About these ads

Retraction Watch

Tracking retractions as a window into the scientific process

Chutzpah: Authors blame PLOS ONE for failing to find plagiarism in paper on Botulinum toxin

with 35 comments

plosonelogoHoly Chutzpah, Batman! A team of researchers in India has retracted their 2012 paper in PLoS One on botulinum toxin for plagiarism — while blaming the journal for failing to use its “soft wares” to catch the plagiarism.

The article, “Small-Molecule Quinolinol Inhibitor Identified Provides Protection against BoNT/A in Mice,” was written by a group from the Defence Research and Development Establishment, in Madhya Pradesh.

According to the retraction notice:

It has been brought to the attention of the PLOS ONE editors that substantial parts of the text in this article were appropriated from text in the following publications:

Identification and biochemical characterization of small-molecule inhibitors of Clostridium botulinum neurotoxin serotype A.
Roxas-Duncan V, Enyedy I, Montgomery VA, Eccard VS, Carrington MA, Lai H, Gul N, Yang DC, Smith LA.
Antimicrob Agents Chemother. 2009 Aug;53(8):3478-86

Eubanks LM, Hixon MS, Jin W, Hong S, Clancy CM, et al. (2007) An in vitro and in vivo disconnect uncovered through high-throughput identification of botulinum neurotoxin A antagonists. Proc Natl Acad Sci USA104: 2602–2607.

PLOS ONE therefore retracts this article due to the identified case of plagiarism. PLOS ONE apologizes to the authors of the publications above and to the readers.

The corresponding author, Ram Kumar Dhaked, took to the journal’s website to comment on the retraction. In a Sept. 6, 2013, post, titled “The response of the editors to the paper,” he wrote:

I have discussed your response with a Senior Editor. We appreciate that the data reported in the PLOS ONE article are original, and there have been no allegations about misattribution of data., however, the overlap in text with that from previous publications is substantial, and you have acknowledged that te [ends here]

A week later, Dhaked wrote:

There was some text overlapping in the article especially in introduction. That was went un-noticed by us. However, PLOS ONE could have checked for the overlap as they have soft wares etc. The editorial board has written to us that Data is original and there is no allegation of results attribution. This is very unfortunate.

Unfortunate indeed. Because the authors might be looking at more than embarrassment. Here’s their funding statement, which lists U.S. government support:

National Cancer Institute, National Institute of Health, USA acknowledged for providing chemicals bearing NSC prefix. Ms. Padma Singh is a JRF working in ICMR (Indian Council of Medical Research) scheme (Ref No: 3/1/3/JRF-2008/MPD-78, 31500). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

About these ads

Written by Adam Marcus

September 25, 2013 at 10:07 am

35 Responses

Subscribe to comments with RSS.

  1. “Holy chutzpah, Batman!” indeed! “PLOS ONE could have checked for the overlap”?! Why couldn’t the authors have done so themselves? Why is it up to an external body to hold the authors’ hands?

    Jennifer R. Ewing

    September 25, 2013 at 10:28 am

    • I defend neither party, but criticize both. The authors need to write more carefully. It is unclear EXACTLY how much text was plagiarized. So, PLoS One needs to come forth with exact information. These loose statements using adjectives like substantial and significant are totally unacceptable. The publishers need to start giving us, the scientific community, EXACT numbers of words that were plagiarised, and what text or source was plagiarized. This quantitative information should form a standard part of retraction notices that reflect a retracted a paper based on plagiarism. I don’t see where COPE specifies the quantitative amount of words that constitutes plagiarism. Why is COPE not forthcoming on setting up a CLEAR and quantitative threshold? I disagree with Jennifer Ewing. Free software does a (frankly-speaking) crappy job of detecting plagiarism. Google and other DB searches are insufficient and time-consuming. And authors are not always tech buffs. And many simply do not have the finances to pay companies for plagiarism detection for every document. So, there is a certain aspect of human and technical limitation (am I being too naive?). Nonetheless, the authors are partly correct in blaming PLoS One. I personally am tired of seeing these so-called hi-tech publishers who charge no less than 1500-3000 US$ for ONE PDF file, so yes, a good service, including plagiarism check, and a solid peer review, should be one of the responsibilities of PLoS One. One of the problems is the commercialization of ethics and plagiarism verification systems. Why hasn’t the scientific community come forward with FREE, robust and reliable software that is open for all scientists to use? If such software existed, I am sure that more scientists would run their papers through such software prior to submission, which would reduce the level of plagiarism, increase the level of self-awareness and reduce the risks of fraud related to plagiarisms. Thus, the levels of retrraction could potentially decrease. But because we have to feed the sharks at companies that sell the community a service that should be a free service, and keep the CEOs healthy and comfortable, such situations will continue to arise. We seriously need to get movement from the IT specialists who are a little more altruistic and can see a greater good of giving a pro bono product than to fatten their pockets. And no, I am not referring to the assistance of Soros.

