (i.e. it wouldn't be a zero/zero study, but it would be a positive/positive study which would be unhelpful).

    An image came to light that was a Western Blot result from the original study which had "5AZA" written on it. The media interviewed Mikovits and Ruscetti about it, and they said that it was not central to the results of their study.


    I've answered the questions to the best of my knowledge, so I might have some details wrong.
    Bob, Oct 7, 2011
    #22

  3. Lee

    Messages:
    82
    Likes:
    75
    One addition to Bob's answer:

    "Why did Mikovits only recently mentionned 5-AZA? Was there a specific occasion for it?"

    At the Ottowa meeting, JM tried to explain why no one else was finding her retroviruses. She argued that they were integrating and becoming cryptic, but that they could be found if one used a demethylation agent 5AZA. So, in her official version of the story, use of 5AZA was legitimate - checking to see if there were cryptic provirus maintaining the viral infection
    But the get she showed to make that point turned out to be the same as Figure 2C of the Science paper, but with different labels - thus setting off this kerfuffle.
    Lee, Oct 7, 2011
    #23

  4. Deatheye Senior Member

    Messages:
    153
    Likes:
    16
    I thought there are others that Found the Virus family? Also without 5-aza? Not sure if I will Be able to find it again.
    Deatheye, Oct 7, 2011
    #24

  5. slayadragon Senior Member

    Messages:
    953
    Likes:
    112
    twitpic.com/photos/SlayaDragon
    Lee,

    Thanks for explaining the situation in a succinct and clear way.

    Not related specifically to the Mikovits situation, but are there things other than lack of methylation and 5-aza that can make endogenous retroviruses go active? What would those things be?

    Thanks much for your comments.

    Best, Lisa
    slayadragon, Oct 7, 2011
    #25
    Francelle likes this.

  6. alex3619 Senior Member

    Messages:
    4,211
    Likes:
    3,660
    Logan, Queensland, Australia
    Hi Lee, my position is basically the same as Bob. First let me give a caveat that I was not at the Ottawa conference so do not have the full details of what was presented, especially the spoken commentary.

    I agree with the technical side of your argument Lee, the broad implication of your technical analysis was obvious from the beginning and I had already considered it.

    I disagree with your non-technical analysis. My analysis is closer to Bob's position.

    In exploratory science many different tests will be run, some of which will be put aside as not relevent. From the facts I have at present, only two samples, ever, were treated with 5AZA. If you have evidence of the contrary I would like to hear it. Has Ruscetti or anyone else admitted it was routinely used?

    If 5AZA was a standard technique then your analysis is correct. If it was a small exploratory study that was mentioned at Ottawa to make a point, then your analysis is incorrect. We are dealing with cutting edge exploratory science, this is not the same as routine lab testing. Many experimental tests may be tried and most discarded. Should scientists have to list every test and method they have tried even once and never used again?

    We do not have the facts to decide which, it is only speculation at this point. Science is investigating, lets allow that to proceed. If there is anything to the allegation we will hear that in time.

    However, even if the WPI team failed to mention this technique and used it routinely it does not follow that Lo and Alter or anybody else used it. We don't know that. Contrary to media opinion there are many labs that have reported finding MLLVs in patient samples, its just not being published. If they only found MLLVs after using 5AZA this will come out in time. It also does not explain the antibody data though.

    On the gag proteins it is possible to determine what they are. The bands can be excised and sequenced. Their protein sequence will probably determine if they are endogenous or not, but it depends on what they were. It is possible that they could match an endogenous retrovirus and still be from an infective retrovirus, but in this case additional evidence would be required to prove that.

    One of the things I find very frustrating about the WPI research is that they failed to sequence enough findings. I understand that they were under resource constraints, which is much worse in ME or CFS research than is normal in other fields due to minuscule budgets, but many of the issues which are plaguing them now could have been resolved if more sequencing had occurred. In-house sample blinding would have been good too.

