In 2001, Chris Orvig was happy when his team had synthesized a molecule with potential therapeutic applications. He and his colleagues published their findings in a 2002 paper in Inorganic Chemistry.
Over a decade later, Orvig discovered a pivotal error in the paper: The authors had misidentified the compound.
Orvig, a professor of chemistry at the University of British Columbia in Vancouver, told us:
It was an honest mistake that we discovered in the last couple of years while trying to re-purpose the ligand (the title compound) for a new project.
Orvig explained that the issue first came to light a few years ago, in 2014 or 2015. Orvig had asked his PhD student, David Weekes, to re-examine several promising compounds the lab had synthesized in the past. But when Weekes—now a postdoctoral fellow in Orvig’s lab—repeated the methods from the 2002 paper, he was only able to create a molecule that was about half the size.
In November 2016, Orvig and his team contacted the journal to explain the mistake and submitted a new paper that corrected the literature.
Here’s the retraction notice for “Synthesis and Solution Studies of the Complexes of Trivalent Lanthanides with the Tetraazamacrocycle TETA-(PO)2:”
The authors retract this article on the basis that all of the characterization data were consistent with the proposed structure of the ligand being a 2:2 product of the reactants, but subsequent work showed that in fact it is a 1:1 product. Thus, all the solution chemistry and interpretations in the original publication were based on an incorrect assumption of the ligand structure. The corrected structure and related data are being published concurrently with this retraction notice, in an article entitled “Di- and Trivalent Metal-Ion Solution Studies with the Phosphinate-Containing Heterocycle DEDA-(PO)”, DOI: 10.1021/acs.inorgchem.7b01117.
The original article was published on January 31, 2002 and was retracted on August 15, 2017.
The now-retracted paper has been cited 13 times, according to Clarivate Analytics’ Web of Science.
Here’s the recent article “Di- and Trivalent Metal-Ion Solution Studies with the Phosphinate-Containing Heterocycle DEDA-(PO),” published in August, which details the correct structure and how the researchers discovered their error.
Anatomy of a molecule
Orvig told us that, back in the early 2000s, “we had a hell of a time characterizing the compound because we couldn’t grow crystals of it.” The team had relied on different, less precise methods for identifying the compound.
Weekes, however, had better luck growing crystals. And, according to the new paper, the crystal structure “proved unambiguously” that the molecule reported in 2002 was incorrect.
After contacting the journal about the error, Orvig says he and the editors discussed the best course of action. Ultimately, because “pretty much everything in old paper is wrong,” Orvig said, they agreed that a retraction was the best option, and the journal published the new paper.
Orvig also explained that the smaller compound reported in the new paper doesn’t have the same therapeutic potential as the larger one described in the 2002 paper. The compound needs to be large enough to encase a radioactive metal and deliver the metal — known for its therapeutic properties — to the bone to treat bone disorders. If the molecule is too small, it can’t wrap around the metal. Orvig explained:
It would be equivalent to grabbing a marble with your hand and completely encasing it in your palm versus trying to grab an American football.
We asked the journal whether it would have published the paper in 2002 with the different, less potentially useful structure. The editor-in-chief of Inorganic Chemistry, William B. Tolman, chair of the Department of Chemistry at the University of Minnesota, told us:
Your question is a difficult one to answer definitively at this juncture, but my sense is that the answer would be yes, the journal likely would have published the paper with the correct structure back then.
Given that the journal retracted the 2002 paper and replaced it with a corrected version, we asked Tolman if he would consider this case an example of a retract and replace. Retraction and replacement has become an increasingly popular option for some journals to correct honest, but pervasive mistakes. Tolman said that although the journal doesn’t really have that specific category of paper:
…essentially, this is what it is: a retraction, and then a paper that corrects the original work (which was NOT in error due to some ethical issue).
Orvig noted:
It’s important that we could correct the literature. I wonder how many people might have tried to make the ligand, failed, and moved onto something else.
Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.
A retraction for an honest error should not be in any way a detriment to the authors.
It’s doing the right thing.
As this example shows, the parts the paper that are not incorrect are free to be republished.
Why, if you are really a serious scientist, would you not want to correct the scientific literature containing an honest error? Pride? Hubris? Denial?
It’s a fact that the act of authors and their institutions fighting tooth and nail to keep a paper from being retracted demonstrates (to me at least) that honest error was not involved.
This fact is something that I believe all editors should keep in mind.
In some cases the reason to avoid issuing a retraction may be “I do not trust my institution or my peers–I believe they will hold this against me.” If this mistrust is in fact warranted, it is hard to attach too much blame to the authors (though they should try to move to a better institution if they can!)
All academics have a duty, when serving on tenure or promotion committees and reviewing papers, to make sure that researchers are never punished for doing the right thing. (It’s also good to try to avoid rewarding doing the wrong thing. For example, an implausibly high publication rate should be treated as a red flag, not a badge of honor. Hardly anyone can really publish a good paper every month, yet there are researchers whose CVs indicate they have….)
I’m sympathetic to that argument to a certain extent. Institutions and esp peers need to have a better understanding of the difference between an honest mistake, which is part of the natural scientific process, and misconduct.
I suggest a change in terminology. Withdrawal for a honest mistake. Retraction for misconduct.
Admittedly the name Withdrawal Watch doesn’t have the appeal of RW.
Yes an early career investigator with limited grant support publishing a paper a month is a red flag. It’s possible for lead of a large heavily funded consortium to achieve that level of productivity, but it’s a team effort.
Do you have any examples of the early career researcher and the heavily funded lead in a consortium?
“Withdrawal” is a euphemism Elsevier uses when retracting “articles in press”. In withdrawal situations, they claim since the article is not officially published, they also do not have to provide a retraction notice. It’s a way to obfuscate. There’s examples of this on RetractionWatch but you’d have to go back a few years.
Elsevier should change their terminology.
That actually sounds more like a recall, a term already in use for a recent email sent in error.
I still like the term withdrawal for a paper with serious, but honest error(s) that is confusing the literature. Otherwise, just correct the honest mistakes.
Using withdrawal for an honest error would put more teeth into the term retraction, which would be reserved for misconduct or other ethical failings.
I have a feeling wide adoption of this terminology (COPE?) would encourage corrections and even withdrawals (a good thing) to avoid the stigma of a retraction.
The comments make good points. Still, I wonder about situations where tenure or promotion or receiving a grant or award were involved that hinge on a single, high impact paper or results. Would the committee have decided favourably if that paper had not been published? Just as importantly, would SOMEONE ELSE have gotten the tenure, promotion, grant, or award instead or have at least been looked upon more favourable? Hard to tell of course, but are there any examples of litigation where say someone who was denied tenure concurrently with someone who was awarded it but who eventually had to retract (or, I guess in a more cut and dried case, had conducted gross misconduct etc)? It seems like a zero sum type of argument could be made and a court might buy it.