      JATdS

      September 25, 2013 at 10:59 am

      • JATds, PLoS did give the sources from which text was plagiarized. You can compare the text in the retracted paper with the text in the source documents if you wish to do so.

        I have a bit of a mental disconnect with your post, though, because you seem to be saying that authors themselves need to be protected against plagiarism as if the latter was some sort of miasma that might inadvertently envelope one unawares.

        But writing papers isn’t like that. You avoid plagiarism by not copying someone else’s work! It’s that simple. The idea that one might need to put one’s own work through plagiarism software to check that you haven’t plagiarised is surely ludicrous.

        In this case, and in every case of plagiarism the fault is ALL with the authors. They copied other peoples text and passed it of as their own.

        chris

        September 25, 2013 at 1:27 pm

        • Metrics can be tricky, for reasons shown later, but let me suggest a software tool that would help a human display a text and its known antecedents in ways that would quickly make the similarities obvious, and then perhaps help with metrics.

          As an example of an end result, see examples, starting p.5. Text is on left, antecedents on right, formatting adjusted to visually align them.
          Word-for-word, in-order text is highlighted cyan, easily findable by automated text comparisons (an early version of which might be the UNIX diff(1) command from ~1970s). As soon as a reader trusts the cyan is done correctly, that text drops our of perception, allowing readers to focus elsewhere. It seems a waste to present a text and its antecedents in ways that force humans to scan both texts for similarity, when sufcdh has already been determined anyway.

          Having done the cyan highlight, trivial changes then become obvious, in these examples, highlighted in yellow, and then sometimes rearrangements of text become more easily visible. Yellow is mot obvious when a few words are changed amidst cyan passages. I have seen texts where 50% of the the words were cyan, 20% yellow, but all this was done by hand.

          However, it might be a nice MS project to display a text, then compare against one or more antecedents, generate candidate cyan (or other colors if found preferable), but then give a person a chance to adjust, select areas to be marked yellow, or another color for obvious paraphrases (very hard for computer). Much work could be done by computer, but still, human intervention is very useful.

          John Mashey

          September 25, 2013 at 3:06 pm

        • Isn’t it amazing how a set of inverted commas can convert a plagiaristic text into an acceptable text. Chris, I think you misinterpret what I had to say. I am in no way suggesting that sloppy authors like these need any form of protection. I am saying that they require a fair system, and that part of the responsibility is in fact that of PLoS One. I mean, honestly, did PLoS One explain why they DIDN’T find the plagiarism, assuming they use iThenticate, etc? As I say, I am not defending the authors, I am just saying how could plagiarized text get under the supreme radar of the PLoS One screening system? And why are authors not given the benefit of the doubt? If indeed there was plagiarism, why can’t the authors be given an opportunity to correct that text that was plagiarized, and allowed simply for an erratum with a statement that plagiarized text had been corrected? I am not sure why the community has to go from one extreme to the next? Surely an erratum with a statement of guilt is punishment enough, especially if the scientific data is sound?

          JATdS

          September 25, 2013 at 3:47 pm

          • for some countries publication charges are USD 500 and for some countries it is free!