    Your anger over this is appropriate if its used to insist on an investigation. Your conclusion is not appropriate until after we evidence that things are exactly as you claim. So far as I am aware, 5AZA was used only twice in a minor experiment. Lets wait for the investigation.

    Bye
    Alex
    alex3619, Oct 7, 2011
    #26
    jace likes this.

  7. leela Slow But Hopeful

    Messages:
    1,568
    Likes:
    909
    Couchland, USA
    Okay, non-scientist over here just wondering out loud if there is any connection between the fact that most (if not all) PWMEs are poor methylators, and could that be why RVs and viruses in general express more in us than in the general population? Maybe the 5AZA thing
    brings Rich VanKs theory in through a new door, if demethylating is a similar enough condition to undermethylating.
    leela, Oct 7, 2011
    #27

  8. Lee

    Messages:
    82
    Likes:
    75
    Alex:
    "In exploratory science many different tests will be run, some of which will be put aside as not relevent. From the facts I have at present, only two samples, ever, were treated with 5AZA. If you have evidence of the contrary I would like to hear it. Has Ruscetti or anyone else admitted it was routinely used?"

    I would agree, but:

    Sure ,people do failed, or exploratory, or side-project experiments all the time. If the gel in question had been an exploratory experiment, left to languish in a lab notebook until they realized it told them something - no problem.
    But it wasn't.

    We now know that, in the Science paper, this was THE gel that showed us gag expression in patient PBMCs. It is THE DATA THEY SHOWED. That makes it not minor - it is the actual data they used.

    And what it actually showed is gag expression induced with 5AZA, but they didn't tell us that.

    The fact that they used 5AZA here without telling us, means we can not trust that they didn't use it everywhere without telling us. It mens we have no reason to believe anything in that paper, pending even more work by other people, trustworthy people, cleaning up after their fiasco.

    My point isn't that the rest of the methods and data is or is not what they claim. it might be, it might not - the investigation will hopefully tell us that.

    My point is that we have no reason now to trust what they've told us - we don't know what they did - because where we've been able to verify, we know that what they told us is either wrong (the PCR results) or not what they did (the western blot). We therefore have no reason to believe that any of the rest of that paper is valid.

    My personal suspicion is that everything remaining in that paper is an artifact of 5AZA - it would certainly explain all the negative followup results. But I don't and cant know that yet. What I do know is that nothing they said in that paper can be trusted - right or wrong, it is now worthless - because they have admitted to telling us things that just ain't so.
    Lee, Oct 7, 2011
    #28

  9. gu3vara Senior Member

    Messages:
    330
    Likes:
    37
    That's the first thing I thought also, don't know if it makes any sense scientifically, sure sounds plausible to me!
    gu3vara, Oct 7, 2011
    #29

  10. alex3619 Senior Member

    Messages:
    4,211
    Likes:
    3,660
    Logan, Queensland, Australia
    methylation and retroviruses

    For the most part low methylation capacity might not make a difference. However in any rapidly growing tissue it could be critical. Tissues that grow rapidly include the gut wall and activated B cells. So this raises two possibilities: one that viral replication via clonal expansion or routine cell division could result in active virus; two that endogenous retroviruses as Lee keeps saying could be activated.

    This is interesting at this time because new evidence supports earlier hypotheses that ME symptoms are directly correlated to, and might be caused by, translocated LPS from the gut. (LPS lipopolysaccharides) (See my Maes thread)

    If something compromises the gut wall, this could lead to this problem. Gut wall is rapidly replicating to replace damaged tissue. However, this does not resolve the question of whether or not its a virus or an activated endogenous retrovirus. This is because viral proteins might be present and make the gut wall a target of the immune system, leading to damaged capacity. This is complicated by the fact that the gut has its own unique immune balances different from much of the body and which I do not really understand. For example, the gut has the capacity to suppress many immune responses.

    It also does not escape my notice that up to 80+% of patients have large enteroviral presence in gut, with up to 50% of cells being infected.

