It seems that CEA just published my rather critical view of the recent ORMDL3 association paper. My letter had been first submitted to “Nature” but rejected after review.
A rebuttal seemed to be necessary as the authors repeatedly highlight ORMDL3 as a new asthma gene – in the printed Nature paper, in the Nature podcast and in the accompanying press releases. They even continue with reviews saying that
Completion of the human genome sequence and the advent of genome-wide association studies have resulted in the identification of two novel asthma susceptibility genes, ORMDL3 and CHI3L1, in the past year.
Besides all the facts mentioned in my CEA letter, there even more inconsistencies like unreported population stratification and unreported relationship in the German case-control study. I even believe that their figure 3 could not have been produced by the available data. During our recent correspondence the authors insisted that
our figure 3 contained all the information in Dr Wijst’s figure, with the exception that our LD map was drawn from the subjects and markers in our study and not a limited number of individuals in a public database.
So lets have a look at a (partially cropped and) zoomed version of their original figure 3a.
When plotting their data in the online deposit it turns out that the authors did not have such a high SNP coverage to produce such a high resolution figure.
The redrawn map is clearly different from the printed version while the reason for this inconsistency is unclear. At least the Nature editor wrote
publication of this debate is not justified: this is because in our view it would not add sufficiently to our understanding or otherwise clarify the issues for our readers.