Comparing PreXMRV-2 gag sequence diversity in laboratory and wild mice

asleep

Senior Member
Messages
184
It's awfully hard to fault other groups for not following Dr. Mikovits's methodology when that methodology was not properly described in her published work. For instance, the Lombardi paper makes no mention of the need for multiple PCR runs or that 5-azacytidine was used in the production of the (mislabelled) Western blot.

Again, this seems like a post hoc rationalization to me. I grant that the original paper might be insufficient to reconstruct her precise methods. But unless I'm mistaken, the ultimate goal of science is to move our understanding closer to the truth and not merely to preserve egos and tear each other limb from limb for having the audacity to lack perfect foresight in publishing an initial finding.

Imagine a group that tries to reproduce the Lombardi results and comes up empty. (Perhaps they did their very best to reconstruct precise methods from the published paper, or perhaps--as seemed far more common--they decided to do their own thing, maybe even going so far as to defy basic sensibility by doing something like excluding patients with signs of infection. Either way, it doesn't matter.) This group has a choice. They can a) defiantly publish their results under the presumption that their isolated efforts were sufficient and cannot have overlooked important details known to others or b) email/call/meet Mikovits (or others with requisite expertise and/or success) to see if there might be un-published or unanticipated ways of improving your methodology. In my opinion, one of these is far more conducive to scientific progress while the other, the one we've seen chosen repeatedly, is antagonistic and counterproductive. The only group I recall choosing the latter route is that of Dr. Hanson, who did subsequently improve their ability to find evidence of the virus, albeit inconsistently still.
 

Sam Carter

Guest
Messages
435
Again, this seems like a post hoc rationalization to me. I grant that the original paper might be insufficient to reconstruct her precise methods. But unless I'm mistaken, the ultimate goal of science is to move our understanding closer to the truth and not merely to preserve egos and tear each other limb from limb for having the audacity to lack perfect foresight in publishing an initial finding.

Imagine a group that tries to reproduce the Lombardi results and comes up empty. (Perhaps they did their very best to reconstruct precise methods from the published paper, or perhaps--as seemed far more common--they decided to do their own thing, maybe even going so far as to defy basic sensibility by doing something like excluding patients with signs of infection. Either way, it doesn't matter.) This group has a choice. They can a) defiantly publish their results under the presumption that their isolated efforts were sufficient and cannot have overlooked important details known to others or b) email/call/meet Mikovits (or others with requisite expertise and/or success) to see if there might be un-published or unanticipated ways of improving your methodology. In my opinion, one of these is far more conducive to scientific progress while the other, the one we've seen chosen repeatedly, is antagonistic and counterproductive. The only group I recall choosing the latter route is that of Dr. Hanson, who did subsequently improve their ability to find evidence of the virus, albeit inconsistently still.

From Shin et al 2011:

""""
The protocol for the viral replication assay using spin inoculation was adapted from the one used in the original study that found XMRV in CFS patients (12), with extensive help from Frank Ruscetti (Leukocyte Biology Section, NCI).
"""

W.r.t those patients experiencing apparent persistent viremia, was it not later determined (by Robert Silverman) that their PBMC DNA samples contained VP62 plasmid contamination, thus significantly undermining confidence in this finding?
 

barbc56

Senior Member
Messages
3,657
It's awfully hard to fault other groups for not following Dr. Mikovits's methodology when that methodology was not properly described in her published work. For instance, the Lombardi paper makes no mention of the need for multiple PCR runs or that 5-azacytidine was used in the production of the (mislabelled) Western blot.

Thanks Sam, I feel the same.

Barb C.:>)
 

barbc56

Senior Member
Messages
3,657
b) email/call/meet Mikovits (or others with requisite expertise and/or success) to see if there might be un-published or unanticipated ways of improving your methodology. In my opinion, one of these is far more conducive to scientific progress while the other, the one we've seen chosen repeatedly, is antagonistic and counterproductive. The only group I recall choosing the latter route is that of Dr. Hanson, who did subsequently improve their ability to find evidence of the virus, albeit inconsistently still.

Why would scientiest do this? If anything it should be the other way around. Mikovits was unable to replicate her own findings. In fact Singh did do this and was still unable to find anything.