    Bye
    Alex
    alex3619, Oct 7, 2011
    #30
    Glynis Steele and anne_likes_red like this.

  11. Mark Moderation Team Lead

    Messages:
    3,818
    Likes:
    1,233
    Sofa, UK
    Some observations on why it is sensible to wait for more complete contextual information and not get carried away with all the wild and aggressive speculation out there about what these discrepancies mean for the validity of HGRV research and the integrity of Mikovits, Ruscetti and others...

    - It is perhaps highly relevant to remember some of the history of how Lombardi et al was reviewed by Science. The paper went through a very extensive review process over (as I recall) nearly a year from the time when it was first presented for publication. According to what I read, it was cut down to about 1/3 of its original length, and many details which the researchers wanted to include were omitted at the request of Science. In that context, omission of what now seems to be key information may be more understandable and excusable.

    - The slides in figure 2c of Lombardi et al only refer to a very small subset of the samples tested. Many tests were run by Mikovits et al over the course of several years. There's no reason at this stage to assume that the selective application of 5AZA to patient samples featured in any of the other tests - in fact it would be so ridiculous if it did that this seems quite implausible.

    - Mikovits et al have previously stated that the core of the experiments they conducted were blinded and coded, and that the tests on controls and patient samples were treated in exactly the same way. Although it now appears this didn't apply to the specific slide referred to in figure 2c, there's no reason (that I'm aware of) to assume that the 5AZA treatment was applied selectively to the rest of their tests. It is an extrapolation to suggest that it did, based on insufficient evidence.

    - Indeed, since Ruscetti says that the failure to mention the 5AZA treatment of patient samples in Figure 2c was a trivial omission not necessary to the paper, and since Mikovits has asserted that the bulk of samples were treated exactly the same for patients and for controls, the most reasonable way to reconcile all these statements is that Figure 2c was included in the paper as the most clearly-defined image from their various experiments, because it best illustrated the contrast. That may seem a little misleading, but in the wider context of all the materials the researchers had at their disposal, and in the context of the review process re-editing the original paper down to a fraction of its original length, it seems to me most likely that they just presented the case in the strongest possible way by selecting the strongest image. If other slides which did not depend on the use of 5AZA, or where 5AZA was applied to both patient and control samples, showed similar contrasts, then this choice may not have been particularly misleading.

    - At this point, trust in the researchers and references to their professional reputations and their career history are relevant because the attempt is now being made to destroy the reputations and careers of these scientists based on little more than this single omission in the labelling of one figure in the Science paper. As Lee has noted, the publication process does always critically depend on trust in researchers; like any other researchers it would have been possible for them to just make the whole thing up, and when one reads a paper, one really has no way to prove that the scientists involved really did what they say they did. This applies to pretty much any paper. Insinuating misconduct and claiming a wholesale loss of trust in such experienced and high-profile researchers, based on a single discrepancy, before the researchers have had adequate opportunity to fully explain their side of the story...this is harsh and premature, to say the least.

    - A tangential point: failures in the methylation process which Lee mentioned have for many years been posited as central to ME/CFS, and treatments exist specifically aimed at unblocking the methylation cycle. Those treatments are not yet widely accepted scientifically, but they worked out just fine for me and a friend of mine, both of us experiencing much improved health after decades of illness and ineffective treatments. The wider context of ME/CFS science, which fits so well with many aspects of the XMRV/HGRV hypothesis, is relevant background knowledge which many scientists from outside the ME community are unaware of.

    - Finally: IMO the biggest issue that has caused so much consternation on this forum in recent days is the readiness of sceptics to selectively attack specific types of researchers, but not others, at the slightest pretext. This is a phenomenon we have all observed throughout the XMRV saga, and beyond it. If only it were true that the following principle were even-handedly applied:

    If only it were true that such behaviour is universally a 'career-ender', then why have Wessely and Chalder not been consigned to the wastebasket because of misrepresentations like this one?
    http://www.bmj.com/content/320/7233/515.2.full
    "Somewhere between the analysis and the printed copy we have been attacked by gremlins." - and that was all they needed to say. Where was the outrage and criticism from sceptics about this? What happened to the peer review process on this paper, allowing obviously false and statistically impossible results to pass through unchallenged?

