Sing
Senior Member
- Messages
- 1,784
- Location
- New England
XMRV and chronic fatigue syndrome: Questions I have
Category: XMRV
Posted on: February 1, 2010 9:23 PM, by ERV
So here are my answers to The Brainstorm Challenge.
Some of you got real damn close to the 'answers' I was thinking of, but you all missed a great big one (which I think will make sense to you after I bring it up hehe!)
1-- Lets say youve isolated white blood cells from CFS patients. You treat these cells with chemicals that interfere with normal DNA/histone methylation. What do you think will happen? Do you think that is a good diagnostic test for retroviral infection?
The Reno groups decision to use this as a diagnostic test is absolutely baffling. The idiomatic definition of 'epigenetics', histone and DNA modifications, probably evolved as a method of controlling pirate DNA. Pirate DNA like endogenous retroviruses.
Put 'ERV methylation' into PubMed. 'LTR methylation'. You screw this up, you get particle production.
Treat cells with a chemical that messes up methylation... and you get retroviral particle production... XMRV or not.
2-- Magic Johnson was diagnosed 'early' and got on antiretrovirals. Do you think there is any chance Magic Johnson will develop AIDS?
Maybe he will, maybe he wont. I would not say "Magic Johnson will not develop AIDS" in a million years.
Heres what happens with HIV-1 infected individuals:
Lets say you are diagnosed early. Get on HAART, viral load goes down, CD4+ T-cells stay up, YAY!
Well, there are always drug resistant variants present in the patients quasispecies.
Drug resistance comes at a fitness cost.
So, there are still HIV-1 viruses replicating in the patient. They might be real shitty, replicating real slow and awkward like, but theyre still going.
Some people are very very very unlucky, and in those few crappy replication cycles, the virus stumbles upon a secondary compensatory mutation. A mutation that allows it to be drug resistant AND able to replicate at a normal rate.
Some people are very very very lucky, and in those few crappy replication cycles, the virus just keeps banging its head against a wall.
The latter is like Magic Johnson. But there is no guarantee, with anyone who takes their antiretrovirals religiously, that they wont be unlucky tomorrow.
With todays technology, with todays antiretrovirals, we can extend the lives and improve the quality of life of people with HIV-1. But we cannot say they will 'never' develop AIDS.
3-- Lets say you isolate a retrovirus from a sample from 1984. The sequence from that virus is not significantly different from sequences you are isolating from patients 25 years later. In other words, this 'retrovirus' is not acting as a quasispecies. What are possible explanations for this? If the virus does not mutate, why could the British group not find MLV sequences we know are conserved? If this virus does not mutate, why would the PI looking for this virus be worried about PCR giving 'false negatives'?
*sigh*
Fish gotta swim.
Birds gotta fly.
And retroviruses gotta act as a quasispecies.
They have to. They cannot help it. Its a side-effect of an error-prone reverse transcriptase and inter- intra-strand recombination. Even if it finds the most perfectest sequence EVAH!, it cannot keep it.
And that most perfectest sequence in Patient #1 might be awful in Patient #2, and Patient #3. Every individual is a different environment...
Certainly there are regions of a retrovirus that are functionally constrained-- if they do not have sequence ABC, then the proper structure doesnt form, and viruses are non-infectious, therefore, sequence ABC is always there, but in a region like env? There is genetic plasticity, there is functional plasticity, there is selective pressure by everyones individual antibody repertoire! You cant stop the virus from mutating! If the virus stops mutating, the Red Queen race between us/retrovirus stops, and the virus is gone. I am not currently aware of any instance of anyone or any organism being 'cured' of a retrovirus ever.
But, quote Mikovits, "XMRV doesnt act as a quasispecies."
I just dont see how this is possible.
