Hi Cort,
This statement from your Buzz Page is incorrect:
He notes that the Hot Taq polymerase enzyme has mouse antibodies and can carry XMRV.
Weiss does
NOT say that HotStart Taq Polymerase contains XMRV. He
speculates that it
might sometimes contain traces of X-MLVs, the mouse ERVs to which XMRV appears to be (in part) related.
I hope it's possible for you to edit that post on the XMRV Buzz page? :worried:
I agree with Kurt that too little has been made of confounding possibilities and looking back it's true that in general the WPI did present the virus in a very positive light.
Not in their paper, which is what matters to other researchers. It did not "oversell" their positive findings, and the study was in fact designed to provide multiple corroborations of their PCR results and to reduce any potential contamination concerns.
I definitely agree with you about the Weiss article being lightweight. When a scientist raises concerns about flaws and potential artifact in another scientist's results, s/he is supposed to suggest an alternative experiment to rule out said artifact - which Weiss really does not, as even a multi-lab study wouldn't be immune to some of the vague concerns he raises - and to provide a critique of their own statement, e.g. point out the potential flaws in the contamination argument. No one who has raised contamination as an issue has done this so far.
Erlwein et al brought up the Taq polymerase issue already in their comment to PNAS, and as you mentioned Lo et al provided a very strong response.
Too worn out to critique Weiss's spiel thoroughly, but a few points on this paragraph of his:
"CSF [sic] patients would then be assured of having a recognized infection with the possibility of effective treatment - indeed, some of them are already so convinced they have started treatment with anti-retroviral drugs (first developed against HIV) in the hope of clearing infection and their symptoms. Blood banks would have to consider whether to screen donations for the implicated retroviruses. But before such steps could be justified, it will be essential to perform truly blinded tests on cases and proper controls in several laboratories. Profoundly disappointing as this would be for patients, without such additional studies, laboratory artifacts cannot be ruled out; also, with the signal exceptions of HIV and human T-lymphotrophic virus, the history of retroviral associations with human disease is not encouraging."
(1) Blood bank donor screening does not have to wait on the results of his proposed studies; that is a public health precaution which needs to be instituted more swiftly than the slow wheels of science can turn (his own government's blood supply agency, by the way, will likely disagree with him on this matter).
(2) Carefully controlled antiretroviral trials technically do not
have to wait until his proposed requirements are met, either.
(3) The Lombardi study
was "truly blinded"; for him to to imply otherwise, as I believe he was (he clearly wasn't just referring to the Alter/Lo study) is deceptive.
(4) He does not mention the ongoing Blood Working Group on XMRV, which
is a multi-laboratory coordinated study "with blinded tests on cases and proper controls". [By the way, what the heck is a "proper control"? Spiked water, like his colleague McClure used?]
(5) The fact that there is a history of false "retroviral associations with human disease" should have no bearing on a scientist's attitude towards current research in this area, unless specific mistakes from those failed efforts are being repeated. To say otherwise would be to bring unacceptable researcher bias into the equation (which, frankly, Weiss seems to be doing).