It appears that some are starting with the premise that because they respect Mikovitx, it follows that her science has to be correct. This is a logical fallacy. It has nothing to do with how much we like Mikovits or not, it has to do with the scientific knowledge we have at the present and not the scientist. It's possible to admire her and at the same time admit that in hindsight she made mistakes.

I am anxiously waiting for the Lipkin study to come out so this matter can be settled as much as possible.

Barb C.:>)
 

currer

Senior Member
Messages
1,409
Barb, you are trying to put people into a false position. It is not to do with "liking" Mikovits.

It is to do with her willingness to study CFS/ME and think about this disease and then build a hypothesis from her observations of how this disease presents clinically and her lab. findings. She has tried to build the diverse facts into an integrated approach. This is why she looked at cytokine levels and NK cell abnormalities.
No real advance can be made unless someone is prepared to actually study our disease.

We have just seen on this thread how misleading assumptions drawn from other diseases can be when applied to ME.

The other element which deeply concerns many posters on this thread has been the bogus arguments purporting to be "science" which have been used to discredit this research.
Now Mikovits research may be wrong.
But why use bogus arguments against her if you have better ones to hand?

There is a word for this. It is "propaganda."
 

Mula

Senior Member
Messages
131
So you mean that they rely purely on PCR, and they haven't confirmed their results in any other way?
I agree with you on that point.

I don't know what you mean about being 'complimentary' to VP62.
Do you mean that the primers were not verified by testing them on VP62 positive controls or VP62 clones? So they haven't validated their primers?
I agree with you on that point.

The Paprotka paper is not a conclusive paper, and the methodology is weak. It's just explores a hypothetical model of XMRV recombination.



That doesn't prove your earlier point. You were wrong about that point. They had enough sequences from both 22RV1 and CWR-R1, separately, to say that they had detected XMRV in each. However, it's only a minor issue, compared with the overall weakness of the study, as per the other issues we have discussed. And the env genes were detected, as far as the results of their primers indicate, just as the rest of the genome was detected using primers (all except a section of the LTR region, by the look of it.)

PCR was the only approach, but the primers are assumed to be complimentary to VP62 without confirmation using other necessary experiments. The assumption is that they are amplifying complimentary sequences but there could be any number of sequences that those primers will amplify. Clones don't imitate the actual conditions within the sample matrix under examination.

Why would I be wrong when the cell line of the gag-pro-pol sequence is /cell_line="22Rv1/CWR-R1" ? As I explained the primers are only assumed to be complimentary. The red bars are for unidentified sequences assumed to be VP62. The env gene is not to be found but there is a "putative gag polyprotein". This is a proposed env. Primers alone do not establish a gene sequence.
 

Mula

Senior Member
Messages
131
Word count is limited in journal publications but multiple draws were included in the response to comments by May 2010 and the addition of AZA was of public knowledge to those who had not worked on the paper by October 2009.
Samples were prepared within 6 hours of blood draw and frozen immediately in -800C or liquid nitrogen depending upon the sample type. Samples were taken from patients on at least two different time points within three months of each.
http://www.sciencemag.org/content/suppl/2010/05/12/328.5980.825-d.DC1/Mikovits_SOM.pdf
 

asleep

Senior Member
Messages
184
From Shin et al 2011:

""""
The protocol for the viral replication assay using spin inoculation was adapted from the one used in the original study that found XMRV in CFS patients (12), with extensive help from Frank Ruscetti (Leukocyte Biology Section, NCI).
"""

Ok, so there is another group that worked with one of the original researchers, but it appears that they adapted (read: modified) the assay in unspecified ways, which introduces unknown variables. Plus, there is more to methodology than details of a single assay. Deviation in cohort selection, sample collection, storage, and processing, and possibly other aspects all present realistic stumbling blocks.

W.r.t those patients experiencing apparent persistent viremia, was it not later determined (by Robert Silverman) that their PBMC DNA samples contained VP62 plasmid contamination, thus significantly undermining confidence in this finding?

I don't think we can say that for sure. VP62 was introduced at some point to these samples which fouled up the sequencing, but both labs claim evidence it wasn't them. I submit that the WPI must have consistently detected viral presence across multiple samples from patients in this subset in order to have considered characterizing the patients as persistently viremic. It sounds like you are proposing, then, that the WPI was somehow more consistently contaminating samples from this arbitrary subset of patients than samples from all the rest of the patients (some of which were presumably also analyzed similarly for persistent viremia). If we are to believe that the results of Lombardi et al are due to contamination, as you seem to suggest, we must necessarily also believe that they not only differentially handled patient samples from controls, but also differentially handled samples from arbitrary but consistent subsets of patients.