4-- Lets say we just discovered a new virus in humans. While most laboratories are being conservative/cautious about their statements and approach to this discovery, another lab is verbally, though not scientifically, 'connecting' this virus to CFS, breast cancer, chronic lyme disease, autism, and a cadre of other 'medical mysteries'. Furthermore, the PhDs in these labs are giving medical advice like 'take supplements X, Y, Z and immune modulators' and suggesting 'detox'. They are also heavily emphasizing 'early detection' of this new virus to prevent this list of diseases, and why, they have a test for sale right here. Do you think that is the most scientific approach to this new virus? What advice would you give this group of scientists?
This is example #918356125 of how unprofessional the Reno group is. There has been nothing published connecting XMRV to autism. Nothing. There has been nothing published connecting XMRV to chronic Lyme disease. There has been nothing published connecting XMRV to breast cancer. So when youre talking to the general public, you say general things like "Lots of other labs are trying to see if there is a connection between XMRV and their disease of interest. None of this, including XMRV-->CFS, has proven to be causal yet. This is currently a neat phenomena in CFS that might turn out to be something real fantastic! But right now, everything is preliminary."
Standing up in front of a group of laymen saying "THEYVE CONNECTED XMRV TO AUTISM AND BREAST CANCER AND LIEK EVERYTING!" screams insecurity and immaturity.
And a PhD, in any field, giving medical advice? Thats down right irresponsible.
Look, my epigenetic research, I just tell people "You know what? I eat my broccoli, LOL!"
I do not tell people failing chemo "OMFG YOU NEED TO TAKE X, Y, Z SUPPLEMENTS AND DETOX WARBLEGARBLE!"
I have no doubt CFS is a real disease. PhDs are not medical physicians qualified to treat diseases. End of story.
Furthermore, Ive heard it through the grapevine that a nice, normal diagnostic test for XMRV is in the works. It looks for anti-XMRV antibodies. Awesome!
Its not from the Reno group.
It will be for research purposes only, at this point, to study the epidemiology of this virus.
There is also lots of nice, normal basic science, basic virology being done on XMRV.
Not from the Reno group.
There is going to be lots of information coming though the pipeline on XMRV. Maybe it causes CFS, maybe it doesnt. Maybe it causes certain kinds of leukemia, maybe it doesnt. Maybe it causes certain kinds of prostate cancer, maybe it doesnt.
This information is going to come out through hard work done by normal scientists doing normal scientist things.
Not by PR releases accusing other labs of fraud.
Not by doing confusing, scary, and misleading conferences for prostate cancer patients.
Find more posts in: Medicine & Health
Share this: Facebook
Twitter
Stumbleupon
Reddit
Email + More .
TrackBacks
TrackBack URL for this entry: http://scienceblogs.com/mt/pings/130833
Comments
1
There has been nothing published connecting XMRV to autism. Nothing. There has been nothing published connecting XMRV to chronic Lyme disease. There has been nothing published connecting XMRV to breast cancer.
Smith, on this issue you have been nothing short of brilliant. Clearly, this is a case of a "virus in search of a disease." These jerks are sub-par scientists trying to give false hope to CFS patients and make some big bucks by patenting a "test" for XMRV antibodies. Next thing, they'll try to convince the American Red Cross to use this "screening" test on the blood supply -- a lotta money to be made with this scam.
You have smoked out these quacks with facts, logic and good science. Well done.
Posted by: Ben Rabb | February 1, 2010 10:54 PM
2
I do not know who you are, I am not a scientist myself, and I get lost in most of your arguments. But for some reason it seems to me you have something personal against WPI and Judy M. which I do not understand why... maybe because you need visits into your blog that will generate traffic, which also will generate add revenues into your pocket? Could be... What I know is that WPI has no economical interest, is a foundation, the mother of a CFS patient is in charge. The benefits of the XMRV test are used for research, and you are wrong assuming that XMRV has not been linked to Autism, Fibromialgia or Rare Multiple Sclerosis. You are wrong because it was detected in 40%, 60% and 100% of the small samples they analyzed but was not published in Science, because in order to publish you need a big study as they did with CFS. But is worth to have a look given the small studies. In my view anybody can understand that...