            Ressci Integrity

            September 25, 2013 at 6:41 pm

        • Of course you are right! If you are copying from another text you know you are plagiarising, no need for a d… software!!

          Toby

          September 25, 2013 at 4:39 pm

      • PLoS is a nonprofit publisher; the publication fees cover actual production costs, and do not make anyone rich.

        Anonymous crystallographer

        September 25, 2013 at 4:00 pm

        • Is there any evidence that this is true. Assuming that not everyone pays to publish, they get maybe 1000 per paper which works out at around 24 million last year. Currently they are publishing about 3000 articles/month. As recently as mid-2010, they were publishing 500/month. Has their staff increased 6-fold to cope with this increase?

          I don’t know the answer to these questions but I am interested to find out.

          Gerry

          September 25, 2013 at 4:51 pm

          • This is really not difficult information to find. Their finances are discussed here:

            http://www.plos.org/about/what-is-plos/financials/

            Moreover, while the compensation rules for non-profits are somewhat flexible, operating as a for-profit enterprise while registered as a non-profit with the IRS is a good way to get audited.

            Anonymous crystallographer

            September 25, 2013 at 5:53 pm

          • My friends who have published in Plos One have not paid one cent. I am rather confused at these claims that everyone has to pay?

            CR

            September 25, 2013 at 6:15 pm

        • Yes PLoS as an organization is non-profit. But I always thought that PLoS One was a vehicle that generates income – to compensate for the more high-end/low-revenue PLoS journals.

          Arthur

          September 25, 2013 at 5:48 pm

          • Think of it as a tax on salami-slicers and uninterpretable phenomenological studies the world over.

            BoDuke

            September 26, 2013 at 12:45 pm

  2. I think an even greater concern is that the corresponding author apparently writes incoherent messages on web sites.

    It’s not just typos, bad grammar, or the fact that Indian English is somewhat different from western English. I really can’t understand what Dhaked was trying to say in those comments.

    Dan Zabetakis

    September 25, 2013 at 11:02 am

  3. what kind of western blots are they? very strange! Journals need to be careful.

    Ressci Integrity

    September 25, 2013 at 11:06 am

    • Figure 3 is a very obvious copy/paste. How could reviewers not see that?

      Arthur

      September 25, 2013 at 11:27 am

      • 24000 papers published last year. One would wonder how it is possible to get enough good reviewers.

        Gerry

        September 25, 2013 at 11:44 am

        • well, they have more than enough editorial board members, they could review the articles submitted. I thought that isone of the reasons they have huge number (i.e. almost 5000) of editorial board members (check this out http://www.plosone.org/static/edboard.action#)

          Ressci Integrity

          September 25, 2013 at 11:51 am

          • Right… So each paper submitted to PLOS ONE needs at least two reviewers, and there are about 25000 to 30000 papers submitted there each year, which means each editor has to be reviewing 10 to 12 papers a year for PLOS ONE alone on top of their PLOS ONE editorial duties, their other commitments and their real jobs.

            Umm, no. Especially since every paper I’ve ever been asked to review for PLOS ONE has been rubbish and sometimes the English has been barely intelligible.

            Bobo2

            September 25, 2013 at 3:32 pm

          • Bobo2. I am curious, do PLoS editors or reviewers get remunerated? If so, please indicate actual amounts per paper or job.

            JATdS

            September 25, 2013 at 4:34 pm

          • ten to 12 papers a year for their own journal is not bad. I review many more articles for different journals in a year. For being in the editorial board, they can do their part at least.

            Ressci Integrity

            September 25, 2013 at 6:39 pm

  4. ” PLoS editors or reviewers get remunerated?”

    I can answer. No.

    Dan Zabetakis

    September 25, 2013 at 4:48 pm

    • Thanks for that. I are back my comment, it appears that, financially at least, things are sound. I think I was letting my inherent bias against their model color my attitude.