    Where was the censure from the sceptic community when Myra McClure claimed "1000% certainty" or when Reeves and Wessely spoke of their "expectation" that the Lombardi results would (should?) not be replicated, before a single attempt had been made to investigate the findings?

    And where was the intense scrutiny and criticism of the PACE trial from the scientific community? No, it is not irrelevant to the issues at hand to highlight the double-standards applied here, certainly not, when some people's reputations are being impugned and some researchers are once again being scared away from researching ME/CFS while others continue to get away unchallenged. It is fundamentally important to notice that some science is waved through, while other science - highly inconvenient and controversial science - is bombarded with vicious attacks until the researchers crack. That kind of distortion of the process is not scientific at all.

    The PACE trial started out by promising to use pedometers to objectively measure the actual changes in participants' activity levels. Measurements of activity levels at baseline were taken, but part-way through the trial, research was published by psychologists elsewhere in which the pedometer evidence clearly showed that, although patients self-reported increased activity levels after CBT and GET, in actual fact their activity levels had slightly decreased. The PACE trial promptly ditched the use of pedometers part way through - changing their experimental methods halfway, and at a time when they almost certainly had data indicating whether this, the only objective measure in their experimental design, would show what they wanted the trial to show. The reason they gave for this? It wasn't fair to ask the participants to wear these wrist-watch-sized devices after everything they had been through in the trial.

    Was this a credible explanation? Was this good practice, changing the agreed experimental design partway through? Does this behaviour qualify as deception, even as fraud? None of these questions, of course, were asked by the peer reviewers or sceptics about this particular research. Nor were the many other serious flaws in the paper addressed by these people.

    Wessely, Chalder, and the PACE trial authors have been cut the most enormous amount of slack by the scientific world, so it would only be fair to apply similar caution before calling for the heads of researchers with decades of experience based on a single mislabeled slide. As Lee put it:

    That should read "Had to have known this". Or, rather, "must have known this". But it isn't fair or reasonable to call somebody out or undermine their entire reputation based on simple mistakes...

    What Lee set out above is really basic virology, and it was a well set out and useful summary of the technical issues involved here. But as others here have noted, most of those of us here who have been following the XMRV saga are well aware of all this background information by now, and as Lee notes, Ruscetti and Mikovits are surely well aware of all these issues too, and much more besides. Nobody expects their science to be accepted based purely on an appeal to their authority. But their credentials, and the fact that they continue to stake their reputations and stand by their research - based on greater knowledge of the science than any of us here, and based on their unpublished as well as published data - do strongly suggest that there is something in this science that is still worthy of further investigation.

    Whether that something eventually turns out to be XMRV, HGRVs, or ERV activation, few ME/CFS patients would really care either way. What should concern us all, though, is that this promising line of research should be allowed to run its course, and the different results found by Lombardi et al in ME/CFS patients vs controls should be fully and completely explained. A premature end to the research, and the destruction of the career of yet more researchers who showed an interest in ME patients, would do nothing to diminish the anger, distrust and sense of alienation felt by large sections of the ME patient community - and such an outcome would be in nobody's best interests in the long run.
    Mark, Oct 7, 2011
    #31
    Dainty, currer, Ash and 9 others like this.

