When a disaster happens in science, like the XMRV fiasco, the most important lesson everyone can take away from the incident is ‘Why?’
If you understand the ‘Why?’ behind what happened, you can avoid the problem in the future.
So specifically in the case of the XMRV fiasco, the question was, “Why did samples in this study appear positive, while no one else could replicate the findings?”
The answer was contamination, contamination, contamination.
MLV viruses are in our cell lines (we already knew that) and mouse gnomic DNA, chock full of mouse ERVs, were in all of our reagents.
Well, neato. This means that if you are doing experiments in the future, you need to be on the lookout for these forms of contamination. Hurray.
… No, not really. I am not satisfied at all with the answers put forth to ‘Why?’
See, Bob Silverman already answered a ‘Why?’ long before the official retraction. He found XMRV plasmid (amp resistance genes, unquestionably, incontrovertibly from plasmid) in patient samples. Okay, fine. I guess that totally could have happened on accident (**WINK!!**). Whatever.
And I (and others) figured out that ‘XMRV’ protein expression was artfully/creatively/falsely demonstrated by treating patient samples (and only patient samples) with epigenetic modifiers, capable of inducing expression of any/all endogenous retroviruses in the patients genome.
But we were still left with a cartload of ‘Why?’s in this paper.
I was hoping Frank Ruscetti retesting some of the Science samples with the help of John Coffin would answer the remaining issues with the Science paper, so we could all understand the ‘Why?’ behind the rest of the mistakes so we dont make them ourselves.
Multiple Sources of Contamination in Samples from Patients Reported to Have XMRV Infection
It did not.
Technically, it made things worse.
See, we *KNOW* from Bob Silvermans independent retraction that there was XMRV, VP62 plasmids spiked into the patient samples he had. We *KNOW* that.
And yet…
That is not what Kearney et al found in their samples. They found mouse DNA/mouse ERVs.
The three methods yielded concordant results for all 9 samples and provided unequivocal evidence that the plasma samples provided to us from 4 of the CFS patients originally reported [12] to be XMRV infected were contaminated with mouse DNA.
Now.
But in 2009, those sequences were unquestionably iterations of VP62. If Lombardi et al had ‘found’ mouse DNA contamination before, then their data should have mirrored Lo et al. It didnt.
What??
This paper does explain *one* ‘Why?’– XMRV plasmid or mouse DNA in serum samples will not induce expression of infectious retroviruses. This group tried to figure out what went on with Mikovits/WPIs super magical co-culture experiments. Turns out all the cell lines they used are infected with XMRV.
But then how were their ‘mocks’ negative?
And we are still left with how 100% of their patient cells stained positive for flow, when an elaborate voodoo ritual needed to be performed in order to get the samples to test positive for XMRV via PCR. And now we have more, new, ‘Why?’s.
Ugh.
I wish they would just come out and say “We faked it all, and heres how.”