Posted by: pochoams | February 2, 2010 7:28 AM
3
pochoams, if you read and understood erv's arguments about the handling of the science, you would understand why she's unimpressed.
100% of a small sample of people with autism had two legs. Let's start the amputations right away, shall we?
Posted by: Stephen Wells | February 2, 2010 9:07 AM
4
pochoams - there's one thing I don't understand about your argument (OK, that's a lie) - why do you accuse this blog of having a conflict of interest, and in the same paragraph say that because the Reno group is led by a mother with CFS that is a good thing?
Surely they are both 'conflicts of interest'.
On a side note - as a (somewhat) medically trained person, your blog does a very good job of explaining virology. Keep up the good work.
Posted by: F Hamilton | February 2, 2010 10:08 AM
5
This was a fantastic post, Abbie. Very clear and emphatic. Too bad at least one person still doesn't get the point...
Posted by: minimalist | February 2, 2010 10:16 AM
6
Which branch of the shadowy international cabal is paying you to weave your web of lies?
Do they have any openings in legal?
Have robe, will travel.
Posted by: Prometheus | February 2, 2010 11:19 AM
7
One aspect of this issue confuses me. I've heard that CFS is kind of an umbrella diagnosis, and that people whose symptoms have no confirmed cause often get a CFS diagnosis.
Is it a disease, many diseases, a group of symptoms? I'm not a researcher, so I don't know whether one assumption is more practical than another. Is it best to assume all CFS diagnoses have the same cause until proven otherwise? Are researchers trying to find one identifiable marker (such as XMRV) and eject all non-XMRV-infected, chronically fatigued people from the CFS diagnosis? Was my impression about the umbrella diagnosis all wrong?
Posted by: Stella | February 2, 2010 3:31 PM
8
Excellent question, Stella. I have heard CFS refered to as a garbage can diagnosis, a receptacle for patients who insist on an answer but the science is not available to supply one. Certainly, once the science does come up with something (perhaps XMRV) to define the disease, patients given an incorrect diagnosis will be shuffled off into the next garbage can diagnosis.
And, erv, as for the connection to other diseases as mentioned in the original post, people with a diagnosis of CFS have scientifically and anecdotaly shown higher incidences of breast cancer, leukemia and other diseases. Families show evidence of being communally infected, and children in those families do have a higher occurance of Autism. Provided the connection between XMRV and CFS holds up, it is a legitimate theory, but, as of know, it is just a theory.
As for a test that looks for XMRV antibodies, that brings up an interesting dilemma. Many of the usual tests for other diseases are not effective on people with damaged immune systems, simply because they do not have the ability to produce antibodies. Now, as of today, I do not have a way to scientificly prove that I have CFS. But tests have proven beyond any doubt that my immune system is lacking several components. Here's another THEORY for you. Patients who become ill with CFS do so because they are unable to produce the antibodies needed to to stop XMRV in its tracks.
Posted by: heidi | February 2, 2010 4:10 PM
9
Heidi, antibody deficiency is not common in CFS (nor is it common in HIV). That notion would directly contradict the claim of opportunistic infections (EBV, HHV-6, CMV, etc) in CFS patients. Most of those opportunistic infections are detected, after all, via abnormal antibody tests. If your body is not producing antibodies to common pathogens, it would be extremely unlikely that you'd end up with a CFS diagnosis.
Posted by: thomas_bernhard | February 2, 2010 4:27 PM
10
Despite the media gaffes, the WPI has made a huge contribution to the sufferers of CFS, their families, and to those yet to become sick with this truly pernicious and misunderstood illness. It's a net positive, no matter how you break it down. More funding, more news, more hope, more research. Obsessively condemning a few speculations that J.M. made during a web seminar strikes me as really, really silly. Yeah, JM is not so savvy in front of the cameras. Are you?
