A perfectly clean spectrum of a newly synthesized compound is a work of art to a synthetic chemist—it can require painstaking lab work to get rid of any signs of solvent and impurities. But sometimes a pristine spectrum isn’t possible, either for lack of lab equipment, researcher skill, money, or time.
In such cases, a researcher might touch up the data by removing the offending peaks in a spectrum. Although massaging data in this manner is rare and not as egregious as outright falsifying of data, it violates ethical standards of scientific research.
One journal editor has decided it’s time to pipe up about it. Amos B. Smith III, a chemistry professor at the University of Pennsylvania and editor of the American Chemical Society journal Organic Letters, has written an editorial—published online on June 5—to alert the chemistry community to “a serious problem related to the integrity of data being submitted” (DOI: 10.1021/ol401445g).
To help quash the problem, Smith has added a dedicated data analyst to the Organic Letters editorial staff to help check spectral and other characterization data. The move to add a data analyst follows the lead of ACS’sJournal of Organic Chemistry. The publications appear to be the only two chemistry journals to use data analysts to support editorial staff and external manuscript reviewers.
The editorial is a “warning shot” for chemists that they will be held accountable for data manipulation, comments John A. Gladysz, a chemistry professor at Texas A&M University and editor-in-chief of ACS’s journal Organometallics. Gladysz says the eventual fallout from the increased scrutiny could make it more acceptable for data presented in a paper to not look perfect.
Smith was prompted to write the editorial after uncovering several cases in which reported spectra had been edited to remove evidence of impurities. “Such acts of data manipulation are unacceptable,” Smith writes. “Even if the experimental yields and conclusions of a study are not affected, any manipulation of research data casts doubts on the overall integrity and validity of the work reported.”
He declined to provide C&EN details of the cases his editorial team found or to say if any punitive actions were taken. Punishment typically involves a publishing ban in the journal for the corresponding author.
Smith also opted to not comment further on the topic, preferring to let his editorial do the talking for him. “I am concerned about stoking any fires at this point,” Smith says. “I want to let it go and see how it plays.”
Analyzing the vast amount of data submitted to support the conclusions of a research paper is a challenge for editors and manuscript reviewers. And it’s not just spectral data that can cause a problem. Journals must ensure that data for crystal structures submitted with research papers support the structures being reported and are consistent with the paper’s chemical conclusions. In organic synthesis papers in particular, reporting inflated product yields, diastereomer ratios, and enantiomeric excess values have long been concerns (C&EN, May 30, 2011, page 50).
C. Dale Poulter, a chemistry professor at the University of Utah and editor of the Journal of Organic Chemistry, knows these problems well. They are part of the reason he enlists the help of a data analyst.
When it comes to checking reported data, a reviewer or data analyst must make sure the spectra, elemental analyses, and other data required by the journal are there, Poulter explains. The presented data are then reviewed to be certain there aren’t any blatant misinterpretations. Any anomalies reviewers find could be the result of a simple mistake—such as a typo, math error, or loading the wrong data set. Or they might point to data manipulation.
The occurrence of data manipulation remains rare, Poulter says—only about a dozen cases out of the 3,000 manuscripts submitted to his journal each year. These cases aren’t reflected in paper corrections or retractions, he notes, because the papers are not accepted for publication.
“Everything that improves the quality of the data published in the scientific literature is to be applauded,” says Peter Gölitz, editor-in-chief of Angewandte Chemie, which is published by Wiley-VCH on behalf of the German Chemical Society. Although he doesn’t use dedicated data analysts, Gölitz notes that, like other chemistry journals, Wiley-VCH publications have trained chemists as full-time editors who are scrutinizing manuscripts and supporting information along with the reviewers. “Given the total number of manuscripts and the diversity of topics, there are of course limits to what one editor, or what a data analyst for that matter, can do,” he says.
At Nature Chemistry, the editorial staff doesn’t include a data analyst, but spot checks are made on figures in papers to monitor for manipulated images, says Chief Editor Stuart Cantrill, who formerly served on the editorial staff of Organic Letters.
“We’re starting to look at ways of engaging with the community on the issue of making raw data available alongside spectra and other figures,” Cantrill says. There currently isn’t a widespread culture of sharing such data in the chemistry community, with the exception of files associated with X-ray crystallography, he notes. “We hope that doing so will reduce the likelihood of researchers reporting fraudulent data and increase the chances of detecting it if they do.”
“We are not out to catch people and sanction them, but to make sure papers are accurate,” Poulter says. “Our efforts are part of what I think of as a remarkable service for authors to help improve the quality of the data they publish.”