Recently we had a dust up about supposed leftist tendencies to see crises everywhere. Here, perhaps, we see scientists finding what they want to find -- sort of.
Ian Murray suggested we might be interested in this article. <http://www.nytimes.com/2007/02/28/science/28stem.html> February 28, 2007 Panel Finds Flawed Data in a Major Stem Cell Report By NICHOLAS WADE An inquiry panel has found what it called "significantly flawed" data in a major stem cell paper published in Nature in 2002. The article, which claimed stem cells isolated from an adult could change into all the major tissue types of the body, was seized on by opponents of abortion as showing that embryonic stem cell research was unnecessary since adult stem cells could provide all the predicted benefits. The lead author of the article, Catherine Verfaillie, said yesterday that she had sent a letter to Nature stating that the flawed data should not be relied on but that they did not affect the article's conclusions. She said the journal was resubmitting the article to the original referees for them to make their own assessment. Problems with the article were first reported this month by New Scientist. Two writers for that magazine, Peter Aldhous and Eugenie Samuel Reich, noticed last year that a set of graphs in the Nature article was the same as that in an article by Dr. Verfaillie published in a different journal but ascribed to different mice. Timothy Mulcahy, vice president for research at the University of Minnesota, where Dr. Verfaillie had led the stem cell institute, said an inquiry panel was set up after Dr. Verfaillie asked the university to look into the issue of the duplicated graphs, which she attributed that to a mix-up when preparing two papers for publication at the same time. The panel confirmed Dr. Verfaillie's account, Dr. Mulcahy said, and concluded that there was no issue of scientific misconduct. But it found another problem besides the duplication, involving inconsistencies in the data describing the proteins on the cells' surface. The university then sent both articles to stem cell experts to assess whether the conclusions would still hold if the inconsistent data were to be eliminated. Three experts said that conclusions would still stand, one that they would be weakened. Because of this lack of complete agreement, the university believed that the scientific community should be informed of the situation, through Dr. Verfaillie's letter to the journal editors, and make its own evaluation, Dr. Mulcahy said. Dr. Verfaillie is now at the Catholic University of Leuven in Belgium but still holds a part-time appointment at Minnesota. "I have no idea how this could have happened or why it happened," she said, referring to the inconsistencies described by the panel. Other laboratories have had difficulty repeating Dr. Verfaillie's results, but recently some have succeeded, Dr. Verfaillie said, although they often gave the cells different names. Recently she wrote an article with Irving Weissman of Stanford University, a stem cell expert renowned for his rigorous methods. But that article confirmed only that her cells could make blood, not that they would transform into other tissue types. "I think he is still not totally convinced they can make all the other cell types," Dr. Verfaillie said. Dr. Weissman did not respond to e-mail messages asking for his assessment. This is the second time that a major stem cell paper has come under criticism, the previous episode involving the collapse last year of Hwang Woo-suk's claims to have generated embryonic stem cell lines from adults. Although Dr. Hwang's case involved fraud and belongs in a different category, the problems with the two papers underline the difficulty of working with stem cells and the immaturity of the field, despite the large sums of money being invested in it. "Working with stem cells is pretty tricky," Dr. Mulcahy said. -- Michael Perelman Economics Department California State University michael at ecst.csuchico.edu Chico, CA 95929 530-898-5321 fax 530-898-5901 www.michaelperelman.wordpress.com
