Retraction Watch

Tracking retractions as a window into the scientific process

Cell pulls fruit fly article, citing image manipulation

without comments

The journal Cell has retracted a paper on fruit fly genetics over concerns that the first author, a postdoc in a German laboratory, might have manipulated dozens of electron micrographs in the manuscript.

The article, published in November 2009, was titled “Assembly of Endogenous oskar mRNA Particles for Motor-Dependent Transport in the Drosophila Oocyte.” It has been cited six times since then, according to the Thomson Scientific Web of Knowledge.

Not having the foggiest notion of what those words might mean, other than that the paper was about fruit flies, we called in a ringer, Jeff Perkel, who explained as patiently as he could that the gist of the research involved

trying to understand how fruit fly oocytes first lay down the molecular signposts that will establish the animal’s final body plan during development.

To explore the question, the researchers, from the European Molecular Biology Laboratory in Heidelberg, first used electron microscopy to analyze frozen samples, then validated their findings with live-cell imaging.

As they state in their retraction notice:

In this paper, we used cryoimmuno-electron microscopy and live-cell imaging to investigate the sequential assembly of oskar mRNA into an mRNP competent for transport from the Drosophila nurse cells to the oocyte posterior pole.

So far so good. But:

We have recently identified instances in all of the figures where the cryoimmuno-EM data were inappropriately manipulated by the first author. The manipulations do not affect the live-cell imaging data. We are in the process of reanalyzing the raw experimental cryoimmuno-EM data but can already state that the published conclusions are not fully consistent with the raw data. We are therefore retracting the paper. We sincerely apologize for any inconvenience that this might have caused.

Although it might seem like good news that the live-cell data might be unscathed, those images account for just five of the 45 scans in the paper. The rest were generated by cryoimmuno-EM.

Even the Texas Rangers had a better batting average than that in the recent World Series.

Robin Wharton, a drosophila expert at Ohio State, says he read the retracted article when it first appeared and didn’t see any obvious signs that the images had been manipulated.

We have tried in vain to find contact information for the first author of the paper, Alvar Trucco, a postdoc in the lab of Anne Ephrussi. Ephrussi, the corresponding author on the article, has not responded to e-mailed requests for comment.

We also tried contacting the editors at Cell a number of times, to no avail.

Written by amarcus41

November 5th, 2010 at 3:15 pm

  • Pinko Punko May 23, 2011 at 10:36 pm

    The examination of images is the entire deal in cryo-EM. If the images are processed incorrectly, noise can be modeled as an actual result.

  • Post a comment

    Threaded commenting powered by interconnect/it code.