Posted by: thomas_bernhard | February 2, 2010 4:46 PM
11
Also, during her lecture, JM went out of her way to advise CFS patients to wait for a better diagnostic test. I.e. an antibody test. She did not once explicitly recommend the VIP XMRV culture test for CFS patients.
Posted by: thomas_bernhard | February 2, 2010 4:52 PM
12
I enclose the last XMRV presentation given by WPI in case anybody is interested.
http://www.wpinstitute.org/news/docs/WPI_JAM_012210.pdf
Posted by: pochoams | February 2, 2010 5:19 PM
13
"Yeah, JM is not so savvy in front of the cameras. Are you?"
Yes, she is.
Posted by: Prometheus | February 2, 2010 5:50 PM
14
Prometheus, even us lurkers know you're in love with her.
Posted by: thomas_bernhard | February 2, 2010 6:19 PM
15
"But, quote Mikovits, 'XMRV doesnt act as a quasispecies.'"
What? Just, what?
Posted by: Vene | February 2, 2010 6:51 PM
16
Thomas-- You are not a lurker. Lurkers dont comment, or do so rarely. Youve only commented on a few XMRV posts, a very new topic on this blog, so technically, youre a newb.
Also, your 'newb dead giveaway' is lurkers know that Ive done/do lots of presentations to the general public, which are available here on ERV, and some of them I didnt even know I was being recorderd. So Prom was poking fun at your newbness.
hehe. Newb.
Vene--
One of the really interesting things about these studies is we only isolate one thing out of these people. When we do the sequencing, it’s clean. We don’t isolate quasi-species. We don’t have the virus have these changes here in one week or one year… we have patient samples across dozens of years. We isolated XMRV from a 1984 plasma sample from a patient. So we got it in 2008 and we got it in 1984, which again suggests that the virus has been around at least 25 years and it might have a role in disease. But it’s not plausible, so yes indeed, it could play a role in other things… Did I answer the rest of that question?
Posted by: ERV | February 2, 2010 7:28 PM
Post a Comment
(Email is required for authentication purposes only. On some blogs, comments are moderated for spam, so your comment may not appear immediately.)
Category: XMRV
Posted on: February 1, 2010 9:23 PM, by ERV
So here are my answers to The Brainstorm Challenge.
Some of you got real damn close to the 'answers' I was thinking of, but you all missed a great big one (which I think will make sense to you after I bring it up hehe!)
1-- Lets say youve isolated white blood cells from CFS patients. You treat these cells with chemicals that interfere with normal DNA/histone methylation. What do you think will happen? Do you think that is a good diagnostic test for retroviral infection?
The Reno groups decision to use this as a diagnostic test is absolutely baffling. The idiomatic definition of 'epigenetics', histone and DNA modifications, probably evolved as a method of controlling pirate DNA. Pirate DNA like endogenous retroviruses.
Put 'ERV methylation' into PubMed. 'LTR methylation'. You screw this up, you get particle production.
Treat cells with a chemical that messes up methylation... and you get retroviral particle production... XMRV or not.
2-- Magic Johnson was diagnosed 'early' and got on antiretrovirals. Do you think there is any chance Magic Johnson will develop AIDS?
Maybe he will, maybe he wont. I would not say "Magic Johnson will not develop AIDS" in a million years.
Heres what happens with HIV-1 infected individuals:
Lets say you are diagnosed early. Get on HAART, viral load goes down, CD4+ T-cells stay up, YAY!
Well, there are always drug resistant variants present in the patients quasispecies.
Drug resistance comes at a fitness cost.
So, there are still HIV-1 viruses replicating in the patient. They might be real shitty, replicating real slow and awkward like, but theyre still going.
Some people are very very very unlucky, and in those few crappy replication cycles, the virus stumbles upon a secondary compensatory mutation. A mutation that allows it to be drug resistant AND able to replicate at a normal rate.
Some people are very very very lucky, and in those few crappy replication cycles, the virus just keeps banging its head against a wall.
