tech

RIKEN says stem cell research was falsified

65 Comments
By ELAINE KURTENBACH

The requested article has expired, and is no longer available. Any related articles, and user comments are shown below.

© Copyright 2014 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed.

©2024 GPlusMedia Inc.

65 Comments
Login to comment

Every scientist knows that groundbreaking research such as this will result in the experiments being replicated by other institutions. I wonder why she falsified these elements. She has probably ruined her career now.

8 ( +9 / -1 )

And another fraud discovered. How many makes that now? I lost count.

Also can we raise the punishment for falsifying "medical" researches, they should be punished severely. Its not some school paper quiz.

7 ( +10 / -3 )

This is not the end of the story. There is no chance this woman was solely responsible for fraud in a major academic journal. This was not a one person show.

I think the truth is going to be more complex.

7 ( +8 / -2 )

Just two comments on the article itself, not on the right or wrong of the situation. First, the writer identifies Obokata's place of work as the "RIKEN Center for Development Biology", while anyone who writes about this subject matter for an outfit like Associated Press ought to know that Obokata's research area is developmental biology, thus the correct name is "RIKEN Center for Developmental Biology". Get your facts right, Ms Kurtenbach. Second, she refers to Obokata as a "fashionable young woman". What's that got to do with anything, except that it slavishly repeats the stereotyping by the popular press. Surprised the equally irrelevant "she wears a koppogi in the lab" wasn't added for good measure. Stick to the relevant facts, Ms Kurtenbach.

6 ( +7 / -1 )

While falsifying data is unfortunately nothing new in university labs, my sense here is that Obokata is a scapegoat for a bigger issue at Riken. We shall see.

4 ( +7 / -3 )

My mind is blown by the continued white-knighting here. She was the paper's lead author (and she got the recognition for it), she recycled images from her thesis, no one was able to reproduce the same results. She is the main culprit here, and the investigation determined this, not some "scapegoat for being female"...

4 ( +7 / -3 )

There is no chance this woman was solely responsible for fraud in a major academic journal.

But she is the first author of this paper and will be liable for any fraud that was committed. I mean she wrote the paper, she put it together, so if she falsified the two sets of images (falsification is unfortunately quite common in Japan), she did it by herself, probably without the knowledge of the co-authors. The co-authors and the reviewers of the paper are of course also responsible for not having noticed that. The other question is how in a hell this paper could be accepted with apparently obvious falsifications of the images?

@rickyvee

It doesn't really matter if she fabricated the results. The point is that she tried to make them look better. Which means that they are probably not as appealing as the paper tried to make it appear in the first place.

Falsifications or manipulations like this one means that her career in science is not looking good at all at this point. And as cracaphat said, she should better come up to light and explain herself with solid arguments instead of hiding behind useless statement to the press. She was more than happy to put herself in those ridiculous TV shows and media channels when the news came out about her "research". Why isn't she more willing now when facing serious problems?

3 ( +5 / -2 )

”There is no chance this woman was solely responsible for fraud in a major academic journal."

Yeah, right. That journal did reject the paper at first and was quite dismissive of her "research." But it later accepted it. Hmmm. I wonder what kind of pressure it was under, given the head researcher's media and star appeal.

3 ( +6 / -3 )

On something as sensitive as this science, you'd think there would have been a least one independent replication BEFORE the paper was published in Nature -- Harvard prof not withstanding, and all her supposed mentors at Rikken.

3 ( +4 / -1 )

I have to admit that when Haruko Obokata, a movie-start-look and fashion-savvy young female scientist, burst into the scene, it was quite media firework display. :)

But, if one pays attention to the hype beneath the surface, one can certainly find enough evidences that attitude and tolerance have been built in today’s Japan already as to forgery and hoax from hotel chain food menu to music production.

It’s no surprise that Haruko Obokata will continue to act as if she were a victim of other person(s) or circumstance.

If people here still wonder how this episode may end, look no further, the predictable trajectory would that a group of people or an individual takes a deep o-jigi in front of cameras, showing regret and remorse. Then everything will be alright, and life goes on as usual - no one would get punished for wrong doings. (from the board of the directors at TEPCO to the person holds the high office in the gov such as Taro Aso. The list goes on and on, you name it.)

3 ( +3 / -0 )

Ok, so here's what we agree on. There is a contributor called Yoshiyuki Seki. The blog is run by Paul Knoepfler. That's about it I suppose.

Are you seriously suggesting that the sentence: "Note from Paul–Yoshiyuki now reports this green signal is determined to be autofluorescence." is referring to a person called Paul-Yoshiyuki, somebody not mentioned anywhere else on the site (or internet it would seem, at least not in connection with any kind of cells according to my Google-fu)? Do you seriously suppose that my scenario of Paul, the blog runner, commenting on Yoshiyuki's results is not more realistic?

Note also the long dashed line, not normally used with names per se, but rather as a separator of names, for example Lennard-Jones is one person, but Black--Scholes are two. You said that the case was clarified in the comments section and indeed it was; there "Yoshiyuki" writes in a thread that "Yoshiyuki Seki" started: "Jeanne: Thank you for your advise. Unfortunately, because I can detect Red channel, this signal is autofluorescence. I will retry. Thanks". I should also add that in the comments it was pointed out that Yoshiyuki Seki might not be an entirely 3rd party researcher, as he shares some affiliations with some of the authors of the STAP papers.

I am not saying that STAP cells are not real, I claim no expertise. What I am saying is that there are no successful replications despite a lot of effort put into it by scientists working in the field. In fact, some have come out clearly stating that they do not believe that the cells exist (I am of course referring to Prof. Lee). Your argumentation wholly rests on STAPs having been replicated by others, which is untrue.