      Gearoid

      September 25, 2013 at 7:01 pm

      • “Take back”. Sorry

        Gearoid

        September 25, 2013 at 7:02 pm

  5. Somewhat at random, I copied this phrase from the retracted article,

    mechanical ventilation which is necessary once BoNT-induced paralysis compromises thoracic muscle contraction

    popped it into Google, and the first hit is

    Pharmacophore Refinement Guides the Rational Design of …
    http://www.ncbi.nlm.nih.gov › Journal List › NIHPA Author Manuscripts
    by JE Nuss – ‎2010 – ‎Cited by 4 – ‎Related articles
    Oct 14, 2010 – Botulinum neurotoxins (BoNTs) are the deadliest of microbial toxins. … via the proteolysis of SNARE proteins, terminates neuron-to-muscle signaling, resulting … only after enervation) and 2) mechanical ventilation, which is necessary once BoNT-induced paralysis compromises thoracic muscle contraction.

    with URL http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2992458/

    NIH Public Access
    Author Manuscript
    ACS Med Chem Lett. Author manuscript; available in PMC 2011 October 14.
    Published in final edited form as:
    ACS Med Chem Lett. 2010 October 14; 1(7): 301–305. doi:10.1021/ml100056v.

    This article was not even listed as one of the sources of plagiarized materials in the retraction notice, so the list of plagiarized articles is by no means complete.

    Furthermore, this article is not listed in the references of the retracted article.

    The sentence following the one I Googled is interesting.

    From the retracted article:

    “However, the latter form of treatment would also be impractical, even a limited act of bioterror employing BoNT(s), as critical care resources would likely to be overwhelmed.”

    From the plagiarized original:

    “However, the latter form of treatment would also be impractical following even a limited act of bioterror employing a BoNT(s), as critical care resources would likely
    be overwhelmed.”

    The Indian English “tells” include the additional “to” in “likely to be overwhelmed” so the plagiarism was not purely copy/paste, but rather it appears that the Indian authors read the original and retyped the material, introducing and deleting various particles/articles/prepositions etc. that make English such a vexing language to learn and understand.

    There is no way someone in India would innocently type out such a sentence, matching word for word (save the vexing particles/articles etc.) with a previously published article.

    The plagiarized sentences do not appear in quotes, with a reference to the original, as any such closely matching phrases should appear in a decent scientific article.

    So, JATdS, how exactly should PLoS One report such plagiarism? Can you provide the algorithm to allow EXACT reporting of the number of words plagiarized? Do you count the extra “to”? Do you count, or not count, the omitted “following”? If the PLoS One report includes the “to”, is their report EXACT? Since the Indian authors dropped the “following” and added the “to”, is it now original?

    If I can grab a fairly random phrase from the article and get a hit so readily from Google, my sympathies are with the editors of PLoS One. Tracking down and reporting every bit of plagiarized material for this paper would require substantial effort, far more effort than reasonably required to readily demonstrate unacceptable levels of near-verbatim copying resulting in the charge of plagiarism.

    For reviewers, iThenticate and other specialized software is not essential, given the awesome text matching algorithms in Google software.

    Steven McKinney

    September 25, 2013 at 7:53 pm

  6. Maybe the problem is that I’m not natively English-speaking (Italian) but why it appears that I’m the only to consider this as a minor issue? I’m not referring to the PLoSONE paper specifically but more generally to plagarism in the introduction. The introduction is usually the section where most of the citations are inserted, isn’t it sufficient to correctly cite all the articles? Do you have to produce an original piece of prose for every manuscript you write?

    I was a little sad reading some comments regarding discounts for developing countries… as it seem that it was subtly suggested that because they are paying less they doesn’t deserve a full and fair editor and peer-reviewed process (but maybe it is because I can’t fully understand the tone of the posts).

    Sorry for my sloppy English/Italian

    Andrea Coletta

    September 26, 2013 at 3:06 am

    • I don’t agree at all with this argument,

      First, the introduction is where an author gives the background to the field in which they are supposedly expert and explains their motivation for performing the specific work.

      If they can’t do that in their own words, they really shouldn’t be publishing in the field.

      Second, wherever they did it, it’s fraud to take someone else’s writing and pretend it’s yours.

