Fraud and retractions in scientific writing - (Sep/20/2012 )

Just read a Nature news about a researcher who is risking hundreds of his research articles being retracted for fraud:
http://www.nature.com/news/retraction-record-rocks-community-1.11434

Another interesting read on this issue:
http://www.nature.com/news/2011/111005/full/478026a.html


I would think that most retractions were published in low impact journals but there's also a few high impact ones. I read a comment that the demands of higher impact journals to produce novel results with pretty images/diagrams could be pushing the trend towards image modifications (e.g. Photoshop). Thoughts?

Yeah it's becoming more of a problem... so much so that journals are starting to ask for the original images and using software to detect manipulations of images.

Having said that, the fact that an article has been retracted often doesn't mean that it gets fewer citations, as the retractions are often not noticed. I guess in this day or electronic journals etc, it is a bit easier to notify readers -just delete the paper or place a big notice on the .pdf file.

science noob on Thu Sep 20 07:30:13 2012 said:

Photoshopping an image is something that pretty much everybody seems to do!

When I take pictures with my fluorescence microscope for example, I neve get the clean cut images you see in papers..
A lot of the people make their images nicer, cleaner....
(this is not the same as creating an image from zero yourself or adding stuff.. but still.. I find it horrible how much clean up is done just to get really pretty pictures.

I find that journals should also publish the original images (the ones you see on your computer prior to cleanup)

this website tries to give some overview, though surely not exhaustive...the problem is that the journals or publishers usually won't shout it from the housetops that a paper was retracted...for them the damage to their public image seems more important than science.

@pito: what you are describing is beautification and this is not ok either; on the other hand the definition of how much of that you can do is: any linear changes to the images are ok (talking about the contrast curves from photoshop which you can manipulate non-linearly). Also many journals started asking for the full gels in the SI and the original data (e.g. Nature Cell biology)

ascacioc on Thu Sep 20 17:41:16 2012 said:

I know that its beatification.. and I do not like it, but almost everyone does it.. there are even journals out there that "stimulate" this by requesting a "nicer" image because yours is not "clean" enough or something.

ANd yeah: the full gels I understand, I know that many cut and paste pieces of gels.. just to get the best image (for example if the negative control went bad , they just cut it out and replace it with a negative control of another gel that went ok , but were for example the positive control was not good).



And about the retractions: I also dont understand it why they dont put a big red "retracted" on the pdf file, on every page of the paper.. Now people dont even know its retracted and often it take ages before the paper is removed from the website.

I guess if you look at the grand scale of things, photoshopped images are actually nothing compared with deceiving (and most probably harming) cancer patients (http://www2.nbc17.co...ppe-ar-2277973/ and funding organisation and damaging the reputation of a good university. Such was Duke’s Pottigate (60 minutes made a full feature about it here: http://www.cbsnews.c...in-cancer-care/)….involving a rising star oncologist (A Potti) who “discovered” how to match the patient’s tumour to the best chemo therapeutic drug” (personalized cancer treatment) and was touted as having found the holy grail of cancer research….and this resulted in publications (later on retractions) in Lancet, Nature medicine, NEJM, PNAS…a substantial grant from the American Cancer Society, patent applications and even a start-up company.

Keith Baggerly, the forensic bioinformatician who “challenged” Potti’s data (busted him so to speak) reported the cool reception of the journals to his inquiries and I guess, throughout this entire business, nobody really wants to admit that they were wrong or was duped or in case of the journals, they haven’t lived up to the highest standards that they demand from contributors.