Nature talks to Sholto David about his process for flagging image manipulation and his tips for scientists under scrutiny.

The prestigious Dana-Farber Cancer Institute (DFCI) in Boston, Massachusetts, acknowledged this week that it would seek retractions for 6 papers and corrections for an additional 31 — some co-authored by DFCI chief executive Laurie Glimcher, chief operating officer William Hahn and several other prominent cancer researchers. The news came after scientific-image sleuth Sholto David posted his concerns about more than 50 manuscripts to a blog on 2 January.

In the papers, published in a range of journals including Blood, Cell and Nature Immunology, David found images from western blots — a common test for detecting proteins in biological samples — in which bands seemed to be spliced, stretched and copied and pasted. He also found images of mice duplicated across figures where they shouldn’t have been. (Nature’s news team is editorially independent of its publisher, Springer Nature, and of other Nature-branded journals.)

It was not the first time that some of these irregularities had been noted; some were flagged years ago on PubPeer, a website where researchers comment on and critique scientific papers. The student newspaper The Harvard Crimson reported on David’s findings on 12 January.

The DFCI, an affiliate of Harvard University, had already been investigating some of the papers in question before David’s blogpost was published, says the centre’s research-integrity officer, Barrett Rollins. “Correcting the scientific record is a common practice of institutions with strong research-integrity processes,” he adds. (Rollins is a co-author of three of the papers that David flagged and is not involved in investigations into them, says DFCI spokesperson Ellen Berlin.) The DFCI is declining requests for interviews with its researchers about the retractions.

David, based in Pontypridd, UK, spoke to Nature about how he uncovered the data irregularities at the DFCI and what scientists can do to prevent mix-ups in their own work...