The latter is like Magic Johnson. But there is no guarantee, with anyone who takes their antiretrovirals religiously, that they wont be unlucky tomorrow.
With todays technology, with todays antiretrovirals, we can extend the lives and improve the quality of life of people with HIV-1. But we cannot say they will 'never' develop AIDS.
3-- Lets say you isolate a retrovirus from a sample from 1984. The sequence from that virus is not significantly different from sequences you are isolating from patients 25 years later. In other words, this 'retrovirus' is not acting as a quasispecies. What are possible explanations for this? If the virus does not mutate, why could the British group not find MLV sequences we know are conserved? If this virus does not mutate, why would the PI looking for this virus be worried about PCR giving 'false negatives'?
*sigh*
Fish gotta swim.
Birds gotta fly.
And retroviruses gotta act as a quasispecies.
They have to. They cannot help it. Its a side-effect of an error-prone reverse transcriptase and inter- intra-strand recombination. Even if it finds the most perfectest sequence EVAH!, it cannot keep it.
And that most perfectest sequence in Patient #1 might be awful in Patient #2, and Patient #3. Every individual is a different environment...
Certainly there are regions of a retrovirus that are functionally constrained-- if they do not have sequence ABC, then the proper structure doesnt form, and viruses are non-infectious, therefore, sequence ABC is always there, but in a region like env? There is genetic plasticity, there is functional plasticity, there is selective pressure by everyones individual antibody repertoire! You cant stop the virus from mutating! If the virus stops mutating, the Red Queen race between us/retrovirus stops, and the virus is gone. I am not currently aware of any instance of anyone or any organism being 'cured' of a retrovirus ever.
But, quote Mikovits, "XMRV doesnt act as a quasispecies."
I just dont see how this is possible.
4-- Lets say we just discovered a new virus in humans. While most laboratories are being conservative/cautious about their statements and approach to this discovery, another lab is verbally, though not scientifically, 'connecting' this virus to CFS, breast cancer, chronic lyme disease, autism, and a cadre of other 'medical mysteries'. Furthermore, the PhDs in these labs are giving medical advice like 'take supplements X, Y, Z and immune modulators' and suggesting 'detox'. They are also heavily emphasizing 'early detection' of this new virus to prevent this list of diseases, and why, they have a test for sale right here. Do you think that is the most scientific approach to this new virus? What advice would you give this group of scientists?
This is example #918356125 of how unprofessional the Reno group is. There has been nothing published connecting XMRV to autism. Nothing. There has been nothing published connecting XMRV to chronic Lyme disease. There has been nothing published connecting XMRV to breast cancer. So when youre talking to the general public, you say general things like "Lots of other labs are trying to see if there is a connection between XMRV and their disease of interest. None of this, including XMRV-->CFS, has proven to be causal yet. This is currently a neat phenomena in CFS that might turn out to be something real fantastic! But right now, everything is preliminary."
Standing up in front of a group of laymen saying "THEYVE CONNECTED XMRV TO AUTISM AND BREAST CANCER AND LIEK EVERYTING!" screams insecurity and immaturity.
And a PhD, in any field, giving medical advice? Thats down right irresponsible.
Look, my epigenetic research, I just tell people "You know what? I eat my broccoli, LOL!"
I do not tell people failing chemo "OMFG YOU NEED TO TAKE X, Y, Z SUPPLEMENTS AND DETOX WARBLEGARBLE!"
I have no doubt CFS is a real disease. PhDs are not medical physicians qualified to treat diseases. End of story.
Furthermore, Ive heard it through the grapevine that a nice, normal diagnostic test for XMRV is in the works. It looks for anti-XMRV antibodies. Awesome!
Its not from the Reno group.
It will be for research purposes only, at this point, to study the epidemiology of this virus.
There is also lots of nice, normal basic science, basic virology being done on XMRV.
Not from the Reno group.
