Nearly 10 years ago, reports of fraud rocked the crystallographic community.  Scientists that had long prided themselves on validating and sharing their data were surprised to find instances of faked crystal structures published alongside genuine datasets. Fraudsters did not discriminate: cases were found in the protein and small molecule literature and even in specialist journals that had developed what they believed to be rigorous data validation workflows. The crystallographic community was quick to rally, with concerned members developing protocols and algorithms to help ferret out faked data.<br/><br/>Although the number of individual frauds might be considered relatively small, their reach can be broad. Crystallographers are known for sharing their structural data so that other researchers – outside of crystallography – can make use of their findings. Fabricated structures can thus easily become embedded in the research methodologies of others. What recourse do scientists have when fraudulent data are deposited alongside of real data? What steps can data repositories take to ensure the quality and authenticity of the data entrusted to them? Might mandating publication of raw data help reduce the chances of fraudulent structures slipping through? This talk will discuss the efforts that the crystallographic community are undertaking to keep fake data from polluting the well.

Crystallographic crime: Detection and prevention of fake data

256th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.