Furthermore, the main reason I introduced the quote from that paper was to demonstrate that Mikovits was actively aware of a divergence of viremic consistency across patients, a concept that seems to have been overlooked in almost every follow-up attempt, including those that came well after the Onlamoon study.
 

Bob

Senior Member
Messages
16,455
Location
England (south coast)
PCR was the only approach, but the primers are assumed to be complimentary to VP62 without confirmation using other necessary experiments. The assumption is that they are amplifying complimentary sequences but there could be any number of sequences that those primers will amplify. Clones don't imitate the actual conditions within the sample matrix under examination.

I agree with all of that Mula.

Why would I be wrong when the cell line of the gag-pro-pol sequence is /cell_line="22Rv1/CWR-R1" ? As I explained the primers are only assumed to be complimentary. The red bars are for unidentified sequences assumed to be VP62. The env gene is not to be found but there is a "putative gag polyprotein". This is a proposed env. Primers alone do not establish a gene sequence.

I'm basing my understanding of it all on the Paprotka paper. I might be wrong about some of it, but you haven't provided me with any evidence that demonstrates that I'm wrong.

The wording that you are quoting: "gag-pro-pol sequence is /cell_line="22Rv1/CWR-R1" doesn't prove your point, as far as I understand. It just tells us that these sequences were found in the two cell lines. The Paprotka paper shows that all of the sequences were found in each of these two cell lines.

I agree with what you are saying about assumptions made in the Paprotka paper. But if you look at the Paprotka paper, they detected the env gene in the same way as all the other genes, using PCR.

Dusty Miller explained what 'putative' means in this context. I think he said it's because they have not actually detected viral RNA, but only detected the associated gDNA (mouse genetic DNA), just as they did with the rest of the genes, as far as I can see. So this means that they used the same method to detect the env gene as the rest of the genes. I don't know why the env gene would be labelled differently in genbank though.

The fact that they've only detected gDNA for all the viruses in Paprotka (XMRV, PreXMRV-1 & PreXMRV-2), again demonstrates how they've made assumptions, as we've been talking about.

As we've been agreeing, they have made a lot of assumptions, and they've only assumed that they detected XMRV, based on the results of the PCR. Their assumptions re PreXMRV2 were even worse, as I've outlined earlier in the thread. And they made tons of other assumptions as well. They've created a neat and interesting hypothetical model, but that's all it is.
 

anciendaze

Senior Member
Messages
1,841
To narrow the focus of arguments on that env gene I would like to ask those of you who feel up to using sequence matching software to tell me where else we might find the immunosuppressive env sequence referenced in that paper and patent by Heidmann, et al.? That domain is highly specific and quite sensitive to mutations. It could also have health implications all by itself.

I am painfully aware of cognitive deficits which could be obstacles to doing so myself, verbal intelligence is much less affected. I now also try to avoid cognitive overload from masses of raw detail. I once worked with sequence-matching algorithms in a non-biological context. Biological sequences are often like the test cases you create to test algorithms, with any number of repetitions, overlapping sequences and generally weird subsequences, as if created by an opponent with malice aforethought. If I felt confident I could spot check things by looking at raw data, to see where an algorithm might be misleading, I'd tackle the problem.

An algorithm usually incorporates tacit assumptions prevalent at the time it was designed. Fixing bugs is unlikely to remove major conceptual errors. There are always some users who want the original behavior, for example people interested in host genes coding for proteins want the clutter of viral and bacterial sequences removed. In this case the assumptions which concern me are about "junk DNA" and ERVs. This is one reason why I favor novel sequencing methods with a different conceptual basis.
 

Mula

Senior Member
Messages
131
I agree with all of that Mula.



I'm basing my understanding of it all on the Paprotka paper. I might be wrong about some of it, but you haven't provided me with any evidence that demonstrates that I'm wrong.

