After a nearly year-long investigation into anonymous allegations of data and image falsification in numerous papers, a University of Tokyo committee today announced it had confirmed that one research group falsified images and graphs in five papers. The panel cleared five other research groups of wrongdoing.
The panel's judgment "is very severe," says Yoshinori Watanabe, leader of a team that studies chromosome dynamics at the university's Institute of Molecular and Cellular Biosciences. He previously acknowledged having departed from best practice in handling images, but maintains that the problems the panel cited in his group’s publications had no impact on the papers' conclusions.
The committee, headed by Hiroaki Aihara, a physicist and University of Tokyo vice president, recommended holding off on disciplinary measures pending a review of additional published papers.
The investigation was launched last September after an entity called “Ordinary_researchers” raised questions about images and data in 22 publications by six prominent research groups at what is arguably Japan's most prestigious university. In more than 100 pages of documents posted online, Ordinary_researchers detailed charges of image falsification and manipulation in the biomedical papers. In a 49-page report released at a press conference today, the investigative committee, which includes experts from outside institutions, found “minor inconsistencies” among images found in the papers of five research groups, but “no indications of falsification.” All of the exonerated groups are affiliated with the school of medicine.
But the panel noted "improper handling” of 16 images and graphs in five papers published by Watanabe’s group. The problems include image alteration and graphs with data apparently not derived from the reported experiments.
In June, Watanabe posted online detailed corrections to the figures that the committee later identified as problematic. Watanabe blamed the errors on such things as mix-ups in cell lines, merging data sets into a single graph, and improper processing of western blot assays.
Of the five papers the committee flagged as problematic, two appeared in Science, two in Nature, and one in EMBO Reports. They all cover processes affecting chromosomes during cell division. Watanabe says the publishers of EMBO Reports have accepted corrections to the paper. He is awaiting decisions from Science and Nature.
In an email to ScienceInsider, Ordinary_researchers praised investigators for their work on the Watanabe papers but said the examination of other publications “was not sufficient to dispel doubts” about questionable data. Ordinary_researchers also criticized the journals for a lax attitude toward suspected fraud that is “undermining the credibility of scientific research.” The group is hoping the incident “will deepen future discussions on the prevention of fraud.”
*Update, 2 August, 10:47 a.m.: The status of the Watanabe group's papers has been corrected and a statement from Ordinary_researchers added.