3 ( +4 / -1 )

Yoshiyuki and Yoshiyuki Seki are the same person (of Kwansei Gakuin University). In the comments (of the blog) callname "Yoshiyuki Seki" starts a thread reporting on his experiments. Other people ask for details and the person who answers is called simply "Yoshiyuki". It is clear that these are the one and same person, but that simply he could not be bothered to write his whole name again and again. Having established this, let's turn to the last piece, Paul–Yoshiyuki. Is the supposed Paul–Yoshiyuki (who does not exists according to Google) the same Yoshiyuki, i.e. Yoshiyuki Seki? He is, for it is indeed "Yoshiyuki" who in the comments section writes about the autofluorescence results, which exactly matches the interpretation that Paul wrote and update on Yoshiyuki's progress.

I did not realize that you had not read a book before, or written large portions of text. Well, this must come as a surprise: there are different types of dashes. Nobody separates two names with two short dashes, but rather with one long one. Let me elaborate. Indeed it is true that the dash from Paul–Yoshiyuki "–" is singular, but so is "--" in any text editor for in fact the latter becomes the former (in Word/LibreOffice/LaTeX). Look up "em dash" on Google if you don't believe me (more specifically what Paul–Yoshiyuki contains is an "en dash", LaTeX for example has this as -- and the other long one, the em dash, as --- if I recall correctly. The usage of these dashes is of course different). I did not think using "--" in place of a clumsy copypaste or Alt-code could possibly be misunderstood by anyone.

There might have been two experiments, I'll give you that much for I am not an expert. However, if you follow the comments section of the blog, you'll see that the four day results too, are due to autofluorescence, and thus a failure.

If you Google Prof. Lee's attempts, you'll see that a few days ago the press heralded him as having succeeded in replicating STAP cells. This turned out to be a bit too soon, and upon further analysis, Lee found that the experiments had in fact failed, and is now of the opinion that STAP cells do not exist. Surely had Yoshiyuki, Yoshiyuki Seki, or Paul-Yoshiyuki succeeded, these would too have been reported by the media (even though this particular researcher is not completely independent) and mentioned in conjunction with the news of Lee's success.

I don't think I can make this any clearer: You are plain wrong. Instead of even trying to provide evidence of a successful replication (which surely would reported by other sources than just one blog), you keep making more and more complicated and elaborate excuses.

3 ( +3 / -0 )

Read the blog, there is absolutely no way that "Yoshiyuki Seki" and "Yoshiyuki" are different people. "Yoshiyuki" is answering to queries directed at "Yoshiyuki Seki" with such intimate knowledge that only "Yoshiyuki Seki" could possibly possess it. It's not like there are two Yoshiyukis talking about two different trials, but one and the same.

Having programmed internet applications for some 15 years now (for leisure, never professionally), I can tell you that -- turning into – is expected behaviour on (modern) web sites. Also, being quite the typography buff that I am, I assure you that – is an en dash, not the Japanese ー, which clearly looks different in every possible way and is far from the others (en&em) in the Unicode table. Finally, WordPress, the popular blogging platform (which the iPS cell blog is using) automatically converts -- into a long dash (I checked), so it is quite obvious that the blog owner wrote --, a separator of names. I assumed that the platform Japantoday is using would also work as expected and didn't bother to click "Preview", for even without the conversion it should be obvious to everyone what was meant. As you have clearly not have been using the internet for too long, I am not even going to go into why "--" is usually actually better than "–" on simple discussion forums, and why you might expect the latter to render completely wrong.

You put too much focus on appearances instead of the content. The green light on the iPS blog means nothing when on the very same page the results are refuted. Whether I write the long dash as -- or – should not matter, the meaning "a long dash" should be immediately obvious from the context and the usual behaviour of computers.

As for cell research, again, I claim no expertise. I have not been picking and choosing facts, in fact the tobacco stem cells you wrote about was why I initially joined in the discussion. If it is well known that external stress should create STAP-like cells, the ethical implications of Obokata et al.'s sloppy research are even graver. This I tried to point out with the Jan-Hendrik Schon example in my first post here. I am not saying that external stresses cannot convert cells to stem cells. I am saying that the evidence has it that the protocol of Obokata et al. is not the way to do this. This is to say that STAP cells, as defined by the protocol, do not exist, but STAP-like cells of course might.

3 ( +3 / -0 )

They said researcher Haruko Obokata, the lead author of the paper in Nature, had manipulated or falsified images of DNA fragments used in the research.

this was already widely known (and the fact that there was some plagiarism). i still haven't read anywhere that she fabricated the results of the experiment, which obokata and vacanti still stand by.

2 ( +4 / -2 )

It will be a real shame if the results cannot be replicated as this was (or at least seemed to be) a breakthrough that could have had amazing benefits for all of us. The question of interest to me is whether these manipulations actually mean that the underlying research is of no value or not.

2 ( +2 / -0 )

It would be good to hear more from Obokata than “I will file a complaint against RIKEN as it’s absolutely impossible for me to accept this,”

I mean, most people will be wondering what possible excuse she could have? Claim ignorance?

2 ( +3 / -1 )

The investigation was to find out about irregularities with the research report.

But noone has mentioned the most important fact; are STAP cells for real? Everything else is relevant only if the STAP cell is a fake in itself.

2 ( +3 / -1 )

Frungy, did you even read your source? You say some labs have been able to replicate it, but according to your own source, from a post made yesterday:

To my knowledge no one independent of the STAP groups themselves has gotten this technology to work and even Dr. Teru Wakayama, who is an author on the papers, can no longer get it to work.

And then you call STAP cells undoubtedly real...

http://www.ipscell.com/2014/03/stap-cell-nature-papers-cases-for-against-retraction/

2 ( +2 / -0 )

Ms. Obokata, may have had her reason, we haven't heard what she has to said. Even if she did all that... then how was possible that the research paper was published? (usually there supposed to be controls for that).

1 ( +5 / -4 )

@Beer4me

Yes true, the number of women being employed in science in Japan is abysmally low. Totally shameful that this country discriminates women like that.

