October 7, 2008

U. of Minnesota Panel Says Stem-Cell Scientist Faked Data

A panel convened by the University of Minnesota-Twin Cities has determined that a high-profile laboratory there falsified evidence in experiments on adult stem cells, according to a statement issued by the university.

The panel’s investigators concluded that parts of four images had been falsified in a paper published in the journal Blood in 2001. The paper reported that stem cells isolated from adult bone marrow could develop into different types of tissue, but researchers subsequently had great difficulty reproducing the Minnesota lab’s results.

The university said that the investigation had focused on two individuals: Catherine M. Verfaillie, a professor and former director of the university’s Stem Cell Institute, and Morayma Reyes, who was then a student in the combined M.D./Ph.D. program at Minnesota. The panel determined that Dr. Verfaillie was not guilty of the image manipulations. By extension, that would leave Dr. Reyes as the person responsible, but the university did not release the panel’s findings related to her because it considers information about students private and protected by state and federal laws.

Dr. Verfaillie has retained a position at Minnesota but has moved to the Catholic University of Leuven, in Belgium. Dr. Reyes is an assistant professor of pathology at the University of Washington, in Seattle.

The panel criticized Dr. Verfaillie’s laboratory for “poor scientific method and inadequate training and oversight for this research.” It contacted Blood and asked the journal to retract the paper. The investigators also found discrepancies with images in a second paper from Dr. Verfaillie’s laboratory, published in the Journal of Clinical Investigation in 2002. Those problems did not rise to the level of academic misconduct, the university said.

Tim Mulcahy, vice president for research at Minnesota, told The Chronicle that the case underscored the need for all universities and faculty members to be vigilant about misconduct. “The message here is that everyone needs to fulfill their responsibility to the public and to science,” he said. The university did not have plans to alter its policies related to training or oversight, he said.

The New Scientist, a magazine, first detected problems with images from Dr. Verfaillie’s laboratory last year. In an article today, the magazine quotes Dr. Reyes as saying that the problems with the images in Blood were “honest errors.” The University of Washington may investigate Dr. Reyes, according to the New Scientist. —Richard Monastersky

Comments

1. Seems like a lot of scientists out there faking it these days. What happens when they get caught? Do they ever work in science again? Or do they end up working the Tilt-a-Whirl at the church fair?

   — Dee Dee Pontiac    Oct 7, 06:42 PM    #

2. We tried unsuccessfully to replicate the Minnesota laboratory’s results for the better part of a year. On several occasions, we requested information about specific troublesome protocol procedures from the laboratory, but we never received a response. Ultimately, we abandoned the project with the conclusion that either we were incompetent or the Minnesota results were simply not verifiable.

   — Dean O. Smith    Oct 8, 01:39 AM    #

3. What I don’t understand (one thing anyway) is why such a high proportion of the big-time cases of falsified data, unreplicable experiments, disgraced scientists, etc., involve biological research. Yes, there’s cold fusion. But outside of a very few such cases in physics/physical sciences, so many of these cases involve biological research. Why is that?

   — Tim2    Oct 8, 09:14 AM    #

4. Concluding that “a lot” of scientists fake data seems unwarranted. Yes, there are regular reports of faked data, plagiarism, etc., but there are thousands of scientists and other scholars out there. The few don’t represent the many.

   — Mark    Oct 8, 09:16 AM    #

5. Tim2 – I think there are a few reasons. First, biological research is very hot right now, so there is tons of competition. Second, many (most?) biological projects take a long time to get going. For an asst prof, that may mean that if the big one doesn’t work there isn’t time to do anything else before tenure time. Third, I see less collaboration in the biological sciences than the physical sciences. Or maybe I should state that as more, and more brutal, competition. It’s easier to fake something if nobody else is seeing your data in real time. Fourth, much of the data relies on easily manipulated images, rather than pure quantitative data. Finally, and this also plays into #4, many of the techniques are so difficult and there are so many unknown variables that it’s easy for another lab to decide that it’s their incompetence preventing replication of the experiment (see Dean O. Smith above), rather than misconduct. I find chemical and physical experimental setups to be more straight forward than biological ones. Of course, I’m a chemist, so that may have something to do with it.

   — Chem Prof    Oct 8, 10:16 AM    #

6. The irony is that the data that was manipulated was not the primary results of the study. In the case of the Blood article, the results HAVE been replicated by several other groups.

   It’s an odd scenario where sloppiness leads to this kind of misconduct.

   As someone who specializes in technical innovation, I can attest that oftentimes when a lab tries to replicate or use our methods and initially fails, the issue is lack of training. We are trying our best to provide better access to our methods, but sometimes the issue is not the wand, but the magician.

   That does not explain why a lab would not respond to queries, however.

   — Former UM faculty member    Oct 8, 10:24 AM    #

7. It certainly isn’t confined to the natural sciences. I know of several cases in the social sciences and humanities in which extensive word-for-word plagiarism in published works has not proven any impediment to the malefactor’s academic career.

   — Gustave    Oct 9, 04:33 PM    #

Previous: $125-Million Pledged to Harvard Program in Biologically Inspired Engineering
Next: New Report Outlines Barriers to Study Abroad at Community Colleges