Wednesday, February 15, 2017

A peculiar case of remarkable similarities

Tables 4 and 5 from Mandapati et al. and Seelam et al.
Via Twitter, Lana Hiscock found some very interesting similaries between two Tetrahedron Letters papers. She also wrote it up on Reddit: 
"Recently I discovered two almost identical, and in places copied and pasted, articles on synthetic methodology (both published within the last 6 months). Amazingly, neither journal has any link on their website which I could find for reporting apparent plagiarism. Not only that, but I caught the outright copying easily and it's something that should have been picked up in peer review.... 
...Additionally, and highly suspiciously, they all refer to water as a "grenary" (or "greenary" in one case) solvent. If you Google "grenary solvent" you get the ChemistrySelect paper as the fourth result. If you can read the articles, it's obvious, in my opinion, that these papers are all related to a common ancestor."
In her Reddit post, she points to four separate papers (a 2016 Tet. Lett., a 2017 Tet. Lett., a Synth. Commun. paper and a ChemistrySelect.). This Reddit comment from Auntie Markovnikov indicates there may be both plagiarism and self-plagiarism going on.

I haven't been able to check all four papers, but I have downloaded both Tetrahedron Letters papers and as you can see above, there are remarkable similarities in the methods table.

Amusingly, if you search a phrase in both of the Tetrahedron Letters papers ("The reactions are rapid and facile and accomplished at room temperature"), you find yet another paper, this one from 2011 and published in Green Chemistry Letters and Reviews. The subject matter is rather similar; one wonders if this is the original source.

As more and more of the literature becomes open-access, isn't there some way to program bots to crawl the literature and look for plagiarized text? Might be a useful thing... 

13 comments:

  1. ACS uses some antiplagiarism software. It has caught us redhanded within Computational Methods, which is quite funny as we of course do not change parameters of simulations when we change from one protein to another...

    ReplyDelete
  2. If you don't think that plagiarism is a major problem in modern scientific literature, you've never looked.

    ReplyDelete
  3. At least the reactions aren't identical, just one is an extension of the other. Copying substrates is fine IMO, as why waste money ordering -and/or a grad student's time in making- new contrived substrates that, like with most methods papers, tell you almost nothing other than you can fill out a table.

    ReplyDelete
    Replies
    1. I think your points are fair, but there are a number of exact copied sentences in the text as well.

      Delete
    2. This is supposed to be scholarship. Science Madlibs isn't scholarship.

      Delete
    3. Of course it's actually pretty badFebruary 15, 2017 at 12:41 PM

      According to grant agencies it is. In fact, long substrate tables (a certain prominent group now makes tables in the 100's...) are valued more highly than novelty, both by grant reviewers and journal editors. How I would wish for people to value scholarship; however, this is not the climate we live in (and why organic synthesis is rightly dying).

      Delete
    4. Does anybody else think that substrate scopes are the most boring things in the world? Even more so when they use almost identical substrates. Hmmm...pentanol gives 85% yield and hexanol 86%. Better try heptanol and octanol too.

      Worse, they usually don't examine the effects of changing substituents other than %yield (measured by NMR, based on conversion, blah blah.)

      Delete
    5. gratifyingly, ortho methyl was tolerated in addition to meta and para methyl...

      Delete
    6. "gratifyingly, ortho methyl was tolerated in addition to meta and para methyl..."

      Ah, that gratifying moment when your ortho methyl group reacts similarly to para methyl. Groundbreaking stuff.

      Delete
  4. To me this appears to be more of a symptom of the "quantity over quality" aspect of publishing. I would argue that these publications would make up one good article rather than four communications. At the very least combine the two cobalt-catalyzed ones, as the make the isothiocyanates in one, and then in the next paper they do the same thing and then add amine to make thiourea.

    I know the fear of getting scooped, and/or the fear of not having enough publications when you apply for tenure is the main reason, but it leads to overall dilution of the literature.

    ReplyDelete
  5. This is not "bots to crawl the literature", but there is a program that can check statistics in papers.

    http://finance.yahoo.com/news/scientists-furious-famous-psychologist-accused-164203508.html

    Scientists are furious after a famous psychologist accused her peers of 'methodological terrorism'

    Business Insider
    Rafi Letzter
    Business InsiderSeptember 22, 2016
    "The science of psychology is in crisis."

    "One by one, many of its flashiest and most famous results have collapsed in the last decade as a new generation of researchers have re-examined famous findings."

    "Forcing a smile can't really make you happy. The smell of chocolate cookies doesn't make you a better test taker. A robot named 'statcheck' is trawling through published papers for statistical errors, and posting public comments when it finds them."

    "It's a scary time to be a member of the psychological academic establishment. As the social psychologist Michael Inzlicht wrote on his blog back in February, 'Our problems are not small and they will not be remedied by small fixes. Our problems are systemic and they are at the core of how we conduct our science.'"


    https://mbnuijten.com/statcheck/

    statcheck
    "R package “statcheck”: Extract statistics from articles and recompute p values
    (Epskamp & Nuijten, 2016)

    "Conclusions in experimental psychology often are the result of null hypothesis significance testing. Unfortunately, there is evidence that roughly half of all published empirical psychology articles contain at least one inconsistent p-value, and around one in eight articles contain a grossly inconsistent p-value that makes a non-significant result seem significant, or vice versa."

    ReplyDelete
  6. More links on Statcheck, that checks for statistical errors in publications:

    https://www.theguardian.com/science/2017/feb/01/high-tech-war-on-science

    The long read
    The hi-tech war on science fraud
    The problem of fake data may go far deeper than scientists admit. Now a team of researchers has a controversial plan to root out the perpetrators
    by Stephen Buranyi
    Wednesday 1 February 2017 01.00 EST

    http://www.nature.com/news/smart-software-spots-statistical-errors-in-psychology-papers-1.18657

    Smart software spots statistical errors in psychology papers
    One in eight articles contain data-reporting mistakes that affect their conclusions.

    Monya Baker
    28 October 2015

    Researchers have created an automated software package that extracts and checks statistics from psychology research papers — and it has confirmed that the literature contains plenty of errors.

    http://www.nature.com/news/stat-checking-software-stirs-up-psychology-1.21049

    Stat-checking software stirs up psychology
    Researchers debate whether using a program to automatically detect inconsistencies in papers improves the literature, or raises false alarms.

    Monya Baker
    25 November 2016

    ReplyDelete
  7. It's another mediocre, by the numbers methodology paper, in Tet Lett published from India. You could probably find about a dozen others in the same issue. Good luck repeating any of the chemistry.

    ReplyDelete