There is going to be lots of information coming though the pipeline on XMRV. Maybe it causes CFS, maybe it doesnt. Maybe it causes certain kinds of leukemia, maybe it doesnt. Maybe it causes certain kinds of prostate cancer, maybe it doesnt.
This information is going to come out through hard work done by normal scientists doing normal scientist things.
Not by PR releases accusing other labs of fraud.
Not by doing confusing, scary, and misleading conferences for prostate cancer patients.
Find more posts in: Medicine & Health
Share this: Facebook
Stumbleupon
Email + More .
TrackBacks
TrackBack URL for this entry: http://scienceblogs.com/mt/pings/130833
Comments
1
There has been nothing published connecting XMRV to autism. Nothing. There has been nothing published connecting XMRV to chronic Lyme disease. There has been nothing published connecting XMRV to breast cancer.
Smith, on this issue you have been nothing short of brilliant. Clearly, this is a case of a "virus in search of a disease." These jerks are sub-par scientists trying to give false hope to CFS patients and make some big bucks by patenting a "test" for XMRV antibodies. Next thing, they'll try to convince the American Red Cross to use this "screening" test on the blood supply -- a lotta money to be made with this scam.
You have smoked out these quacks with facts, logic and good science. Well done.
Posted by: Ben Rabb | February 1, 2010 10:54 PM
2
I do not know who you are, I am not a scientist myself, and I get lost in most of your arguments. But for some reason it seems to me you have something personal against WPI and Judy M. which I do not understand why... maybe because you need visits into your blog that will generate traffic, which also will generate add revenues into your pocket? Could be... What I know is that WPI has no economical interest, is a foundation, the mother of a CFS patient is in charge. The benefits of the XMRV test are used for research, and you are wrong assuming that XMRV has not been linked to Autism, Fibromialgia or Rare Multiple Sclerosis. You are wrong because it was detected in 40%, 60% and 100% of the small samples they analyzed but was not published in Science, because in order to publish you need a big study as they did with CFS. But is worth to have a look given the small studies. In my view anybody can understand that...
Posted by: pochoams | February 2, 2010 7:28 AM
3
pochoams, if you read and understood erv's arguments about the handling of the science, you would understand why she's unimpressed.
100% of a small sample of people with autism had two legs. Let's start the amputations right away, shall we?
Posted by: Stephen Wells | February 2, 2010 9:07 AM
4
pochoams - there's one thing I don't understand about your argument (OK, that's a lie) - why do you accuse this blog of having a conflict of interest, and in the same paragraph say that because the Reno group is led by a mother with CFS that is a good thing?
Surely they are both 'conflicts of interest'.
On a side note - as a (somewhat) medically trained person, your blog does a very good job of explaining virology. Keep up the good work.
Posted by: F Hamilton | February 2, 2010 10:08 AM
5
This was a fantastic post, Abbie. Very clear and emphatic. Too bad at least one person still doesn't get the point...
Posted by: minimalist | February 2, 2010 10:16 AM
6
Which branch of the shadowy international cabal is paying you to weave your web of lies?
Do they have any openings in legal?
Have robe, will travel.
Posted by: Prometheus | February 2, 2010 11:19 AM
7
One aspect of this issue confuses me. I've heard that CFS is kind of an umbrella diagnosis, and that people whose symptoms have no confirmed cause often get a CFS diagnosis.
Is it a disease, many diseases, a group of symptoms? I'm not a researcher, so I don't know whether one assumption is more practical than another. Is it best to assume all CFS diagnoses have the same cause until proven otherwise? Are researchers trying to find one identifiable marker (such as XMRV) and eject all non-XMRV-infected, chronically fatigued people from the CFS diagnosis? Was my impression about the umbrella diagnosis all wrong?
Posted by: Stella | February 2, 2010 3:31 PM
8
Excellent question, Stella. I have heard CFS refered to as a garbage can diagnosis, a receptacle for patients who insist on an answer but the science is not available to supply one. Certainly, once the science does come up with something (perhaps XMRV) to define the disease, patients given an incorrect diagnosis will be shuffled off into the next garbage can diagnosis.