  12. Wayne Senior Member

    Messages:
    1,723
    Likes:
    675
    Ashland, Oregon
    Excellent post Mark. Puts a number of things in perspective, something I wish some of JM's detractors would make a little more effort at.
    Wayne, Oct 7, 2011
    #32

  13. Angela Kennedy *****

    Messages:
    1,026
    Likes:
    145
    Essex, UK
    Yes, thank you Mark. Excellent comments. Absolutely key issues highlighted there.
    Angela Kennedy, Oct 7, 2011
    #33

  14. Angela Kennedy *****

    Messages:
    1,026
    Likes:
    145
    Essex, UK
    And THERE it is, the elephant in the room. Lee has done a great job spouting out techinical details (not a crime!). It's the unsafe prejudicial value judgements that get tacked onto the end of these long technical explanations that people have to watch for, which show Lee is here to push an agenda: and that is to claim Mikovits and her colleagues as disreputable.

    Those of you not technically au fait with retrovirology (and that will be many of you- indeed us! ;) ) - please don't be sidetracked by the technical jargon that's being used. Look for the value judgements and the good faith or bad faith indications nestling in the technical comments. You might not know how to change a tyre or do plumbing - but it doesn't mean you would accept without demur any old flannel from these 'experts' either. :) No offence to to those tradesfolk by the way - they mostly acknowledge that reasonably.

    There are important methodological and epistemological issues in science and academic practice which are being ridden over roughshod here in the race to discredit Mikovits et al. The techies are unaware of these it appears, and this is where they are showing their naivite (along with the bad faith).
    Angela Kennedy, Oct 7, 2011
    #34

  15. Firestormm Senior Member

    Messages:
    2,126
    Likes:
    961
    Cornwall England
    Thanks Lee. This is the real point about the Ottawa conference not the context in which the slide was used (though that was important) but in terms of the Lombardi et al. paper as published, the Ottawa conference revealed the original film of the Western Blot experiment and the use of the 5-AZA.

    As Stoye said:

    I have a copy of that paper on my desk, says Stoye. It has a line through Fig 1. It has a line through Fig 2c. Where is the next line going to be? Is there going to be anything left?

    It isn't so much about how this might undermine the original claims of an association between those with my condition and XMRV - but that it introduces further doubt in terms of reliability (Request for voluntary retraction, partial retraction, continuing review by Science and now an investigation in light of the above).

    Fire :cool:
    Firestormm, Oct 7, 2011
    #35

  16. Esther12 Senior Member

    Messages:
    4,024
    Likes:
    1,852
    Nah - that's a poor example. It was an error, but not an important one for supporting their argument, and one that was admitted as soon as it was pointed out. It looks as if originally all of the data was correct, then the scoring method was changed, so some figures on a graph became inaccurate (ugh at the paper though. These 'send out the questionnaires' CFS papers, and they way they end up being cited to make really broad and tenuous points, are a bit of a bug bear. People who do a lot feel more tired? There's a moderate correlation between psychological distress and feeling of fatigue? These are findings that should guide the way we treat CFS for the next two decades!)

    I don't think it's fair to say that people shouldn't talk about their expectations either. McClure's 1000% thing was poor politics, and given her belief at the time that PC was related to XMRV, didn't really make sense, but no-one could take literally the claim that she was 1000% certain about anything.

    Only some of the things around PACE look likely to have been intentionally misleading.

    (Actually - maybe I misunderstood you, and you were just making fun of the claim that any researcher saying something not true about their work would have their career ended by pointing out trivial examples of this? I do have bonus fatigue at the mo, and may not be reading things properly. Sorry if I misunderstood).

    We need clarification on this point. Were all the controls and patient samples treated the same from the point of blinding? I don't think anyone would claim that 5AZA added only to patient samples would not matter, and it would be a crazy thing for Ruscetti to have done and not reported. Given the trouble there has been with replication though, it does seem strange that 5AZA may have been a key part of the testing used, and yet not mentioned in Science, or in their later paper on methods. (I'm starting to ramble...)
    Esther12, Oct 7, 2011
    #36
    Firestormm and Sam Carter like this.

  17. Angela Kennedy *****

    Messages:
    1,026
    Likes:
    145
    Essex, UK
    It introduces no more doubt than any area of science where there are technical disagreements.

