However, a report setting out the finding also reveals that Imperial experienced considerable difficulties in investigating its suspicions due to the reluctance of the industrial collaborator on Dr Ahluwalia’s studentship to grant access to his lab books.
The investigation was announced in August 2011, after Dr Ahluwalia’s co-authors agreed to retract a 2003 Journal of Neurochemistry paper, of which he was first author, following the failure of his former supervisor, Istvan Nagy, to replicate its findings.
In 2010 a paper written while Dr Ahluwalia was a postdoctoral researcher at University College London was retracted by his former boss, Anthony Segal, after a UCL committee found that he had manipulated his results and had probably interfered with colleagues’ experiments to cover his tracks.
It subsequently emerged that Dr Ahluwalia had been dismissed from the University of Cambridge’s PhD programme in 1997 after his supervisor suspected him of faking results.
He then did a PhD at Imperial between 1999 and 2002, funded by a Medical Research Council “Case” studentship, in collaboration with the pharmaceutical company Novartis.
In 2009, while the UCL investigation was ongoing, Professor Segal informed Imperial of his suspicions about the 2003 paper. No misconduct was found during a subsequent investigation, but the paper was corrected in 2010 after “an arithmetical error” was identified.
Following its 2011 retraction, a six-person panel investigation panel – which included Imperial’s pro-rector for education, dean of students and student union president – was formed to check Dr Ahluwalia’s PhD work for fraud. None was found.
According to the panel’s report, Dr Nagy said that although he had recently succeeded in broadly verifying Dr Ahluwalia’s findings, he recognised, with the benefit of hindsight, that his student’s methodology had been “careless” and “flawed”.
The report suggests the problems were exacerbated by a “separation” between Novartis and Imperial.
“Had Dr Ahluwalia been more closely supervised during these series of experiments at Novartis, the methodological and arithmetic errors could have been spotted at an earlier stage. This is something that should be addressed for future Imperial-industry collaborations,” it says.
“Collaborative PhDs should always be truly collaborative, supported by strong communication between partners.”
The report also notes that “protracted negotiation” was required before Imperial was granted supervised access to the relevant lab books, which were held by the firm. It recommends that in order to avoid similar difficulties in the future, agreements with industrial research partners must specify that Imperial will retain access to all data and lab books.
It adds that the institution should offer staff and students help to improve the long-term archiving of data.