Geneticist Haruko Obokata, who drew global attention with an apparently revolutionary paper on STAP cells, will retract the report, the government-backed Riken institute announced on Wednesday.

The retraction would be a significant move, as both the paper and an auxiliary study were carried by one of the world’s most influential and prestigious journals, Nature, which itself would bear scrutiny for having published them.

An inquiry conducted by Riken earlier identified two problems with images used in the main paper and recommended that Obokata retract it.

Obokata previously had said she opposed doing so. Speaking at a news conference in April, she said retracting the study would be tantamount to declaring its findings wrong.

On Tuesday, however, Obakata signed a document agreeing to the paper’s retraction and submitted it to her co-authors, Riken said. She had already agreed to withdraw the secondary paper.

Whether Nature will withdraw the article remains unclear, as one of the co-authors, Charles Vacanti of Harvard University and Boston’s Brigham and Women’s Hospital has not agreed. Typically, all authors must agree to a withdrawal.

Also Tuesday, Riken researcher Takaho Endo reported that genetic data on STAP cells Obokata said she cultivated were likely obtained by combining two different kinds of cells. Endo analyzed data that Obokata and her co-authors released via the Internet, and reported the findings to Riken.

Riken does not intend to investigate the matter further.

The cells in question involve stem cells that are created by cultivating STAP cells, or stimulus-triggered acquisition of pluripotency cells, which the researchers said have the ability to develop into any type of tissue. Obokata and her co-authors said in their paper that the stem cells were created from an “F1” mouse and were capable of developing into a placenta.

However, Endo’s analysis of the genetic data found that the stem cells cultivated possessed mixed features of embryonic stem cells and trophoblast stem cells. The latter are capable of developing into a placenta, while the former are known to lack the ability to develop into certain parts of a placenta.

Also, the type of mice created in the experiment were “B6” and “CD1,” not “F1” as claimed by the researchers, Endo said.

“It is quite unlikely that this happens as a result of an accident or mistake,” Endo said, adding that the two kinds of cells could have been mixed intentionally.

Hitoshi Niwa, a Riken project leader who co-authored the paper, said at a news conference in April that embryonic stem cells and trophoblast stem cells could not be mixed uniformly.