Posts Tagged ‘error’

XMRV: More nails in the coffin

5 June, 2011

Except that the title could be “More nails from the Coffin”, given the involvement of someone of that name in amassing the growing weight of evidence against XMRV as an actual natural pathogen – but I digress.

The Nature News blog of 31st May has a very damning collation of views and evidence from around the scientific community – but chief among these is the fact that Science, which published the original paper describing the finding of XMRV in human-derived specimens, has called on the authors to retract it.  The evidence – partly gathered by John Coffin – seems clear: XMRV is a recombinant retrovirus which is a chimaera of two mouse viruses which got into cells derived from a human prostate tumour when these were cocultured with mouse cells.  It is not a “natural” virus, but a laboratory accident; it probably has no relevance to any human disease.

Another interesting and more philosophical view derived from the XMRV saga is that of The Independent, of 3rd June: Steve Connor in “Science Studies” points out that this is, in fact, how science really works – or should work.  That is, that someone publishes something that is really interesting – but which becomes contentious because other can’t replicate it, and eventually is wholly or partially discredited.  All out in the open, in the scientific press.

Some folk – acting with perfect hindsight – then bemoan the fact that the original article was published at all; others are horrified at the waste of money as people dig around and around in the same hole.  What they forget is that progress has been made, whether or not the initial revelation was in fact true.  And that is how science should work.

And because of that sort of iteration, XMRV is going the same way as cold fusion, folks.  And here’s a goodbye….