The wording that you are quoting: "gag-pro-pol sequence is /cell_line="22Rv1/CWR-R1" doesn't prove your point, as far as I understand. It just tells us that these sequences were found in the two cell lines. The Paprotka paper shows that all of the sequences were found in each of these two cell lines.

I agree with what you are saying about assumptions made in the Paprotka paper. But if you look at the Paprotka paper, they detected the env gene in the same way as all the other genes, using PCR.

Dusty Miller explained what 'putative' means in this context. I think he said it's because they have not actually detected viral RNA, but only detected the associated gDNA (mouse genetic DNA), just as they did with the rest of the genes, as far as I can see. So this means that they used the same method to detect the env gene as the rest of the genes. I don't know why the env gene would be labelled differently in genbank though.

The fact that they've only detected gDNA for all the viruses in Paprotka (XMRV, PreXMRV-1 & PreXMRV-2), again demonstrates how they've made assumptions, as we've been talking about.

As we've been agreeing, they have made a lot of assumptions, and they've only assumed that they detected XMRV, based on the results of the PCR. Their assumptions re PreXMRV2 were even worse, as I've outlined earlier in the thread. And they made tons of other assumptions as well. They've created a neat and interesting hypothetical model, but that's all it is.

There is no sequence with all three genes in the NCBI database for 22Rv1 or CWR-R1. Putative is a term for proposed and for the sequence in question from the database there is no envelope gene. Sequences placed into the database represent a single source or no comparison is achievable. Paprotka only features sequences assumed to be VP62 but without data to confirm they are, unlike the sequences for Prexmrv1 and 2 in the database. I wouldn't say you could call this a hypothesis.
 

Mula

Senior Member
Messages
131
Ok, so there is another group that worked with one of the original researchers, but it appears that they adapted (read: modified) the assay in unspecified ways, which introduces unknown variables. Plus, there is more to methodology than details of a single assay. Deviation in cohort selection, sample collection, storage, and processing, and possibly other aspects all present realistic stumbling blocks.



I don't think we can say that for sure. VP62 was introduced at some point to these samples which fouled up the sequencing, but both labs claim evidence it wasn't them. I submit that the WPI must have consistently detected viral presence across multiple samples from patients in this subset in order to have considered characterizing the patients as persistently viremic. It sounds like you are proposing, then, that the WPI was somehow more consistently contaminating samples from this arbitrary subset of patients than samples from all the rest of the patients (some of which were presumably also analyzed similarly for persistent viremia). If we are to believe that the results of Lombardi et al are due to contamination, as you seem to suggest, we must necessarily also believe that they not only differentially handled patient samples from controls, but also differentially handled samples from arbitrary but consistent subsets of patients.

Furthermore, the main reason I introduced the quote from that paper was to demonstrate that Mikovits was actively aware of a divergence of viremic consistency across patients, a concept that seems to have been overlooked in almost every follow-up attempt, including those that came well after the Onlamoon study.

There is enough variance from the Mikovits paper to suggest the presence of at least two viruses.

The simplest explanation for requiring repeat samples is the level of virus. A blood sample taken at one two or more time points could conceivably not contain the virus infecting a host as a consequence of that allocation not containing the virus, but when taking a measure of blood with the virus the tests are positive. Simply put, insufficient virus to infect every sample.
 

RustyJ

Contaminated Cell Line 'RustyJ'
Messages
1,200
Location
Mackay, Aust
It's awfully hard to fault other groups for not following Dr. Mikovits's methodology when that methodology was not properly described in her published work. For instance, the Lombardi paper makes no mention of the need for multiple PCR runs or that 5-azacytidine was used in the production of the (mislabelled) Western blot.

This is myth. Mikovits published no less, no more than other studies. It has also been on record that Science restricted the format of information in the published work. Mikovits has been happy to talk to other researchers about her methods and has been on record as having collaborated with other groups.

I don't know how many times Mikovits stated that PCR was problematic and that her methodology had been refined over a long period. Irrespective of the methodology, other groups were going to find it tough to emulate her work.

Irrespective of Mikovits methods, the bottom line is that in most cases PCR is going to be insufficient. All the negative studies did was support this. They did not prove the non-existance of XMRV.

[Edit] Er, sorry, some of my points were already made by previous posters. Seems I didn't realise there were later posts.

The statement also ignores the reality of the science. No one wanted to copy her methods. The whole point of Coffin's, Stoye's, Levi's etc efforts was to patent their own assays. Once they were satisfied they could not do this, they were happy to abandon their work, or to discredit Mikovits'.
 

Bob

Senior Member
Messages
16,455
Location
England (south coast)
There is no sequence with all three genes in the NCBI database for 22Rv1 or CWR-R1.

The (assumed) entire 22RV1/CWR-R1 genome is given:
http://www.ncbi.nlm.nih.gov/nuccore/FN692043.2

Putative is a term for proposed and for the sequence in question from the database there is no envelope gene.

The (assumed) 'putative' env gene is clearly labelled:
http://www.ncbi.nlm.nih.gov/nuccore/FN692043.2

They detected the (assumed) 22RV1/CWR-R1 putative env gene using PCR primers, just as they did with the rest of the genome. So there is no difference between the putative env gene and the rest of the genome, in terms of evidence.

Paprotka only features sequences assumed to be VP62 but without data to confirm they are...

Yes, we've established that.

...unlike the sequences for Prexmrv1 and 2 in the database.

Paprotka used the same methodology for all the viruses.
 

Bob

Senior Member
Messages
16,455
Location
England (south coast)
Imagine a group that tries to reproduce the Lombardi results and comes up empty. (Perhaps they did their very best to reconstruct precise methods from the published paper, or perhaps--as seemed far more common--they decided to do their own thing, maybe even going so far as to defy basic sensibility by doing something like excluding patients with signs of infection. Either way, it doesn't matter.) This group has a choice. They can a) defiantly publish their results under the presumption that their isolated efforts were sufficient and cannot have overlooked important details known to others or b) email/call/meet Mikovits (or others with requisite expertise and/or success) to see if there might be un-published or unanticipated ways of improving your methodology. In my opinion, one of these is far more conducive to scientific progress while the other, the one we've seen chosen repeatedly, is antagonistic and counterproductive. The only group I recall choosing the latter route is that of Dr. Hanson, who did subsequently improve their ability to find evidence of the virus, albeit inconsistently still.

My understanding was that Hanson was getting positive results for MLVs, or MLV-like viruses, but that she decided it was contamination in the end.
I never found out why she came to this conclusion. (Did she ever publish a paper?)
However, she said that she did get positive antibody results in more patient samples than in controls, and said that these results were worth following up.
With the success of Rituximab, and the info in that interesting abstract that I posted earlier, I'm very interested in the possibility that EBV, or any other viruses, are reactivating HERVs in B cells, and that the antibody results might be a result of cross reactivity with the HERVs. I think this is a line of enquiry worth pursuing whether MLVs are involved in ME or not. (Interesting that the abstract suggests that anti-retrovirals might treat reactivated HERVS.)
 

currer

Senior Member
Messages
1,409
I don't know how many times Mikovits stated that PCR was problematic and that her methodology had been refined over a long period.

I was just checking over Mikovits IiME talk last year for anything about NK cells for Acer and I came across this which is a useful example w.r.t. the discussion on methodology, i.e. how careful you need to be because so many variables will affect the result.

"...and this brings me to another area which is really important, now when you are doing these studies, that is with regard to isolation, we isolate peripheral blood mononeuclear cells, but if you are not a longtime cell biologist you might use off the shelf separation media called "lymphocyte separation media" - it has a different density.

You will only get B and T cells and some NK cells that are smaller. You will miss the entire myeloid compartment - the eosinophils, the neutrophils. They will stay with the red blood cells and you wont be isolating them in the leukocyte compartment.

We think given our work with HTLV1 that these cells may actually be the major reservoir in the blood.
Many of these cells are very short-lived, the granulocyte is a bag of proteolytic enzymes that is gone within four days and any virions that are in those granulocytes would be totally chewed up by the enzymes but neucleic acids remain.

We think this is what we were doing when we delay processed the cells - releasing neucleic acids into the plasma that we could measure."
 

currer

Senior Member
Messages
1,409
I also came across this with regard to immune dysregulation and NK cells;-

After the heading "Identify distinct immune phenotypes and functional profiles in XMRV infection - translate into clinical diagnostic as compliment to cytokine profiles and monitor efficacy of therapy"

."......so we were looking at NK cells and we noticed that (as has been published many times before) that the activity and the function of NK cells - (the CD3 negative and the CD56 postitive cells) - are significantly reduced in CFS patients.