    Again - Lee's argument is not set in stone and irrefutable. Neither is Coffin's. Why isn't Lee using his own name here and arguing this somewhere more official, for example? If he is so confident of his own argument.

    I contend that Lee's own argument is tentative at best. That some of you here find it convincing is a little odd. I don't. And it cannot be excised from the agenda-driven value judgements Lee makes with all his technical details. And it doesn't take into account all the other issues - technical and methodological, around this whole subject, that render his singular argument itself only partial at best, and not comprehensive enough.

    This still takes part in a political context where mud has been thrown at Mikovits from the publication of Lombardi onwards. There has been a marked bad faith in how she has been treated and how the whole retroviral thesis has been treated.

    Now- I'm still waiting for the arbiters of reliability to call for a thorough investigation into PACE with a view to possible retraction or partial retraction. There are serious, multiple discrepancies in methodology there. It does matter because double standards ARE being shown by the various 'outraged' 'bad science debunkers' and 'skeptics', a good few on here, pushing their single-issue 'outrage', in an arena of already multiple exhibitions of bad faith by the retroviral link detractors, that many patients and their supporters are completely wise to.
    Angela Kennedy, Oct 7, 2011
    #37

  18. Angela Kennedy *****

    Messages:
    1,026
    Likes:
    145
    Essex, UK
    Actually, Esther, this was quite a major mistake, which stood for six years without detection, and, to the best of my knowledge, was not corrected at the article's source, even now, meaning people may still be citing incorrect information. It is only your opinion these discrepancies are minor.

    Whatever your opinion, we have multiple sources of clear, serious discrepancies in the PACE trial. This is clear from the threads here and elsewhere. At the very least a thorough investigation by impartial agencies should be undertaken. Unlike the agenda pushers here who want to kill Mikovits career and the possibility of further retroviral link (that is very clear from their value judgements they are expressing, by the way), all those of us seeking investigation of PACE want is for retraction of certain, dangerous claims based on those discrepancies: claims that cause danger to this community! We're not even seeking to end anyone's career!

    The contrast is striking. And we don't even need to use a contrast button.
    Angela Kennedy, Oct 7, 2011
    #38

  19. duckndive

    Messages:
    42
    Likes:
    11
    Hi Lee.
    Thanks for taking the time to explain the situation so patiently. I get now why this is such a big deal to scientists. It's not really about what a WB showed or didn't show, it's about trust in science.

    In terms of where this goes from here, I'm guessing that Science won't just rely on assurances from Mikovits/Ruscetti. I realise this is far from a normal situation but would you normally expect the authors of a paper under investigation to turn their original materials over to the journal - lab notes, gels etc. etc? In this case, these are probably with the WPI? Is it possible/likely that Science might just retract the paper anyway without going back to source materials given the doubts that have been cast. I realise this is speculation, I'm just interested in how this might be brought to a conclusion so that science can move on.
    duckndive, Oct 7, 2011
    #39
    Sam Carter likes this.

  20. omerbasket Senior Member

    Messages:
    510
    Likes:
    13
    That is also what Dr. Mikovits said in the "naturenews" article. They said it was activated, they just didn't think back then that it is worth mentioning by what substance it was activated. Anyway, they wrote that BOTH PBMCS FROM HEALTHY CONTROLS AND FROM PATIENTS were activated. So, unless someone has something that tells us that only the PBMCs from the patients were activated, I don't see how in any way this could be a fraud or something.
    What it is doing, and that is why Dr. Mikovits started talking about 5-AZA, is that it might be causing researchers not to find evidence for HGRVs when they are in fact there. They didn't think about it back then, they do think about it right now, and that is why back than they just wrote "activated" and right now they are also mentioning what it was activated with.

    Anyway, if someone would explain to me why he thinks this is a fraud, it would be nice. Because I don't see why someone would think so.
    omerbasket, Oct 7, 2011
    #40
    jace likes this.
Page 2 of 7

Share This Page