A highly acclaimed study by Haruko Obokata, a rising star in Japanese scientific circles, who said she discovered a new and simple method to engineer pluripotent stem cells, is facing allegations that some of the images documenting her finding show irregularities.
Obokata, 30, who leads a research unit at the Riken Center for Developmental Biology in Kobe, said in late January that she and her colleagues discovered a mechanism called stimulus-triggered acquisition of pluripotency (STAP), whereby strong external stresses, such as an acid bath, could reprogram mouse somatic cells into pluripotent cells, which can develop into body organs and tissue. Her research results were published in a pair of companion papers in Nature, the prestigious British scientific journal.
The Riken national research institute began looking into the case after an outside expert informed it Feb. 13 that some of the images looked questionable. The Nature magazine also opened investigations into the matter.
But Riken maintains the veracity of the research results remains well-grounded.
"We don't plan to have Obokata comment on the matter on her own," a Riken official said.
Online blogs specializing in uncovering irregularities in scientific research, such as the ones over studies on the hypertension drug Diovan that developed into a major scandal, called Obokata's Nature articles into question on Feb. 13.
The blog sites said two photos in one of the papers, which purportedly show mouse placentas under different conditions, look extremely similar.
Teruhiko Wakayama, a co-author of the Nature papers, admitted a "simple mistake" resulted in a mix-up of photo images. He admitted the two photos showed the same animal.
"We used two shots of an identical mouse taken from different angles," Wakayama, a professor of developmental engineering with the University of Yamanashi, told The Asahi Shimbun on Feb. 18. "The second photo ended up irrelevant to the text of our final manuscript, which we had revised over and over. But we forgot to delete it."
According to Wakayama, the team of scientists turned around and flipped over several mouse embryos, engineered from STAP cells, in several hundred photographs. Obokata, confused, ended up using two photos showing the same embryo, he said.
Her extremely heavy workload, whereby she had to conduct additional tests by herself while concurrently preparing figures, was another factor behind the mix-up, Wakayama added.
The blogs also said an image in Obokata's other Nature article contains unnatural linear features that may indicate the photo was doctored.
Bloggers also raised suspicions over an earlier paper that Obokata published in 2011 from the time she studied at Harvard University. It said several images in the publication could be duplications of an identical image.
Obokata, who obtained her Ph.D. from Waseda University in Tokyo, cited her 2011 article in her doctoral dissertation, which also uses the images in question.
The images represent the results of tests to discover which genes were active. Several images in the paper are strikingly similar, although they are supposed to show tests on different genes.
"It certainly appears to have been an honest mistake," Charles Vacanti, an anesthesiology professor and Obokata's supervisor at Harvard Medical School, told Nature magazine. He asked the scientific journal last week to withdraw the images from her 2011 article.
Waseda University opened its own investigation on Feb. 18, but denied there would be far-reaching consequences.
"We don't believe the main points of the Ph.D. dissertation would be affected even if the images in question were to be withdrawn," a Waseda University representative said.
(Michiko Nakamura and Ryoma Komiyama contributed to this article.)
- « Prev
- Next »