But the issue is here goes maybe beyond that, because being man or woman doesn't really matter when it comes to behave with ethic in science. If she did falsified those images, she is in deep sh....

1 ( +5 / -4 )

Being benevolent, it could be a case of naive researchers using the wrong cell types and mistakenly reporting the use of other cell types:

A new article today in Japan suggests that while the STAP paper reported that it used a specific genetic strain of mice/cells called 129 that the cells given by Dr. Obokata to Dr. Wakayama turned out not to be 129 (update: instead it seems to be a mixture of two other strains, most likely B6 and one other, perhaps 129, referenced as “F1″ generation in the STAP paper). If correct, this would be an extremely serious blow to the STAP papers. See screenshot from video on that Japanese news site above.

Source below, scroll down to 'cell mixup/contamination': http://www.ipscell.com/tag/haruko-obokata/

1 ( +1 / -0 )

Another Asian fake stem cell breakthrough. It's sad as I was counting on Asian scientists, particularly Japanese, too lead the way, as the conservative Christian element in the US has blocked so much research. Then again, if PM Abe can state that the nuclear disaster in Fukushima is under control, to sway votes for the 2020 Olympics, why not fudge facts in other fields.

1 ( +6 / -5 )

I just wonder if STAP really exists or not? whaterver panel says about research papers were falsified or maniputated by her. That is the most important thing of all. But the panel did not say its existence. It seems that Riken try to make Obokata scapegoat about everything of STAP research as if Riken is not responsible of it

1 ( +3 / -2 )

She may be young and beautiful and people were drawn in by that but she's an animal tester. Absolutely no respect for her.

1 ( +5 / -4 )

When public finance is committed a full and thorough investigation through an Independent panel of experts is necessary to establish the truth and individual culpability.

Quote from: Knoepfler Lab Stem Cell Blog:

**Notably, the report addressed only six specific potential problematic issues when there are many more key areas of concern that it did not tackle. You know something’s bad when “only six “problems is a statement that makes sense as six problems is certainly a lot one would think.“The scientists said three other co-authors of the papers had not falsified the data but were still “gravely responsible” for failing to fully verify the research findings.The discrepancies in the data showed up as anomalous lines in an image of DNA fragments.”

Members of 'the panel' must be allowed the latitude to ask searching questions into every aspect of how this research paper was compiled, also to consider carefully whether there were commercial interests at play , or connections to any other vested interests, collaborators must be brought before the panel and sworn in.

A close examination of these facts could well find that Haruko Obokata is not the only scientist that 'cut corners'. It could be that other are forces at play here, if so, Haruko Obokata must not left swinging from the nearest lamp post alone.

1 ( +1 / -0 )

With all those involved at RIKEN and only Obokata is reponsible? Sounds like scapegoating and whitewashing. So RIKEN too the research at face value with no verification or checking? Something smells at RIKEN.

1 ( +1 / -0 )

She'll be lucky to get a tutoring position at Kumon.

She doesn't need a baito, she has the talent to become a Prime Minister.

With all those involved at RIKEN and only Obokata is reponsible?

I agree, she has not published the discovery on her twitter, they knew she was going to drop that bomb and they had the option of delaying the publication to check her work before. They have instead given her the green light so she could get her 15 minutes of worldwide fame and then public ridiculing.... I don't know what they were thinking, but they have used her.

1 ( +1 / -0 )

Obokata didn't plagiarize others' work in her 2011 PhD thesis I wish to point out that this is not entirely true, and she seems to have a history of plagiarism http://stapcell.blogspot.jp/ (see the posts on plagiarism). Now maybe her command of English is not that good and she simply copypasted passages fitting her research and changed some of the details. I'm not saying there is necessarily something sinister at play here or that the reason is something other than her being a bit naive, but am pointing out that she has indeed committed this particular no-no in science more than once.

Now, Frungy, as for your claim that STAP cells ought to exist. I am no medical researcher, but I indeed thought that this might be the case, that she accidentally stumbled on something cool, but as it was an accident, did not report all the steps properly. Well in physics (my background) we have/had Jan Hendrik Schon, who has often been called the biggest fraud in science, having had 8 papers from Science and 7 from Nature retracted, among others. What he did was to come up with a way to produce molecular transistors. Everyone knew that this should in principle be possible and the method he proposed seemed plausible. Indeed, there were theories predicting that if you made one, the measured curves should look like so and so. Turns out, he never built the damn things, he was just generating fake data to fit the theories. What he was hoping, I can only guess here, is that someone else would later fine tune his method and get it to work for real, while he gets the fame and the patents for being the "first" to do so. He also had some big name collaborators, which was the reason he really got to publish in Nature in the first place, but I won't go into this in more detail.

Anyway, the parallels with Obokata should be clear, and to me, this is why it is so worrying that Obokata's original procedure did not quite work out, especially so now that you pointed out that everyone knows about the potential effects of a hostile environment. And with the patents, there is an incentive other than want for fame, too.

1 ( +1 / -0 )

No claim that they "ought to exist", that's a given. They've replicated her experiment with a few minor tweaks and they DO exist.

I am not sure if you have vested interest in this or what, but even from an outsider's perspective, your claim is simply not true and the issue is still controversial. Science magazine (among others) reported on the liveblogging attempt by the group of Prof. Lee from the Chinese University of Hong Kong, http://www.researchgate.net/publication/259984904_Stimulus-triggered_fate_conversion_of_somatic_cells_into_pluripotency/reviews/103#532fa277d5a3f2f4188b4570.

This just ended with Lee commenting "Personally, I don’t think STAP cells exist and it will be a waste of manpower and research funding to carry on with this experiment any further".

Furthermore, http://www.ipscell.com/stap-new-data/ reports ten failures and one mixed result by independent groups.

Whatever the case may be, it is still worrying that you'd get published (or patented) for reporting that a method works when it is actually a variant that works and you never understood this. Accidents lead to scientific discoveries for they motivate research and analyses along some previously untread paths. What was done by Obokata et al. seems sloppy science at best, and a far cry from a rigorous analysis. This was my main point in the post above (which you did not comment on) and is why Obokata should be reprimanded regardless of whether STAP cells (by some modification in the procedure) exist or not.

1 ( +1 / -0 )

The one green result (i.e. promising, not successful) turned red upon further analysis as noted on the website. As for the mixed bag, I have not enough expertise to say whether it is even close to a partial success, and I'll be the first to admit this. I know therefore of no other groups that have managed to replicate the results. From what I remember one coauthor already called for retraction, indicating that he too has grave concerns.

You are resting your whole case on the assumption that someone else has replicated the results. I have seen no claims by anyone that they have done so, quite the contrary. Surely the aforementioned Prof. Lee would have stated that he believed in STAP cells if someone else had confirmed the results, and thus settled on merely criticizing the procedure of Obokata et al. instead of calling the hunt for the cells a waste of resources.

1 ( +2 / -1 )

This is the green result:

Yoshiyuki Seki on 2/13/14 wrote:

We used mouse embryonic fibroblast derived from Nanog-EGFP Tg. Now we are performing resuspended culture in B27 + LIF or serum + LIF for 4 days. In B27 + LIF medium, we can’t detect GFP-positive, while we can detect weak-GFP positive cells in serum + LIF. However, we observe many dead cells in GFP-positive clump. Therefore it might be difficult to expand clump of GFP positive cells in adherence culture.

And if you READ stuff and not blindly go with colors, you would see that:

Update from Yoshiyuki on 2/13/14. Suspension culture (100,000 cells/ml) pH5.7, incubation time 25 min. See images for various conditions below. Note from Paul–Yoshiyuki now reports this green signal is determined to be autofluorescence. Bummer.

There are no successful replications and your constant ad hominem attacks are not making your case any more compelling either.

1 ( +1 / -0 )

She has probably ruined her career now.

She'll be lucky to get a tutoring position at Kumon.

0 ( +4 / -4 )

Sorry, but this is just a witch-hunt. She is guilty of fraud, but Riken is guilty of allowing it to happen. If a newspaper writer prints libel he or she can be fired, but it's the editor and publisher that allowed it to go unchecked who should take the lion's share of the blame. They are trashing her because she is a woman who made former Nobel prize winner's science look bad, and a young woman at that. Hence, "the sole person responsible" garbage, while downplaying the role of others involved.

If the company now found it was falsified data, why could they not have before putting it out there and accepting credit for it? And what will be done now besides making the woman the scapegoat and the men thereafter trying to recreate the data for their own means?

0 ( +8 / -8 )

Agree with smithinjapan, young woman = easy scapegoat.

0 ( +3 / -3 )

The Japanese press and RIKEN were the ones who put forth the idea of her wearing a typical Japanese smock/apron as being somehow a great thing. "Look at our housewife looking woman making scientific discoveries." I think it was them trying to aggrandize their culture, which, in reality, subjugates women to a kind of B status in terms of scientific stature. She went along with the whole thing. If she was used as a scapegoat, it's part her fault for accepting the way it is in Japan. If she feels she was used as a scapegoat, she should call out those responsible. But, that's not done in Japan.

0 ( +2 / -2 )

this is so embarrassing. if true, this is the worst crime a scientist can commit. i imagine that many labs around the world are testing the same procedure, without being able to replicate it. this is almost surely pure fraud. in the end, the truth will come out, and if no one can replicate those seemingly fantastic results, it is punishment enough that the fraud be exposed, and no more tax money for fake, unethical researchers. if the results are proven true, then i hope they get a nobel prize.

0 ( +1 / -1 )

From January 2104, the remark from Chris Mason, professor of regenerative medicine at University College London, is poignant to say the least..

http://www.bbc.com/news/health-25917270

0 ( +0 / -0 )

How atypical of the Japanese. Must be the genes.

0 ( +0 / -0 )

@mikeylikesit I have been following Haruko Obokata since January, I am shocked, I'm not trying to defend the indefensible, medical science is not my forte either. I have always been taught a simple basic method , ask the questions, do background research, construct a hypothesis, test your hypothesis by carrying out extensive experimentation, analyze your data and draw a conclusion, communicate your results (verify with peer group), one last important is the independent variable, crucially a series of test to conclude your hypothesis, “trust, but verify”, this is very disappointing.

0 ( +0 / -0 )

This is all so crazy, Haruko Obokata must have known sooner or later making 'breakthroughs' that attract a global audience, any fakery will be found out, it the 'taking their word for it' that's astonishing, it there is no replication, it's science fiction, what were they all thinking?.

0 ( +0 / -0 )

I do hope she is vindicated in the end, because her finding was a fantastic achievement in every sense that it could be.

0 ( +2 / -2 )

For argument's sake, let’s just say Haruko Obokata’s STAP cells research breakthrough is real and sound as she claims, but her academic integrity is absolutely tarnished because her patterns of questionable behaviors and misconducts from deliberately fabricating the data to produce the findings in her STAP paper to plagiarizing others’ work in her 2011 Ph.D thesis.

So far, unfortunately, all of the data point to one direction- Haruko Obokata is guilty as charged.

0 ( +1 / -1 )

She is being criticized for how she graphed the data. They are not talking about the data being invalid. They are talking about the interpretation of the data.

-1 ( +3 / -4 )

She was a newbie when she did all of this. The optimist in me says she probably just didn't have the experience to put together a paper like this, and screwed up.

-1 ( +1 / -2 )

I'm a avid follower Haruko Obokata blog..

http://www.ipscell.com/tag/haruko-obokata/

Jury's out at the moment, Haruko Obokata posted yesterday.

-1 ( +1 / -2 )

@itsonlyrocknroll

The problem being that Chris Mason didn't actually check it himself. He assumed that other people did. But nobody else did either. They merely looked at Okuba's data and findings, gave it a thumbs up, and published it. Nobody actually tried to replicate anything from square one in an independent lab until after the paper was already published.

This is how virtually all science works. Peer review doesn't attempt to replicate results at all. Peer reviewers merely look at what the researchers give them and take their word for it. The process is highly susceptible to fraud.

Three-quarters of highly cited papers in top medical journals were found to be impossible to replicate when a pharmaceutical company attempted to do so. These are the most popular papers in the very best journals. The problem gets worse the further down the totem pole we go.

-1 ( +0 / -1 )

Bunch of old guys covering their asses.... And making Dr. Obokata the escape goat that is what it is.

Though there is no denying in the fact that there were errors on the report, I doubt that the study is a lie as they put it.

-1 ( +1 / -2 )

Anyone who's working in research will know that if colleagues were to thoroughly check your work prior to every paper submission everything would grind to a halt immediately. The responsibility is on the authors.

The only reason her work have been so closely examined is because of her extraordinary claims. Peer review is sometimes thorough, sometimes not, but we're talking about a top publication, Nature, here. Imagine the absolute tosh that gets through in lesser journals without making any waves. I wager if you had the time to investigate properly you'd find a fair percentage.

-1 ( +1 / -2 )

Glad to see that Obokata is fighting back. What about the Harvard prof? Was he so easily fooled so as to add his name to the list of authors?

-2 ( +3 / -5 )

I'm laughing at all the people here saying that this scientist should have known that she'd be caught and musing why the peer review process didn't flag her fraud before it was published. Are people still this naive about science?

Science is no longer about uncovering truth about the world around us. It's an industry, and people involved in Science, Inc., are interested in climbing up the ladder.

The problem is that universities mint more Ph.D.s than there are jobs. The only way to avoid adjunct hell is to have one's name at the top of a flashy article in a well-respected journal. And so people falsify their results. Others horribly misrepresent them. Given the choice between a non-result, which doesn't earn a scientist a career, and a falsified result, which earns university tenure and lucrative research grants, people often choose the latter.

Why isn't this caught? Because nobody is actually trying to catch fraud. When papers are sent to a journal, the peer reviewers look over the results to see if they are interesting and plausible. They don't actually try to replicate the results themselves. Doing so is expensive and would take far too much time away from the peer reviewers' own research. Nobody wants to give up their own careers to do the grunt work of checking other people's research. Peer review ends up being a cursory check at best.

As a result, many scientists are able to publish prominent articles early in their careers, enough to get themselves tenured university positions and some big research grants, and they never end up being caught. Those caught are the exceptions, not the rules. For most people, it pays to fudge their results.

How bad is the problem? As one example, a major pharmaceutical company a couple years ago took a sampling of the most prominent, heavily cited recent articles from several of the most influential medical journals. These papers are the lauded ones in the best journals, the cream of the cream of the crop. Of the 30 or so papers they checked, these researchers could not replicate the results in three-quarters of the papers.

Yep, three-quarters or more of the medical research out there is garbage science. Other scientific fields have been complaining in recent years about similar problems. Social sciences, where data is much easier to manipulate without getting caught, are reputed to have a much worse problem. The withdrawal rate of published papers at respected journals is rising to levels that have many scientists biting their fingernails. The flood of junk science is making it increasingly difficult for people to bring new technology and medicines to the market because they're wasting years chasing after what ultimately proves to be false data. Science, Inc., is in crisis mode.

Why people are at all surprised about a prominent paper getting called out for fraud is beyond me. This is just one of dozens of recent cases. It's the biggest to hit Japan, but it's hardly the most flagrant internationally.

Just remember--for any new scientific result you see published, odds are better than not that a second scientist attempting to replicate the research will not be able to do so. Science does not equal truth; anymore it doesn't even add up to the pursuit of it.

-2 ( +2 / -4 )

CarolingiumApr. 01, 2014 - 06:04PM JST My mind is blown by the continued white-knighting here. She was the paper's lead author (and she got the recognition for it), she recycled images from her thesis, no one was able to reproduce the same results. She is the main culprit here, and the investigation determined this, not some "scapegoat for being female"...

Okay, several points here:

Recycling images - One of the recycled images of a rat... it isn't actually a big deal, its just a generic image of a certain type of rat for goodness sakes. People are acting like it is some sort of high crime, it isn't. The other recycled image is that two fetuses appear to be nearly identical (some minor differences that make people suspect that this is the same fetus just images taken at different times). There's no hard proof here, but it is possible that she misfiled the images or mixed them up. It's bad, but it doesn't invalidate the study.

Replicability - Actually you're mistaken, some independent scientists have been able to replicate the results, although others have met with limited or no success. (http://www.ipscell.com/stap-new-data/). This IS a big deal. It appears that she left out some of the steps in her notes. It has been argued in her defense that this is not entirely uncommon, especially when the technique hasn't been patented yet and it is potentially a multi-trillion dollar discovery. Other researchers have done similar things in the past as a way of slowing down the competition. Of course it could equally just be sloppy note-taking or bad editing.

The key here is that STAP cells do seem to exist, that this young lady has made an important (but not earth-shattering) discovery, and that the media are engaged in a witch hunt without any clue about what is important and what is not important.

This isn't white-knighting, this is simply standing up for someone who's being dragged across the coals for no good reason. Sure she may have made some minor errors, but none of them are disqualification offenses, and it is far from clear if they were even errors. She is entirely correct in appealing the decision, but I hold out very little hope for her given the media pressure to crucify her... despite the fact that the reporters reporting on this issue seem to have absolutely no clue what they're talking about.

-4 ( +5 / -9 )

TakahiroDomingoApr. 01, 2014 - 09:28PM JST i imagine that many labs around the world are testing the same procedure, without being able to replicate it. this is almost surely pure fraud.

You're mistaken, some labs have been able to replicate it, validating Obokata's findings and proving her critics wrong. (Source: http://www.ipscell.com/stap-new-data/)

This means that while Obokata may have made some minor errors in writing up the report the actual key finding is correct and STAP cells do exist.

This is hardly a surprising finding to those of us who know something about this field. Here's a simple example of a similar process in humans. It has long been known that smokers cough a lot, but the "why" is something not known to most people. The answer is that the throat cells in the back of smokers' throats actually change type, changing from regular throat cells to basal cells (a type of stem cell), and then to a different type of cell that secretes more mucus to protect the throat from the smoke. This process, a cell reverting to a stem cell in the presence of a hostile environment, is very similar to what Obokata demonstrated, so this isn't the earth-shattering research that most claim. It is worth noting though that normally this type of change brings about a heightened risk of cancer, which is where I suspect that Obokata may have made a more serious error.

OnsenApr. 01, 2014 - 10:03PM JST The investigation was to find out about irregularities with the research report. But noone has mentioned the most important fact; are STAP cells for real? Everything else is relevant only if the STAP cell is a fake in itself.

STAP cells are undoubtedly real, the results have been replicated by independent scientists and there is a mass of evidence from similar processes (like the change in throat cells in smokers) that makes Obokata's research not only feasible but almost a certainty. We may quibble over some small points of protocol and procedure, but academics will spend weeks arguing heatedly over the placement of a comma, which is something that the media seem to interpret as a serious issue, but is really just academics being academics. Obokata undoubtedly made some errors, but the main discovery of her paper has been validated.

-5 ( +1 / -6 )

CarolingiumApr. 02, 2014 - 02:01AM JST Frungy, did you even read your source? You say some labs have been able to replicate it, but according to your own source, from a post made yesterday:

Yes, I read my source and the important part is that some labs HAVE been able to replicate it. They've all had to tweak the method from the published one though as some steps were omitted, either accidentally or deliberately.

To my knowledge no one independent of the STAP groups themselves has gotten this technology to work and even Dr. Teru Wakayama, who is an author on the papers, can no longer get it to work.

Dr. Wakayama probably didn't do any of the original lab work himself. If you knew how labs work (especially Japanese labs) then you would know that real experiments often mean staying away for days at a time to take readings every hour or even every 5 minutes during critical phases of experiments. This sucks bigtime, and older researchers tend to delegate running the actual experiment to junior staff members or even graduate students. As a result if Obokata omitted something from her notes then Dr. Wakayama wouldn't be able to replicate the experiment because he never did the experiment originally.

And then you call STAP cells undoubtedly real... http://www.ipscell.com/2014/03/stap-cell-nature-papers-cases-for-against-retraction/

I've read this blog post, and I also cited a REAL and well-documented example of how cells in smoker's throats undergo a reversion to stem cells in the presence of a mildly hostile environment. There are other examples, like mild acid therapies for reducing scarring in burn patients. It all depends what you mean by STAP cells - if you're referring to cells reverting to their basal state (stem cell state) in the presence of mildly hostile environments then there is no doubt that this is real, it is simply too well-documented to deny.

The bottom line though is that if even a single group can replicate the result and that group's modified method can be replicated elsewhere then Obokata is vindicated, and that appears to be the case from the crowd-sourced data coming in now.

I retain grave concerns about the rate of cancerous cells found in these STAP cells, but the existence of STAP cells is real, and Obokata's research, while flawed, has advanced the field considerably.

I don't know if you paid attention at school, but many great scientific discoveries, such as penicillin and vaccines, were the result of poor laboratory procedure and mistakes. These scientists are icons on science, despite their errors. Science is DIFFICULT, and mistakes happen. Obokata definitely made some mistakes, but pointing out a few small errors doesn't invalidate her results completely, doesn't diminish her contribution to science and doesn't mean she isn't a great scientist. Even the best typist will make a few errors when typing a dozen pages, but that doesn't mean the entire document is incorrect. Stop expecting unrealistic levels of perfection, because I can guarantee that if I observed you at work for a day I'd be able to find plenty of errors in your work.

-5 ( +1 / -6 )

EthanWilberApr. 02, 2014 - 08:24AM JST For argument's sake, let’s just say Haruko Obokata’s STAP cells research breakthrough is real and sound as she claims, but her academic integrity is absolutely tarnished because her patterns of questionable behaviors and misconducts from deliberately fabricating the data to produce the findings in her STAP paper to plagiarizing others’ work in her 2011 Ph.D thesis.

It really gets my goat when people who clearly have no clue feel it is okay to openly slander someone in a public forum while hiding behind internet anonymity:

Obokata didn't plagiarize others' work in her 2011 PhD thesis. What she did is commonly referred to as self-plagiarism. She needed a picture of a particular type of lab rat, and she found that she had one in her 2011 PhD thesis, so she re-used the image and forgot to reference it to the earlier source. It is, at worst, lazy. It in no way misrepresented the study or invalidated the study's results. It would be like someone calling you a plagiarist if you said something on Monday, and then again on Wednesday and someone said, "Hah! You plagiarist! You forgot to reference that you said that originally on Monday!".

As for the vague "questionable behaviours and misconducts" (sic.) you refer to, I'd like to see a full list please so I can debunk them one at a time.

Obokata is young and inexperienced, and she made some mistakes. These mistakes SHOULD have been picked up by her co-authors, which is why she would have sought more experienced co-authors to check her work. If anyone is at fault here it is the more experienced professors who were happy to put their name on the paper and claim another publication, but were too lazy to give her advice and spot her (minor) errors.

-5 ( +2 / -7 )

davestrousersApr. 02, 2014 - 12:44PM JST Anyone who's working in research will know that if colleagues were to thoroughly check your work prior to every paper submission everything would grind to a halt immediately. The responsibility is on the authors.

And that's the point here. Obokata is being crucified while her male co-authors (the people the responsibility is on) are getting away with a slap on the wrist.

No-one is reasonably expecting you to pass the paper to every other academic in the institute, but it is reasonable to expect your co-authors to do due-diligence and read and critique the paper, and point out inconsistencies and errors, and quite possibly go to the simple effort of replicating the experiment before publication.

-5 ( +0 / -5 )

lambdaApr. 03, 2014 - 01:05PM JST

Now, Frungy, as for your claim that STAP cells ought to exist

No claim that they "ought to exist", that's a given. They've replicated her experiment with a few minor tweaks and they DO exist.

-5 ( +0 / -5 )

Lambda - How can I explain this to you simply. There are two entries from someone called Yoshiyuki. They might be the same person, or they might not. You made two MASSIVE assumptions. Firstly that they're the same Yoshiyuki, and secondly that they're the same experiment. We don't know that they're the same person. Your theory is that Paul-Yoshiyuki is the same as Yoshiyuki is not proven. You're also just flat-out editing by putting in two dashes where there is one on the blog. Is it your normal practice to falsify the evidence if it doesn't fit your theory? Here is a cut and paste "Paul–Yoshiyuki" run your cursor along it. ONE dash, not TWO.

Even if it was (which I don't assume) they are two different experiments. One is "Suspension culture (100,000 cells/ml) pH5.7, incubation time 25 min", the other is "Now we are performing resuspended culture in B27 + LIF or serum + LIF for 4 days". 4 days is not 25 minutes. Honestly, that you cannot tell the difference between 25 min and 4 days means that you probably need more than just a diary to keep your appointments straight.

-7 ( +0 / -7 )

lambdaApr. 07, 2014 - 11:42PM JST Yoshiyuki and Yoshiyuki Seki are the same person (of Kwansei Gakuin University). In the comments (of the blog) callname "Yoshiyuki Seki" starts a thread reporting on his experiments. Other people ask for details and the person who answers is called simply "Yoshiyuki". It is clear that these are the one and same person, but that simply he could not be bothered to write his whole name again and again. Having established this, let's turn to the last piece, Paul–Yoshiyuki. Is the supposed Paul–Yoshiyuki (who does not exists according to Google) the same Yoshiyuki, i.e. Yoshiyuki Seki? He is, for it is indeed "Yoshiyuki" who in the comments section writes about the autofluorescence results, which exactly matches the interpretation that Paul wrote and update on Yoshiyuki's progress.

It is possible that the same Yoshiyuki posted twice, but it is also possible that it might be different people. If two people name Smith were posting we wouldn't even be having this discussion, but instead you're defending your assumption like it is a proven fact. It isn't.

I did not realize that you had not read a book before, or written large portions of text. Well, this must come as a surprise: there are different types of dashes. Nobody separates two names with two short dashes, but rather with one long one. Let me elaborate. Indeed it is true that the dash from Paul–Yoshiyuki "–" is singular, but so is "--" in any text editor for in fact the latter becomes the former (in Word/LibreOffice/LaTeX). Look up "em dash" on Google if you don't believe me (more specifically what Paul–Yoshiyuki contains is an "en dash", LaTeX for example has this as -- and the other long one, the em dash, as --- if I recall correctly. The usage of these dashes is of course different). I did not think using "--" in place of a clumsy copypaste or Alt-code could possibly be misunderstood by anyone.

Except that this isn't Word, Open Office, LaTeX, etc. This is a web brower, and as your post shows when you type -- you get drumroll --. If you're using a keyboard configured for Japanese and you shift the language your - will become ー. That also looks a lot like a long dash. It could equally just be a weird font. Do you know how you find out what is behind that code? Well, you copy it and then paste it without formatting... and it is revealed to be .... drumroll a dash. No mystery, no secret code, no hidden language of the ancients, no double-dash. It is a dash. Plain and simple. You're mistaken.

There might have been two experiments, I'll give you that much for I am not an expert. However, if you follow the comments section of the blog, you'll see that the four day results too, are due to autofluorescence, and thus a failure.

Do you even know what autofluorescence means in this context? It means the cells are stressed and dying... but not dead. This was anticipated in the model and is even included in the original paper (in the paper you can see autofluoresence). It doesn't necessarily mean a failure at all.

If you Google Prof. Lee's attempts, you'll see that a few days ago the press heralded him as having succeeded in replicating STAP cells. This turned out to be a bit too soon, and upon further analysis, Lee found that the experiments had in fact failed, and is now of the opinion that STAP cells do not exist. Surely had Yoshiyuki, Yoshiyuki Seki, or Paul-Yoshiyuki succeeded, these would too have been reported by the media (even though this particular researcher is not completely independent) and mentioned in conjunction with the news of Lee's success.

Prof Lee also made the point that there's a lot of pressure to publish quickly because the media's attention span is so short, and meaningful results won't be out for quite some time.

I don't think I can make this any clearer: You are plain wrong. Instead of even trying to provide evidence of a successful replication (which surely would reported by other sources than just one blog), you keep making more and more complicated and elaborate excuses.

I'm not making excuses, I'm pointing out the realities of research. If you can't tell reality from an excuse... well, you already have trouble distinguishing one dash from two, so why am I surprised. I'm confident that STAP cells do exist, and that there is ample proof of their existence in non-laboratory settings. I provided several real-world examples that you just ignored, probably because you seem to have very selective eyesight.

-8 ( +0 / -8 )

This proves that Women still can not get ahead in Japan, most likley she is being discredited because she is female. I really hope she didnt falsify her data.

-9 ( +5 / -13 )

lambdaApr. 04, 2014 - 12:41AM JST I am not sure if you have vested interest in this or what, but even from an outsider's perspective, your claim is simply not true and the issue is still controversial. Science magazine (among others) reported on the liveblogging attempt by the group of Prof. Lee from the Chinese University of Hong Kong, http://www.researchgate.net/publication/259984904Stimulus-triggeredfateconversionofsomaticcellsintopluripotency/reviews/103#532fa277d5a3f2f4188b4570.

This just ended with Lee commenting "Personally, I don’t think STAP cells exist and it will be a waste of manpower and research funding to carry on with this experiment any further".

I don't have a vested interest apart from an interest in the truth.

Other researchers have replicated the results. Prof Lee's opinion is frankly irrelevant. If even one other person can get the same results then Obokata is partially vindicated.

Furthermore, http://www.ipscell.com/stap-new-data/ reports ten failures and one mixed result by independent groups

Correction, one complete success (green), one mixed bag (orange), ten failures (red). The one green is all that matters. But you seem to have "not seen" the green because it doesn't suit your case.

Whatever the case may be, it is still worrying that you'd get published (or patented) for reporting that a method works when it is actually a variant that works and you never understood this.

If you knew something about the field you'd know that this wasn't uncommon. Let's take something a thousand times easier, baking a cake. 20 people can follow the same recipe and get slightly different cakes. Now a cake recipe is maybe 8 steps for a complex cake, using reasonably standardised equipment and ingredients. Try that with non-standard equipment and ingredients.

Frankly you're the one who lacks understanding here. Have you ever even been in a real laboratory? Or are you generalising from your high school experiments?

Accidents lead to scientific discoveries for they motivate research and analyses along some previously untread paths. What was done by Obokata et al. seems sloppy science at best, and a far cry from a rigorous analysis. This was my main point in the post above (which you did not comment on) and is why Obokata should be reprimanded regardless of whether STAP cells (by some modification in the procedure) exist or not.

I agree that Obokata should be reprimanded, ALONG WITH the other authors. She shouldn't have her career ruined though. The media is misrepresenting this entire thing as a serious breach of research ethics, it isn't, it is a mistake, and she really did discover something important... a point that you seem incapable of admitting even when reading a page showing that.

-9 ( +0 / -9 )

lambdaApr. 05, 2014 - 04:17PM JST The one green result (i.e. promising, not successful) turned red upon further analysis as noted on the website. As for the mixed bag, I have not enough expertise to say whether it is even close to a partial success, and I'll be the first to admit this.

Perhaps you should go and check again. The green result is still there. There were two green results, but one turned ORANGE. I don't mean to be insensitive, but are you by any chance color blind? You have my sympathies if you are, but there is software that will code stuff into different, more readable colors for you and avoid these sort of embarassing errors.

I know therefore of no other groups that have managed to replicate the results. From what I remember one coauthor already called for retraction, indicating that he too has grave concerns.

He's just covering his rear. The time to check the research is BEFORE publication, not after. What he's admitting is that he didn't check stuff before.

You are resting your whole case on the assumption that someone else has replicated the results. I have seen no claims by anyone that they have done so, quite the contrary. Surely the aforementioned Prof. Lee would have stated that he believed in STAP cells if someone else had confirmed the results, and thus settled on merely criticizing the procedure of Obokata et al. instead of calling the hunt for the cells a waste of resources.

My case is proven. The green result. I'd recommend you go back and look again.

-10 ( +0 / -10 )

lambdaApr. 08, 2014 - 02:06PM JST As for cell research, again, I claim no expertise. I have not been picking and choosing facts, in fact the tobacco stem cells you wrote about was why I initially joined in the discussion. If it is well known that external stress should create STAP-like cells, the ethical implications of Obokata et al.'s sloppy research are even graver. This I tried to point out with the Jan-Hendrik Schon example in my first post here. I am not saying that external stresses cannot convert cells to stem cells. I am saying that the evidence has it that the protocol of Obokata et al. is not the way to do this. This is to say that STAP cells, as defined by the protocol, do not exist, but STAP-like cells of course might.

And I conceded that Obokata et. al.'s unmodified protocal wouldn't work about from the very beginning. Steps have been left out, and that was clear from the very beginning, for example that the substrate was cut isn't included in the paper. If you had actually READ what I wrote instead of going full-attack-mode and insisting I was wrong before you even understood what I had written.

As for ethical considerations, there are more important and larger ethical considerations here than punishing Obokata's sloppy paper writing skills. If you can't appreciate the wrongness of scapegoating the junior (and only) woman on the team while allowing her senior (and all male) colleagues to walk off with nothing but a slapped wrist then you need a moral compass check. If you can't appreciate that Obokata's physical and mental health has suffered in the media-led witch hunt they you need to develop some compassion. Etc. etc. Finally, consider your own role in this ethically. You're sounding off about how Obokata is so bad when you obviously don't understand the field and haven't even read the paper. Where does that put you ethically?

-10 ( +0 / -10 )

Lambda - I did read the results. Where I differ from you is that I actually understood what was written there.

Firstly, you've incorrectly treated Yoshiyuki (clarified as Paul-Yoshiyuki in the comments) and Yoshiyuki Seki as the same person. They're not. The hint should have been the different names, but I can see how the similar names might be confusing... oh, wait, I can't see how anyone could do that.

So, I'd be a bit slower to criticise other people's reading skills if I were you.

Also, there's more and more research coming out showing that STAP cells are real.

I'll get to say, "I told you so" in about a year when the serious researchers manage to do their experiments, replicate them a couple of times to check they work, write up the research, submit it for review, edit it, re-submit, then wait for the journal's next publication date. By that point though the public's attention will be elsewhere and Obokata's career will be in tatters.

-12 ( +0 / -12 )

Login to leave a comment

Facebook users

Use your Facebook account to login or register with JapanToday. By doing so, you will also receive an email inviting you to receive our news alerts.

Facebook Connect

Login with your JapanToday account

User registration

Articles, Offers & Useful Resources

A mix of what's trending on our other sites