Questions surround alleged stem cell data falsification


Former University of Minnesota graduate student Morayma Reyes , who a University panel last week found had falsified stem cell research data published in 2001, said she refutes that claim and said errors that were made do not invalidate the paper’s conclusions.

The panel that found Reyes data was falsified did not examine the implications of their findings on the conclusions of the paper.

She and former director of the University’s Stem Cell Institute Catherine Verfaillie, who instructed Reyes, said its conclusions have been confirmed by themselves and others.

Verfaillie responded to questions by e-mail from Belgium.

The data in question were images of western blots , a method of detecting specific proteins. Reyes said the blots were used to show that certain cells from human bone marrow could be made into other types of cells, including bone, cartilage and muscle.

Reyes said the panel’s main allegation was that a band present in a western blot image in her notebook was not present in the published image.

She said this was due to the use of darker and lighter exposures of the image, and she denied that it constituted falsification.

Reyes said two other methods were used in addition to the western blots to verify the conclusions of the paper.

She also said she acknowledges making global changes — altering brightness and contrast in the images. And in 1999, when she was putting together the images used in the Blood paper, there was a lack of standards regarding image manipulation, she said.

Inappropriate image manipulation continues to be a problem, she said, and added that lack of education about the proper ways of dealing with it lies on the mentor as well as the student.

Reyes also said the problems should not be considered falsification because the definition of falsification published by the Public Health Services, and used by the committee said:

“Falsification is manipulating research materials, equipment, or processes, or changing or omitting data or results such that research is not accurately represented in the research record.”

Verfaillie said though guidelines weren’t in place when the figures were generated and put in the paper, she should have seen the alterations and prevented them from being published.

She said, however, that she disagrees with the University’s statement that she didn’t provide sufficient oversight and mentoring.

The panel examined whether the data falsification was committed knowingly or recklessly, said John Merritt, spokesman for the Office of the Vice President for Research .

Their definition of recklessness says that even if no harm was intended, the researcher showed indifference to risks that should have been obvious , Merritt said.

This isn’t the first time work done in the lab has been questioned.

In 2005, New Scientist reporters Peter Aldhous and Eugenie Samuel Reich began looking closely at stem cell research findings that hadn’t been replicated.

“There was a lot of talk about the reliability of research in the stem cell area,” Reich said.

They found data had been duplicated between two papers coming from Verfaillie’s lab, one of which was a 2002 Nature paper that showed a certain type of adult stem cell could change into a variety of other cells.

A University panel found that some data in the Nature paper may be flawed, and the authors published a correction. A Nature panel concluded that the paper’s conclusions remained valid despite flawed data.

In this case, however, the Blood paper will be retracted rather than corrected.

Cindy Dunbar, Blood editor , said the paper is being retracted because of the number of errors.

“This would basically be redoing the entire paper,” she said.

The decision to retract the paper was based on Blood’s own investigation, but they waited to retract it until the University had finished its investigation.

Blood magazine started image scanning to look for problems with visual data in 2006.

“Ninety-five percent of what we detect in that process is more educational than actual true fraudulent, designed to change the results type of misbehavior,” Dunbar said.

She said about a third of papers had some image problems, but researchers quickly learned what was and what was not acceptable.

Meri Firpo , a University stem cell researcher who was not part of Verfaillie’s lab, said the findings don’t call into question the entire body of work, but they do sadden the stem cell research community.

She said she doesn’t think the stem cell world is especially prone to this type of problem.

“It’s up to you to work with integrity, but you know, it’s possible there are people who won’t,” she said, and that’s the case in any field.