And, erv, as for the connection to other diseases as mentioned in the original post, people with a diagnosis of CFS have scientifically and anecdotaly shown higher incidences of breast cancer, leukemia and other diseases. Families show evidence of being communally infected, and children in those families do have a higher occurance of Autism. Provided the connection between XMRV and CFS holds up, it is a legitimate theory, but, as of know, it is just a theory.
As for a test that looks for XMRV antibodies, that brings up an interesting dilemma. Many of the usual tests for other diseases are not effective on people with damaged immune systems, simply because they do not have the ability to produce antibodies. Now, as of today, I do not have a way to scientificly prove that I have CFS. But tests have proven beyond any doubt that my immune system is lacking several components. Here's another THEORY for you. Patients who become ill with CFS do so because they are unable to produce the antibodies needed to to stop XMRV in its tracks.
Posted by: heidi | February 2, 2010 4:10 PM
9
Heidi, antibody deficiency is not common in CFS (nor is it common in HIV). That notion would directly contradict the claim of opportunistic infections (EBV, HHV-6, CMV, etc) in CFS patients. Most of those opportunistic infections are detected, after all, via abnormal antibody tests. If your body is not producing antibodies to common pathogens, it would be extremely unlikely that you'd end up with a CFS diagnosis.
Posted by: thomas_bernhard | February 2, 2010 4:27 PM
10
Despite the media gaffes, the WPI has made a huge contribution to the sufferers of CFS, their families, and to those yet to become sick with this truly pernicious and misunderstood illness. It's a net positive, no matter how you break it down. More funding, more news, more hope, more research. Obsessively condemning a few speculations that J.M. made during a web seminar strikes me as really, really silly. Yeah, JM is not so savvy in front of the cameras. Are you?
Posted by: thomas_bernhard | February 2, 2010 4:46 PM
11
Also, during her lecture, JM went out of her way to advise CFS patients to wait for a better diagnostic test. I.e. an antibody test. She did not once explicitly recommend the VIP XMRV culture test for CFS patients.
Posted by: thomas_bernhard | February 2, 2010 4:52 PM
12
I enclose the last XMRV presentation given by WPI in case anybody is interested.
http://www.wpinstitute.org/news/docs/WPI_JAM_012210.pdf
Posted by: pochoams | February 2, 2010 5:19 PM
13
"Yeah, JM is not so savvy in front of the cameras. Are you?"
Yes, she is.
Posted by: Prometheus | February 2, 2010 5:50 PM
14
Prometheus, even us lurkers know you're in love with her.
Posted by: thomas_bernhard | February 2, 2010 6:19 PM
15
"But, quote Mikovits, 'XMRV doesnt act as a quasispecies.'"
What? Just, what?
Posted by: Vene | February 2, 2010 6:51 PM
16
Thomas-- You are not a lurker. Lurkers dont comment, or do so rarely. Youve only commented on a few XMRV posts, a very new topic on this blog, so technically, youre a newb.
Also, your 'newb dead giveaway' is lurkers know that Ive done/do lots of presentations to the general public, which are available here on ERV, and some of them I didnt even know I was being recorderd. So Prom was poking fun at your newbness.
hehe. Newb.
Vene--
One of the really interesting things about these studies is we only isolate one thing out of these people. When we do the sequencing, it’s clean. We don’t isolate quasi-species. We don’t have the virus have these changes here in one week or one year… we have patient samples across dozens of years. We isolated XMRV from a 1984 plasma sample from a patient. So we got it in 2008 and we got it in 1984, which again suggests that the virus has been around at least 25 years and it might have a role in disease. But it’s not plausible, so yes indeed, it could play a role in other things… Did I answer the rest of that question?
Posted by: ERV | February 2, 2010 7:28 PM
Post a Comment
(Email is required for authentication purposes only. On some blogs, comments are moderated for spam, so your comment may not appear immediately.)