What we see quite strongly is that the NK cell numbers are actually decreased but another population comes up that is not seen at all in any healthy individual this is called an NKT. We havent fully characterised it, you can see that there are two sub-populations here.

We are in the process of sorting and phenotyping - is that cell infected ?- what are the aberrant cytokines promoting its production ? - but this would have to do with immune dysregulation.

As Ken Hunter our immunology advisor said...You would have to do thousands (of samples) to prove this...and we didnt have a whole lot of negative controls...he has not seen anything like this before..."
"So the NKT cell may well be a cell that we can follow in therapy."

At this time Dr Mikovits was looking for blood markers to chart the up or downregulation of infection in her infected patients. This is what she did in her study of Dr Snyderman's CLL and CFS. This important work has attracted NO response - which is a disgrace.

She states CLEARLY just after this section that she is not only finding XMRV but other PMRVS.

This talk was given in June last year. Only a couple of days after this conference she was asked to withdraw her Science paper and subsequently forced to withdraw it and sacked from the WPI and all her work in this area stopped.

Now contamination or no contamination it is a shame that these initiatives on blood markers and cytokine profiles have been lost to us.

Dr Mikovits was investigating how MRV infection could result in the type of pathology that is seen in ME patients. She mentions in this talk that certain immune cells can present virions to the neurons in spinal tissue without those virions ever needing to be present in blood.

So you could get an infection in nervous tissue without free virions in the bloodstream.

These are important areas of study, and possibly very relevant to this disease.Dr Mikovits was prepared to think broadly and in depth about this pathology.
So much of the opposing science has been very one - dimensional by comparison - content to stop at the first "negative" result without thinking any further either about our disease, or about the complications of MRV infection.

Now if it is not an MRV but a HERV, - that is just as important because something is activating this HERV. Can a HERV causing all the immune pathology found in ME? Can it cause NK cell dysfunction? Can it infect the nervous system - because there are clear signs of ganglionitis in the sensory nerves in ME.

The virologists who have attacked Dr Mikovits work have no interest in these questions and are content that they remain unanswered. This does not help us in any way.

Many of the pathologies we have are pathologies associated with MRV infection as is known from decades of research with mice. We need to find the underlying cause, not be content with research that only charts downstream effects. If there are NK cell abnormalities the correct question is WHY?
 

Bob

Senior Member
Messages
16,455
Location
England (south coast)
I got the impression that Hanson didn't really know if it was contamination or not, but that she became cautious because of the amount of negative studies. She detected something:

http://forums.phoenixrising.me/inde...retroviruses-maureen-hanson-david-bell.17574/

Thanks Jemal, I've posted all the way through that thread, and I couldn't even remember if Hanson had published! I'm sure my memory has been getting worse recently. :(

Yes, my interpretation is that Hanson validated Mikovit's work, finding P-type MLV-like sequences (similar to, but not the same as, Lo's), but was unable to repeat her test results reliably. She could not understand why she was getting the unreliable results she got, and decided that it was an indecisive study.

Quite a high number of patient samples were positive, if we count samples that were found to be positive at least once, (maybe between 47% and 77%), and the number of positive controls appears to be very low, although I never managed to get a total for them (maybe 3 control samples?):
http://forums.phoenixrising.me/inde...en-hanson-david-bell.17574/page-2#post-268327
http://forums.phoenixrising.me/inde...en-hanson-david-bell.17574/page-6#post-269958
 

lansbergen

Senior Member
Messages
2,512
The (assumed) entire 22RV1/CWR-R1 genome is given:
http://www.ncbi.nlm.nih.gov/nuccore/FN692043.2


I do not see it there. It only shows Gag, Pro and Pol.

At viralzone http://viralzone.expasy.org/all_by_species/67.html I see a gammaretro gnome is about
8.3 kb.

The 22Rv1/CWR-R1 is too short



The VP62 is at http://www.ncbi.nlm.nih.gov/nuccore/121104176?report=graph

That is longer and shows the envelope

The (assumed) 'putative' env gene is clearly labelled:
http://www.ncbi.nlm.nih.gov/nuccore/FN692043.2.

I do not see it. Can you show it to me please?
 
Back