      CBus

      September 26, 2013 at 10:43 am

    • “The introduction is usually the section where most of the citations are inserted, isn’t it sufficient to correctly cite all the articles?” No, you have to make it clear you are copying another source by using quotation marks like I did here, both for the sake of clarity and honesty.

      It is nice having an Italian colleague bringing up this issue. I am Brazilian and I have seen awfully many Brazilians are unaware of the importance of correct referencing and also tend to tolerate what they call small innocent plagiarism, specially in the introduction. Many of my Brazilian colleagues claims this is inherently cultural because non-English speakers MUST resort to copy/paste in order to write a good paper, but I think this is fundamentally a major educational issue — that no one explained referencing practices to them clearly (and about the existence of copyright) and they took little time learning English and plus do not want to invest any time/money on good translators/reviewers.

      What do you or others think?

      CR

      September 26, 2013 at 1:28 pm

      • I do confirm, same feeling on this side. Brazilians are amazingly used to this. Almost irreversibly.

        DEUS

        September 26, 2013 at 1:55 pm

        • Thanks Deus for your impression. Of course I am not generalising here, but this is even hard to do as the practice is widespread. I really think the “small plagiarism” practice has been imprinted in Brazilian academia by omission. This can be easily fixed with a lot of exposure and corrective actions plus educational measures but it seems to me the system leaders do not wish to bring this up for their own sake…

          CR

          September 27, 2013 at 5:25 am

      • I agree with you, but i think also that “est modus in rebus” (from both sides). Can auto-plagarism in the Introduction be tolerated? Moreover when you work in the same field for 5 years writing, let say, 10-15 articles on the same general topic, it is very hard to create new prose and not to accidentally copy some sentence. I was not refferring to people that completely create an introcution by cut & paste, so I thought that a zero-tolerance approach on that type of plagiarism could be considered eccessive, am I wrong?

        losherpa

        September 27, 2013 at 5:48 am

        • One view you may find interesting: http://www.authoraidem.org/ . Go to Researchers (left-hand side of the screen, then Advice for authors, then click on Self-plagiarism duplicate secondary publication.

          Karen Shashok

          September 27, 2013 at 8:09 am

        • Yes, I have also heard much of this line of thought, and I think — (actually it seems to me many scientists think this way) — that one paper is supposed to discuss and add to a previous one building a line of thought.

          Thus repeating what was stated becomes unnecessary. Usually one can just say, “this paper is part of a series of paper investigating on … (e.g. XXX and YYY) which in brief is an important aspect of … because of …”, and this should be enough general introduction for your faithful reader! Then you could add the conclusions of the previous papers from your series and explain why another one yet on the same topic was necessary, and presto, you have made your reader happy and informed.
          I not only do not see any need to keep repeating oneself, while also I think self-plagiarism directly infringes copyright laws and must deeply frustrate readers who have to download several papers (often pay for them) only to find that considerable % is just rerun.
          I actually have a long series of papers (some 20) on the same specific topic and I have never felt like I was rewriting the same paper again. One simply either adds to the other or just confirms one impression and this must be said directly on your paper and maybe make it shorter and clearer, for the sake of our good readers — who easily get bored –, and editors — who like spending time/money on truly original and relevant material.

          What others think?

          CR

          September 27, 2013 at 9:14 am

    • ” The introduction is usually the section where most of the citations are inserted, isn’t it sufficient to correctly cite all the articles? Do you have to produce an original piece of prose for every manuscript you write?”

      Short answers: No; Yes. Putting together a decent introduction takes effort and time, for both native English speakers and foreigners. Ditto for the abstract.

      SA

      September 27, 2013 at 6:14 am

  7. @Adam and @Ivan. This is a bizarre fraud story to say the least, I think it deserve a RW post:

    http://www.nature.com/news/mystery-over-obesity-fraud-1.13810?WT.ec_id=NATURE-20130926

    Junk Science

    September 26, 2013 at 2:00 pm


We welcome comments. Please read our comments policy at http://retractionwatch.wordpress.com/the-retraction-watch-faq/ and leave your comment below.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 35,064 other followers

%d bloggers like this: