IC Study is STILL not a credible replication study
Someone had asked, what did the "Spiking experiment" mean? From what I understand, they added a known quantity of XMRV into the test tubes of patient samples - to see if they COULD indeed find XMRV if it was there.
Many questions were raised about the Imperial College's ability to detect XMRV. For example, the Polymerase Chain Reaction (PCR) bands illustrated in the figures accompanying the IC were very weak - even for the XMRV control. Some readers said things like, "if your control bands are that weak - when you know XMRV IS present, how dim (or nonexistent) might they be when there is a very low concentration of XMRV present? In other words, might a test that is already perceptibly weak, yield false negatives?
On the results of the spiking experiment:
Unfortunately this addresses what was one of the major weakness of the study, and the results after the experiment stand. From an objective standpoint this substantially strengthens the overall results of the IC study. If they can detect XMRV in an in vivo spike like this, then their lack of detection of XMRV in the study carries more weight. Of course there may be other confounding issues in their study, but this removes some of the doubt of their results.
@Kurt, can you help interpret the spiking results further?
1) From McClure's PLoS January 11th response @
http://www.plosone.org/annotation/l...notation/6bfbac4a-5ace-4a2d-a9bb-60f017ae24d8 , she said,
" We randomly chose 22 of the CFS patient samples and spiked the DNA with 10 copies of XMRV plasmid DNA. The XMRV /MLVsequences were amplified in every case.
Positive and negative controls worked beautifully"
Wouldn't that just prove that their test was sensitive enough to pick up THAT particular viral threshold? What if patient samples had say, 1 copy of XMRV plasmid DNA? Or 3? For example, I know from the persistent PVB19 viral research that persistent infections DO often have very low findings on serology. Tests need to be exceptionally sensitive to pick up low viral levels, and even then sometimes they are only found in tissue, not in bloodwork.
2) There still remain, as you rightly say, "other confounding issues in their study", not the least of which is their cohort selection. Do you have any other thoughts on that?
Thanks Kurt!:Retro smile: