About these ads

Retraction Watch

Tracking retractions as a window into the scientific process

STAP stem cell papers officially retracted as Nature argues peer review couldn’t have detected fatal problems

with 12 comments

nature 714A significant chapter of the nearly six-month saga of the STAP stem cell controversy has come to an end, with Nature running retraction notices for the two papers involved. The journal has also published an editorial about the case that’s worth a read.

The retractions for “Bidirectional developmental potential in reprogrammed cells with acquired pluripotency” and “Stimulus-triggered fate conversion of somatic cells into pluripotency” both read:

Several critical errors have been found in our Article (http://dx.doi.org/10.1038/nature12968) and Letter, which led to an in-depth investigation by the RIKEN Institute. The RIKEN investigation committee has categorized some of the errors as misconduct (see Supplementary Data 1 and Supplementary Data 2). Additional errors identified by the authors that are not discussed in RIKEN’s report are listed below.

(1) Figure 1a and b in the Letter both show embryos generated from STAP cells, not a comparison of ES- and STAP-derived chimaeric embryos, as indicated in the legend.

(2) Extended Data Fig. 7d in the Article and Extended Data Fig. 1a in the Letter are different images of the same embryo and not, as indicated in the legends, a diploid chimaera embryo and tetraploid chimaera embryo.

(3) There is an erroneous description in Fig. 1a in the Letter. The right panel of Fig. 1a is not a ‘long exposure’ image at the camera level but a digitally enhanced one.

(4) In Fig. 4b of the Letter, STAP cell and ES cell are wrongly labelled in a reverse manner.

(5) In the Article, one group of STAP stem cells (STAP-SCs) was reported as being derived from STAP cells induced from spleens of F1 hybrids from the cross of mouse lines carrying identical cag-gfp insertions in chromosome 18 in the background of 129/Sv and B6, respectively, and that they were maintained in the Wakayama laboratory. However, further analysis of the eight STAP-SC lines indicates that, while sharing the same 129×B6 F1 genetic background, they have a different GFP insertion site. Furthermore, while the mice used for STAP cell induction are homozygous for the GFP transgene, the STAP-SCs are heterozygous. The GFP transgene insertion site matches that of the mice and ES cells kept in the Wakayama laboratory. Thus, there are inexplicable discrepancies in genetic background and transgene insertion sites between the donor mice and the reported STAP-SCs.

We apologize for the mistakes included in the Article and Letter. These multiple errors impair the credibility of the study as a whole and we are unable to say without doubt whether the STAP-SC phenomenon is real. Ongoing studies are investigating this phenomenon afresh, but given the extensive nature of the errors currently found, we consider it appropriate to retract both papers.

Austin Smith, the author of a News & Views article accompanying the original papers, has also retracted that:

In view of the fact that the authors of ‘Stimulus-triggered fate conversion of somatic cells into pluripotency’ (H. Obokata et al. Nature 505, 641–647; 2014) and ‘Bidirectional developmental potential in reprogrammed cells with acquired pluripotency’ (H. Obokata et al. Nature 505, 676–680; 2014) are retracting their reports (see page 112), I wish to retract the News & Views article, ‘Potency unchained’ (Nature 505, 622–623; 2014), which dealt with these studies and was based on the accuracy and reproducibility of their data.

Nature also published an editorial on the case. While the journal acknowledges that the saga has prompted a review of their processes for checking image manipulation, and admit to at least one mistake, they try to draw lines around what their peer reviewers could and couldn’t have caught, and which of the papers’ problems were “fatal” and which weren’t.

First, they characterize the issues readers found shortly after the papers were published:

As various media outlets including Nature’s independent news team reported, errors were found in the figures, parts of the methods descriptions were found to be plagiarized and early attempts to replicate the work failed.

The problems that initially emerged did not fundamentally undermine the papers’ conclusions. Moreover, replication of such work is not necessarily straightforward or quick, and the ability to use some techniques can be very sensitive to aspects of the experimental protocol.

The figure errors may lead to changes in how Nature checks images:

We at Nature have considered what lessons we can derive from these flaws. When figures often involve many panels, panels duplicated between figures may, in practice, be impossible for journals to police routinely without disproportionate editorial effort. By contrast, image manipulation is easier to detect. Our policies have always discouraged inappropriate manipulation. However, our approach to policing it was never to do more than to check a small proportion of accepted papers. We are now reviewing our practices to increase such checking greatly, and we will announce our policies when the review is completed.

The journal takes pains to say that the public post-publication peer review wasn’t what sank the paper for good; that took an institutional investigation:

But only since the RIKEN investigation has it become clear that data that were an essential part of the authors’ claims had been misrepresented. Figures that were described as representing different cells and different embryos were in fact describing the same cells and the same embryos.

Nature then makes an argument many journals have used when confronted with deception by authors: There’s no way for reviewers to catch those sorts of problems:

We at Nature have examined the reports about the two papers from our referees and our own editorial records. Before publishing, we had checked that the results had been independently replicated in the laboratories of the co-authors, and we regret that we did not capture the authors’ assurances in the author-contributions statements.

We have concluded that we and the referees could not have detected the problems that fatally undermined the papers. The referees’ rigorous reports quite rightly took on trust what was presented in the papers.

It’s always good to see journals note the limitations of peer review, but may we suggest that a group of editors get a list of peer review’s flaws together and publish them before the next scandal involving a hot field such as stem cells? (We also note that Cell, Science, and Nature rejected earlier versions of the STAP papers, according to Science.) Of course, that might undermine a major justification for journals’ attempts to prevent publicity about research before they’ve published it, aka the Ingelfinger Rule. But it would sure be honest.

Nature also highlights some important aspects of retraction policy:

The papers themselves have now been clearly watermarked to highlight their retracted status, but will remain hosted on Nature’s website, as is consistent with our retraction policy. (In our opinion, to take down retracted papers from journal websites amounts to an attempt to rewrite history, and makes life needlessly difficult for those wishing to learn from such episodes.)

About these ads

Written by Ivan Oransky

July 2, 2014 at 8:37 am

12 Responses

Subscribe to comments with RSS.

  1. A most curious defence by Nature. Those who engaged in post publication peer-review of the papers noticed problems with the data. However, Nature editors and their reviewers did not. It was post publication peer review that raised the questions. These questions led to the investigation at Riken and now the retractions. So post publication peer review and the documenting of the problems at Pubpeer

    https://pubpeer.com/publications/24476891

    https://pubpeer.com/publications/24476887

    led directly to the retractions.

    So I am not convinced by Nature’s argument, when the great unwashed of the peer community can pick out the problems on the published figures.

    ferniglab

    July 2, 2014 at 9:59 am

    • This is because it is essential to publishers to sustain the fallacy that traditional, hidden, moderated peer review offers any guarantee of publication quality. Open post-publication peer review does not need editors, and self-publishing does not need publishers. They are merely trying to survive.

      CR

      July 2, 2014 at 10:18 am

  2. Exhibit A: “We can’t possibly police all these images”
    Exhibit B: “The post-pub crowd who did our job for us had no role in the final outcome”

    In other words… “Trust us, we have this under control. Even when we get our rear ends handed to us on a plate by bloggers, AND in the same piece we admit this problem is beyond our abilities, we will never ever admit defeat.”

    Quite an appropriate response, for a week when the supreme court essentially legitimized religious belief by corporations as persons. The church of Nature prevails.

    Paul Brookes

    July 2, 2014 at 12:55 pm

    • Methinks the church of Nature is really a house of cards. If they were to honestly admit that current pre-publication peer review is fatally flawed, the entire foundation of their enterprise would be threatened, including their supposed ability to judge what is or is not worthy science (editorial pre-review decisions).

    • “The church of Nature….” And we all know that every deity has feet of clay!

      Akhlesh

      July 2, 2014 at 4:12 pm

  3. The lack of a rush to comment on this milestone might indicate that everybody knew this would happen months ago and can’t be bothered with yesterday’s news. Job done. Time to move on.

    But for some of us pedants here, it cannot be left to pass.

    “The problems that initially emerged did not fundamentally undermine the papers’ conclusions.”

    Cripes! Didn’t the initially emerged problems as revealed by concerned and diligent Japanese figure sleuths indicate that figures had been made up? How often do we have to hear from apologists that the conclusions have not been affected by transparently fabricated figures. Well, as it turns out, they were! “Null Points” for this whitewashing effort, as we Europeans like to say once a year in our best French accents.

    “Our policies have always discouraged inappropriate manipulation.”

    “Au contraire mon ami” as we Europeans also like to say for dramatic effect! The crappy low resolution figures that you inflict upon the scientific community are a disgrace in the internet age. They are an OPEN INVITATION to fraudsters to make up their figures. Isn’t fixing this a no-brainer? Some of your lesser Nature stable journals now publish something approximating to original data in their supplements. My theory is they do this because they have real scientists on their editorial boards who are pissed off by the effects of abysmally poor data presentation. (Remarkably, given NPG’s business model, those journals even have editorial boards!) It’s just a theory and I’m open to other suggestions though as to why they do this and the flagship Nature doesn’t.

    “We have concluded that we and the referees could not have detected the problems that fatally undermined the papers.”

    Er, the Science commentary indicates that the peer review process did function quite well on several occasions. That said, IT IS NOT the requested function of the referees to go through miniaturised Nature figures to spot image fabrications. Their main function is to advise the Nature editors on trendiness. Their secondary function, if they can even be bothered, is to advise on scientific flaws – logical deficiencies in the experimentation and interpretation, under the assumption that it is all honest and true otherwise why would it have been submitted to Nature? The Nature biology editors all have PhDs, but they either can’t recognise or don’t look for figure fabrications. Maybe they could employ David Hardman? Although, now I think about it, even paid as piecework, he might bankrupt them within the year…

    “A manifestation of these problems has been a growth in the number of corrections reported in journals in recent years.”

    Diversionary tactic! Equivalent problems were always there with microscopy images and WBs. Something has changed though in recent years. So welcome to Fraud 2.0 Nature! PubPeer, Retraction Watch etc. are looking at you. Thanks to them and also due acknowledgement to those earlier pioneering sites who have helped to set the ground rules for the current discourse about data quality amongst scientists. Well amongst some scientists, if not all of ‘em.

    “The Editorial cites a planned increase in the number of articles to be scanned for manipulated images as one measure meant to detect misconduct.”

    Arrgh!!! This means that they still don’t check all the accepted articles for image manipulation? Surely that is more cost effective than random checks at submission? Well unless they only check 10 or 20 or so per year?

    It would appear that if you give scientists a proper platform for discussion then science can be self-correcting. Otherwise it isn’t. Who knew?

    Scrutineer

    July 2, 2014 at 5:23 pm

  4. What will be the next “STAP” that will have been missed b/c many journals didn’t learn from STAP? IDK, but it’s coming surely.

    BTW, you all might find my Q&A w/Nature piece today on their editorial policies in the wake of STAP interesting: http://wp.me/p1xWpk-4wy.

    Cheers,
    Paul

    paulknoepfler

    July 3, 2014 at 4:16 pm

    • Yes, it is good that Nature entered into a dialogue with you to be placed on your ipscell blog. They are in full-on damage limitation mode but their replies to you are certainly more informative than the drivel they spouted in their editorial.

      By the way, your regular STAP Cell Polls were very much appreciated in this quarter. A rare and fascinating capture of PPPR in action. You have captured STAP Cell mind swing for all posterity. (Whether posterity cares about that is quite another matter).

      Scrutineer

      July 4, 2014 at 3:26 pm

  5. Hallelujah!

    Paul

    July 4, 2014 at 3:18 am

  6. Here are some potential problems in NIES.

    http://nieskaizan.doorblog.jp

    A

    July 10, 2014 at 11:00 am

    • I wish to confirm that NIES is extremely problematic, at least in terms of the lack of transparency and not willing to respond to critiques publically. I wrote this e-mail to two researchers who have already been called out publically, but who fail to respond.

      “On Tuesday, July 1, 2014 4:01 PM, [redacted] wrote:

      Dear Dr. Azusa OKAGAWA and Prof. Kanemi BAN,

      I have learnt of some queries about some of your published studies, listed at http://urx.nu/9HZ3 , and advertised at Retraction Watch: http://retractionwatch.com/2014/06/23/riken-inquiry-prompted-by-stap-stem-cell-controversy-generates-three-corrections/#comments

      I think it is important to first comment publically on these criticisms about your studies at Retraction Watch.

      Should no suitable response be made publically, then NIES and Osaka University need to be contacted to initiate an investigation, similar to the Obokata STAP stem cell papers.

      I hope that you will take this opportunity of clarifying the concerns made publically at http://urx.nu/9HZ3

      Sincerely,

      [redacted]“

      JATdS

      July 10, 2014 at 12:41 pm

      • I agree.

        I guess authors think only of their own self-protection and they have no sense of responsibility.
        So they never respond to critiques.

        I think it is necessary to report the concerns made publicly at http://www.webcitation.org/6QytQsxPC to settle the problems.

        So I would like authors of Retraction Watch to write an article of these concerns and to publish it.

        A

        July 11, 2014 at 2:53 am


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 35,846 other followers

%d bloggers like this: