Adventures in Ethics and Science

Abi at nanopolitan nudged me to have a look at Nature‘s recent article on what has become of targets of recent scientific fraud investigations. He notes that, interspersed with a whole bunch of poster boys for how not to do science, there are at least a couple folks who were cleared of wrongdoing (or whose investigations are still ongoing) which seems, to put it mildly, not the nicest way for Nature to package their stories.

So, I’m going to repackage them slightly and add my own comments. (All direct quotations are from the Nature article.)

Scientist: Jon Subdo, researcher at Oslo’s Norwegian Radium Hospital, who studied how painkillers affect smokers’ risk of oral cancers.
Allegation: Subdo made up fictitious patients in his studies — hundreds of them.
Result of Investigation: GUILTY. “An independent commission of inquiry was set up to discern the details of the fraud; last June it reported that the bulk of Subdo’s 30-plus publications were invalid because of the fabrication and manipulation of data.” Also worth noting: Subdo himself admitted to the alleged fabrication.
Current Situation: Upon release of the commision’s findings, Subdo resigned from his position. He was stripped of his degrees, and may face criminal actions for his misdeeds, since his research was paid for with public funds. Subdo is not currently working in science.
Comments: The cases where the investigator admits wrongdoing are obviously the easiest to deal with: he admitted the lies, and lying is wrong. Lies that could endanger people’s health because they throw bad information into the pool of knowledge we think we have about cancer have the potential to do serious harm, too.

Scientist: Thereza Imanishi-Kari, formerly an assistant professor at MIT heading a lab in the cancer center, who collaborated with David Baltimore on a paper published in Cell on immunity in transgenic mice.
Allegation: Imanishi-Kari falsified data in the Cell paper.
Result of Investigation: CLEARED OF ALL CHARGES. The investigations took nearly a decade, and included the US Secret Service analyzing lab notebooks and other records for evidence of fraud.
Current Situation: “After doing no research for ten years, Imanishi-Kari reorganized a research programme at Tufts University in Medford, Massachusetts. Although pleased with the response of colleagues to her publications, she feels the case still overshadows her grant applications. ‘Being a woman in science is one kind of handicap; being wrong accused is another,’ she says.”
Comments: If you’re looking for the scientific equivalent of a beach-book, you might check out The Baltimore case: A Trial of Politics, Science, and Character by Daniel J. Kevles. Kevles’ piecing together of the story is largely sympathetic to Imanishi-Kari, but it suggests ways that issues of sloppy record keeping and data management, not to mention personality conflicts with post-docs and grad students, can create conditions where miscommunications can snowball into accusations of misconduct.

Scientist: Jan Hendrik Schon, physicist and researcher at Lucent Technologies’ Bell Labs.
Allegation: Schon falsified data for papers he published on electronic properties of new materials.
Result of Investigation: GUILTY. “His reputation collapsed following a Bell Labs report that found Schon guilty of falsifying data in at least 16 papers.”
Current Situation: Wherabouts unkown. The university that granted his Ph.D. (University of Konstanz) is currently investigating the research he did for that degree; presumably, their findings could result in his being stripped of his Ph.D.
Comments: This high-profile case spawned new research guidelines in physics. “Some of Schon’s co-authors were criticized for not spotting their colleague’s misconduct, prompting the American Physical Society to issue new rules stating that co-authors should be accountable for important data in papers they sign off.”

Scientist: Anders Pape Moller, an expert on behavioral ecology based in Paris.
Allegation: Moller used flawed data in a paper (since retracted) published in Oikos.
Result of Investigation: CLEARED OF INTENTIONAL FRAUD. Although the Danish Committees on Scientific Dishonesty concluded, in 2003, that Moller had committed “scientific dishonesty”, an independent “committee of wise men” appointed by the National Conference for Scientific Research in Paris came to a different conclusion. “In late 2004, the committee concluded that Moller was innocent: although it found that good lab practices hadn’t been followed, it could not prove intent to commit scientific fraud.”
Current Situation: Still doing research and publishing prolifically.
Comments: This looks like another example where “good lab practices” could have reduced the likelihood of miscommunications leading to allegations of misconduct.

Scientist:Woo Suk Hwang, South Korean superstar stem-cell researcher (and national hero) at Seoul National University.
Allegation: Hwang fabricated results in papers in which he claimed to have derived stem cells from cloned human embryos.
Result of Investigation: GUILTY. “[O]n 10 January 2006, Seoul National University announced after a four-week investigation that all of his claims were fabricated.”
Current Situation: Still embroilled in legal proceedings around chages of “fraud, embezzlement, and violation of the country’s biothetics law.” However, also doing animal cloning experiments in a new lab in Yongjin.
Comments: Snuppy turned out to be the real deal, but I’m not sure how the community of science is supposed to trust the word of someone caught in such spectacular dishonesty. The lure of well-funded collaborators in locales with permissive regulatory climates makes for some interesting decisions about trust.

Scientist: Shinichi Fujimura, Japanese amateur archaeologist who claimed to have discovered 700,000 stone artefacts.
Allegation: Fujimura buried the tools himself before excavating them.
Result of Investigation: GUILTY. An investigation by the Japanese Archaeological Association concluded in 2004 “that all of the 168 sites dug by Fujimura had been faked — forcing Japanese history textbooks to rewrite their descriptions of the Paleolithic period.” The institute of which Fujimura was deputy director was dissolved.
Current Situation: Fujimura is reported to have changed his name and to be “living quietly”. He is not currently practicing science.
Comments: It strikes me that archaeology has a different kind of professional hierarchy than, say, nuclear physics, so it’s not clear that Fujimura’s “amateur” status should necessarily have raised concerns about his “findings”. However, this case might raise reasonable questions about how the connectedness of a scientific community influences the availability of information to judge the credibility of members of that community.

Scientist: Luk Van Parijs, MIT biology professor conducting research on immunology and RNA interference.
Allegation: Van Parijs fabricated and falsified data in published papers, unpublished papers, and grant applications.
Result of Investigation: GUILTY. Van Parijs admitted to altering the data — and then was miffed when MIT publicly outed him as a cheater.
Current Situation: No one in the scientific community seems to have communicated with him since MIT cut him loose. Probably this means he’s not currently practicing science.
Comments: Nature turned up a detail that provides a neat example of intellectual honesty among collaborators:

Surprisingly, Van Parijs has co-authored a publication since the scandal, on a paper published in The Prostate in February 2006. The study’s first author — urologist Qiang Zhang of Northwestern University in Chicago [sic], Illinois — says thathe and his co-authors included Van Parijs as an author because he had provided helpful ideas when giving a seminar in Zhang’s department a few years earlier.

Kudos to Zhang et al. for acknowledging the source of important ideas! (I am hopeful that they also actively involved Van Parijs in authorial duties like reading and commenting on drafts of the manuscript before it was submitted, so that he did everything he needed to in order to properly qualify as an author.) It would be interesting to hear what the impact has been of having Van Parijs listed as an author of this paper.

Scientist: Bengu Sezen, graduate student in chemistry at Columbia University under professor Dalibor Sames.
Allegation: Sezen’s experimental results in carbon-hydrogen bond functionalization were said to be unreproducible.
Result of Investigation: UNKNOWN/PENDING. While Sames withdrew a number of papers which relied on Sezen’s results, it’s hard to know what investigations might be underway, since, as Nature notes, “Columbia University says that it is against its policy to ‘comment on the existence or non-existence of any internal investigation into allegations of research misconduct’.”
Current Situation: Sezen still argues that her experimental work was sound and that it has been reproduced by other researchers within the Sames lab.
Comments: Irreproducibility isn’t in quite the same ethical territory as fabrication and falsification — there are plenty of reasons it could happen even when all the players involved are doing their best to be honest. The fact that in this case, Sames apparently withdrew the papers without first contacting Sezen gives the appearance that he knew there was something more serious going on, but to those of us outside that lab and that relationship, it’s nothing like concrete evidence against Sezen.

Who didn’t make the list?

Dalibor Sames: As noted on ChemBark, Sames just retracted another paper on which Sezen wasn’t an author, which merits some discussion (which you’ll find in the ChemBark post). Minimally, there’s an issue worth exploring here about the responsibilities of collaborators (and graduate advisors) to ensure that reported experiments really are reproducible.

Eric Poehlman: For falsifying and fabricating data in the “preliminary studies” sections of numerous grant proposals submitted to federal agencies and departments, the former University of Vermont College of Medicine researcher will be doing jail time.

Gerald P. Schatten: The University of Pittsburgh biologist who sought out a collaboration with Hwang Woo-Suk was found, by a University of Pittsburgh investigation, to be guilty of “research misbehavior” for misrepresenting his involvement in a paper he “co-authored” with Hwang.

Victor Ninov: Dismissed from his position at Lawrence Berkeley National Labs for apparently fabricating creation and detection of element 118 in 1994.

These are just the omissions that strike me right away. What other guilty parties do you think belong on the list? Ought Nature to have taken steps to make it more obvious which researchers on the list had been cleared of wrongdoing? Are there other scientists you know of who have been cleared yet still labor under a cloud of mistrust?

Comments

  1. #1 Abi
    January 25, 2007

    One big name that’s missing from the list: David Baltimore! If Imanishi-Kari deserves to be in that list, so does he.

    While Poehlman and Schatten are not included in this list, they do make an appearance in other articles on scientific misconduct in the same issue of Nature.

  2. #2 Bob
    January 26, 2007

    There was a postdoc in the institute where I used to work, Meena Chandok, who had done some pretty interesting work on NO production in plants. After she left the lab, her PI was apparently unable to replicate her results, and he invited her back to try to replicate them. She declined, so her PI pulled the papers from the journals without her permission. She later paid for an independent company to replicate the results, and when the company did, she filed suit against her boss for tarnishing her career.

    There may have been some funny stuff going on with her PI and her visa in this process. I don’t really know what happened, so I don’t feel comfortable making any accusations. But I’m sure that her career suffered as a result.

  3. #3 Lab Lemming
    January 28, 2007

    What makes you think the Sames/Sezen case has any merit whatsoever? Seems to me that a jealous professor could try to cripple a talented student by making those claims, resting safe in the knowledge that nobody else in his lab has the talent to reproduce the experiments when the professor wants them to fail.

    Especially if he’s done this to more than one former underling.

  4. #4 HI
    January 30, 2007

    Amitav Hajra: A former graduate student of Francis Collins who fabricated data in papers he co-authored with Collins.

    Kazunari Taira: He published many results on RNA that could not be reproduced by others. When investigated, his group was not able to provide some of the raw data. Reading retraction of one of their papers is quite amusing. They identified the mRNA of a gene called ‘homolog of ES1′ (HES1) as a target of an microRNA. Yet, in their experiment, they showed that the microRNA diminished the protein level of ‘Hairly enhancer of split’ which is also abbreviated as HES1 but is actually encoded by a different gene.

  5. #5 Beating a Dead Horse
    February 1, 2007

    Lemming: the research should be described in the literature in such a manner that it may be reproduced. At least in synthetic organic chemistry, it wouldn’t take a tremendous amount of effort for members of the community to verify the experiments.

    If indeed Sames is a “jealous professor” trying to ruin the career of a talented student, he is doing it at great expense to his own reputation. Also, it would be odd to express jealousy in this way – when a professor trains an exceptional student it is a feather in his/her cap.

  6. #6 per
    February 8, 2007

    The fact that in this case, Sames apparently withdrew the papers without first contacting Sezen…
    so it is a fact, but only apparently ? What does that mean ?

    i am not sure if it is public record if sames
    1) contacted sezen
    2) tried to contact sezen, but was unable to
    3) did not have current contact information for sezen
    4) did not attempt to contact sezen

    I am not sure what difference it makes if you contact someone, and are still completely unable to reproduce their results.

    I would welcome prof free-rides view; if the good prof had papers with a PHD student, and that material turned out to be completely irreproducible once the phd left, despite valiant efforts, what would she recomment doing ?

    per

  7. #7 Bibhas De
    September 21, 2009

    Good discussion on science fraud, but you folks ain’t seen nothing yet. Here is the graddad of all them frauds:

    http://www.bibhasde.com/blackbody.html

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.