Signals Blog

Co-authored with Dr. Zubin Master, Assistant Professor at the Alden March Bioethics Institute, Albany Medical College and Research Associate at the University of Alberta’s Health Law and Science Policy Group.

In the current discussion about the now controversial SCNT human cloning paper by Shoukhrat Mitalipov et al., much has already been said of the reported errors that resulted in the duplication of images and plotted microarrays, as first reported in PubPeer, and whether such examples are evidence of falsification of data. To their credit, both Mitalipov and Cell Press moved quickly to respond to these allegations. Emilie Marcus, Cell’s Editor-in-Chief, stated that the image problems were “minor errors made by the authors when preparing the figures for initial submission” and that “we do not believe these errors impact the scientific findings of the paper in any way.” As reported in Nature News, Mitalipov affirmed that the results are real, but admits:

…he wanted to rush publication this time so that he could present results at the International Society for Stem Cell Research meeting in June. “Maybe it was rushed. But it was nothing to do with Cell,” he said. “It was my mistake.”

Even if there was no intentional fraud, Mitalipov’s rush to publish made their paper sloppy. Huge amounts of carelessness (generally not done intentionally) could appear as fraud, making people skeptical of the data and they may lose trust in the researcher(s).

The pressure to be the first to publish is huge, not just for authors, but also for the journals. As one anonymous commenter on PubPeer pointed out:

Cell has already a fast turn-around time (3 weeks). That’s fast compared to other journals. 4 days from receipt to acceptance, however, lets me suspect that the authors (and the editor) were afraid they would get scooped. The fact that no competing manuscript has been published (yet) makes me wonder about the rush.

Others have questioned whether Cell had been influenced towards expediting the publication, presumably so that Mitalipov could present published, rather than unpublished work at the ISSCR meeting next month – though why this distinction would be so critical is also worthy of debate. This in itself is troubling, but Mitalipov has only stated a desire for publication pre-ISSCR, not what his other motives might have been for a quicker publication – we can only speculate here. One wonders why a research team would, in essence, undermine all the long, meticulous work in conducting and validating their research by not taking the necessary time needed to ensure the accuracy and quality of the submitted publication. Was it so critical that they could not take another week to review it more thoroughly? The same question could also be asked of the publisher and reviewers.

As to the journal’s involvement, Cell points out, a rapid turnaround doesn’t mean the paper received inadequate peer review. Nor does it mean that a peer review period of 30 days will be more robust. It is not necessarily up to peer reviewers to pick up on image manipulation or duplication, which generally falls under the purview of the publisher, and many publishers use image manipulation and plagiarism detection software to identify both types of misconduct. We were unable to find mention on its website whether Cell uses such software prior to publication. Yet, as stated in the journal’s reviewer guidelines, one of the key features of the review is:

A summary of the specific strengths and weaknesses of the paper. In this regard, we encourage referees to comment on the quality and presentation of the figures as well as the validity of the statistical methods used to interpret them. (If necessary, the editors can obtain primary data from the authors for referees’ use in these more detailed evaluations.)

Based on the fact that the duplicated images were allowed to be published in this instance, it would appear that neither the reviewers nor Cell took steps to verify the quality of the figures in question prior to publication.

A comment by Natalie DeWitt on Paul Knoepfler’s blog suggests that it “[w]ould have been better to submit the cells for independent analysis and submit along with the paper.” Should researchers need to have other scientists replicate the data before publication? Publication of Mitalipov’s earlier work in cloning was delayed six months by Nature while the methods were independently replicated. This might be overkill. Surely some form of due diligence is needed by the publisher, through peer review, and we trust researchers to submit things accurately and honestly. But science is a self-correcting system in that results (in most cases) are meant to be replicated or refuted. Yes, we can say that a six month delay would permit others to replicate the experiment but does this mean that all experiments require such robust scrutiny prior to publication? Needless to say, you would think that Cell would be a bit more careful given the previous Hwang cloning scandal in 2006 and wouldn’t be so rushed to accept the manuscript for publication.

Being careless when conducting science or when reporting science is a problem. But it is not an excuse. Carefulness is a principle in the ethics of science and the responsible conduct of research, as many have noted. Carefulness is vitally important to science: it promotes trust and cooperation among scientists and thus being careless, by whatever means it arrives (rushed work, sloppiness) can be said to diminish trust.

We’re not implying that Mitalipov’s careless errors were intentional. Unlike the Hwang case, there does not appear to be any falsification of data on behalf of Mitalipov. Despite the unfortunate situation he and the stem cell community now finds itself in, there are positives: first that Mitalipov (and Cell) responded quickly and have admitted to errors in haste (Mitalipov has graciously and correctly shouldered the blame), and he has openly offered to share their cells for further analysis and verification. All of these measures should go a long way to quell controversy surrounding the paper, alleviate potential distrust and eventually, prove that the paper is worthy of the initial credit it received.

The following two tabs change content below.
Lisa Willemse

Lisa Willemse

Lisa is a science communicator with 15+ years' experience in the fields of regenerative medicine, child development and technology. She launched this blog (first as the Stem Cell Network Blog) in 2009, and served as co-editor until April 2015. She is currently the Senior Communications Advisor for the Ontario Institute for Regenerative Medicine and has recently contributed to Motherboard, Science Borealis and the Genome Alberta and Canadian Blood Services blogs. Follow her on Twitter and Medium @WillemseLA.