Plagiarism earned genomics researchers an erratum, not a retraction, in BioMed Central journal BioData Mining.
We keep a list of best euphemisms for plagiarism, and this one is right up there.
Here’s the notice for “An iteration normalization and test method for differential expression analysis of RNA-seq data”:
Following publication of our paper in BioData Mining[1], it came to light that numerous sections of text were similar or identical to a previous publication by Robinson and Oshlack. In extending the method proposed by Robinson and Oshlack, we inadvertently copied text from their article. We acknowledge that these areas of text should have been clearly indicated as quotations or excluded entirely. We sincerely apologize for this oversight.
Even the title of the paper is too close for comfort – the one they ripped off is called “A scaling normalization method for differential expression analysis of RNA-seq data.” For those keeping score, that’s one word changed and two words added.
We’ve contacted the journal to find out what, exactly, they’d call plagiarism, and reached out to the authors for an explanation. We’ll let you know if we hear back.
Hat tip: Neil Saunders
Shocking that this earned an erratum rather than a full retraction; it’s more or less a verbatim rip-off. Deleting similar text would leave the authors with almost nothing.
The mechanism for “inadvertent copying” intrigues me. Could this happen if you fall asleep at the keyboard and start to dream and your fingers wander to do select-copy-paste? Or is it the the consequence of some malware or mind-alering substances? Perhaps the notice itself needs correcting and the problem is the quotes had a time out, due to the spillage of liquid onto the keyboard?
Dave, we humans are stuck in thinking time is linear. Read Kurt Vonnegut’s “Slaughterhouse Five”, then it is not so straightforward who plagiarised from whom. How do we know it was not/will not be/will not have been actually Robinson and Oshlack who ripped off/will rip off/will have ripped off from Zhou, Lin and Zhang?
It is bound to happen when you own a Guttenberg keyboard:
http://deskthority.net/keyboards-f2/the-guttenberg-keyboard-meme-t189.html
A few additional details:
Impact Factor = 1.54
Publisher = Springer Science + Business Medium
http://en.wikipedia.org/wiki/BioMed_Central
“This journal is a member of, and subscribes to the principles of, the Committee on Publication Ethics (COPE).”
http://www.biodatamining.org/authors/instructions
Absolutely no information about issues related to publishing ethics, plagiarism, retractions, etc. in the IFA.
Editor board: apparently solid
http://www.biodatamining.org/about/edboard
Plagiarism: it is only such when it is detected PRIOR to publication. Afterwards, it is too embarrassing for the journal and authors, ergo it CANNOT be plagiarism.
I hope authors will eventually defend their paper by openly arguing that it was actually plagiarised two years ago due to the proven non-linearity of the time continuum.
The mechanism for “inadvertent copying” intrigues me.
“The cat walked on the keyboard.
“My cat is guilty of plagiarism, I have fined him one bowl of milk for this misconduct”
When in heck are we supposed to be copy-pasting from the articles we read again? Read paper, digest conclusion, state tidbit in introduction, cite paper. I do not see where control-C and control-V came into it.
This journal and those like it should openly declare that they don’t care about plagiarism. At least then they’d be honest.
The question arises as to what sort of actions these authors’ institutions take against those who are caught engaging in plagiarism and other unethical writing and publication practices. Other than corrections, If there are no additional consequences for individuals who commit these types of infractions, then we should consider fastening our seat-belts and get ready for an avalanche of cases.
I will provide a (somewhat) dissenting voice here. While plagiarizing should be avoided it is the quality of data that we should be ultimately concerned about. If what they reported is sound then I would (almost) let them off the hook. Problem is that the modification of the methods that they published is rather minor. To put it mildly, if my student came to me with these modification he would certainly earned a praise, pat in the back and a co-authorship if(when) we use this method in some real data analysis… independent paper, I am not so sure.
Now, about journals and plagiarism. On several occasions I have seen papers that allegedly went through detection software yet clearly had plagiarized elements that I could see by naked eye. Typically thresholds is set to 20-25% match. If you set it lower the number of false positives becomes too high. Herein lies the problem- you can get to, say, 19% match by copying half of the discussion, or by copying parts of sentences, or you can copy parts of sentences and shift the sentence from passive to active voice. The fact of plagiarizing remains, yet proving it let alone auto-detecting it is almost impossible without manual analysis.
But, YC, you are addressing the scientific elements of the case: There would be no wrong-doing per se other than the student not being very original by his/her use of a highly similar methodology. But, what about the plagiarism? Suppose a student came to you with a paper with similar amounts of plagiarism as in the Bio-Data Mining. What would have been your response?
In matters of plagiarism, it is the lopsidedness of how we seem to treat researchers vs. students -sometimes within the same institution!- that I find to be highly problematic.
The journal will be renamed, “BioData Cutting and Pasting” in the near future.
That’s all very well, but how long will it be before he is again allowed to do research unsupervised ?
My cat is looking forward to returning to his groundbreaking research on rodent models.
It is also interesting to note, that the only peer-reviewer of this article is also one of the editors in chief of the journal (Professor Jason Moore – sort of a conflict of interest?)
http://www.biodatamining.org/content/7/1/15/prepub (kudos though for open peer review)
http://www.biodatamining.org/about/edboard
What I find rather shocking is, that Professor Moore is pretty well versed in the world high-throughput sequencing (judging by his extensive Twitter feed), but did not pick up the plagiarism of the original Robinson & Oshlack paper which was published in Genome Biology (also a BMC imprint), and is a pretty well known RNA-Seq related paper – I am not an expert in the field, but I immediately picked up the similarities… see e.g. PubPeer comment on the paper:
https://pubpeer.com/publications/25285156
And on PubMed Commons:
http://www.ncbi.nlm.nih.gov/pubmed/25285156/#comments
In reality this has happened to me (more than once) – cat pawsteps that is – not ‘saving and submitting’ — in part as she is too modest to seek co-authorship