Faking it

Throughout most of 2007, Tony Segal felt like he was banging his head against a brick wall.

The prominent immunologist’s reputation was on the line amid a bad-tempered dispute with rival groups over the validity of a high-profile paper he had published three years earlier. Segal was trying to repeat the key experiments, which had originally been carried out by one of his former postdoctoral researchers, but kept getting contradictory results.

“The experiments worked at some times but not others,” Segal explains. “I just couldn’t understand what was going on. Experiments aren’t easy to do and it is not like turning a switch, but sometimes we got unequivocal results that were absolutely 100 per cent clear [and other times not].

“That level of inconsistency is unusual but not completely unknown. But it was bizarre and confusing, and drove me crackers.”

Five years earlier, Segal, Charles Dent professor of medicine at University College London, had recruited Jatinder Ahluwalia as a postdoc on the strength of a PhD from Imperial College London and a CV that boasted a first-class degree in biochemistry from the same institution.

Back then, Segal was riding high, having recently published a Nature paper that, in his estimation, transformed his field by overturning the accepted wisdom that white blood cells use free radicals to kill bacteria. In fact, he had found, that task was performed by the digestive enzymes activated by incoming potassium. Ahluwalia’s task was to uncover the precise channel the potassium ions were using: before long, he presented Segal with a “perfect” set of results.

After confirmation by an electrophysiologist in a neighbouring laboratory within UCL, the results were written up and published in Nature in 2004.

However, the paper soon began to attract what Segal views as excessive and disrespectful levels of scorn at academic conferences, and his vigorous initial defence of his corner did nothing to dilute the acrimony, some of which lingers to this day.

Segal rejects accusations from some in his field that he persisted too long in his defensive stance. He says scientific accuracy has always been his paramount concern and points out that he set to work on repeating Ahluwalia’s experiments in 2006, as soon as his detractors produced a paper, published in the Journal of General Physiology, that directly contradicted his reported findings.

“I was shocked by the accusations because I have published 220 papers and been in science for nearly 40 years and haven’t made any obvious mistakes. So for people not to be able to reproduce one’s work is really quite a serious thing,” he says.

More than two years of inconsistent results later, Segal hired another electrophysiologist, Philippe Behe, who not only failed to reproduce Ahluwalia’s results but also noticed that the postdoctoral researcher had relabelled computer files containing his results.

Further investigation revealed that some of the reagents used in colleagues’ experiments had been contaminated, while electrical swipe-card records indicated that uncharacteristic late-night visits to the lab by Ahluwalia coincided with several dates on which experiments had suddenly, mysteriously, started working.

“When I heard about all this it was one of the most devastating and shocking experiences of my life. It was beyond anything I could imagine: even a bereavement,” Segal says.

This discovery set in train a series of events that, in Segal’s view, exemplify the difficulties that currently exist in dealing properly with misconduct and the urgent need for the sector to deal more robustly with the issue.

The formal investigation that UCL launched in 2009 concluded late the following year that it was beyond reasonable doubt that Ahluwalia had misrepresented his experiments by deliberately altering the numbering of computer files containing results. It also concluded that it was likely, on the balance of probabilities, that he had used unreported chemicals to alter his results and had deliberately contaminated chemicals used in colleagues’ experiments “so as to falsify the results of those experiments in order to conceal the falsification by him of the results of his own experiments”.

UCL’s three-member investigating panel consisted of an external chairman (a lay member of UCL’s council), a senior scientist from Segal’s faculty and an external scientific expert in the same field. The paper was retracted. Yet the inclusion of an external expert was insufficient to assuage Nature’s fears about a libel challenge from Ahluwalia if it granted Segal’s request - with which it had initially expressed sympathy - to add the panel’s findings to the paper’s supplementary materials.

Meanwhile, Ahluwalia had obtained a lectureship at the University of East London - without a reference from Segal. He refused to sign the retraction. The panel’s report was published on UCL’s website, but Nature declined to include a link to it in the retraction notice.

A spokeswoman for Nature points out that the journal had published details of a contact at the university for anyone who wanted to access the full investigation.

“At Nature we are proactive when it is necessary to clean up the literature,” she says.

The spokeswoman adds that the journal is a strong supporter of the campaign to reform England’s libel law - which helped to ensure that libel reform was included in the Queen’s Speech this year - and has recently “vigorously” defended a libel suit relating to the Egyptian researcher Mohamed El Naschie. (El Naschie sued the journal over a news article published in 2008 that raised questions about peer review of papers he had written and published in his own journal. The case was dismissed last month.)

But Segal is adamant that retraction notices must offer readers clear reasons, partly to make obvious to scientists which parts of papers are faulty and, more importantly, to distinguish between research fraud and honest error.

“The silent withdrawal of a paper has much less of an effect upon the individuals and institutions concerned, but it also protects flawed scientists from exposure, allowing the perpetuation of their deceit,” he says.

Segal is also critical of what he regards as Imperial’s “irresponsible” unwillingness to carry out a thorough investigation into Ahluwalia’s PhD work. Segal first wrote to Mervyn Maze, then head of the Division of Surgery, Oncology, Reproductive Biology and Anaesthetics, towards the end of 2008 to warn him that UCL was about to begin its formal investigation and to make known his strong suspicions about a Journal of Neurochemistry paper published in 2003 that reported results Ahluwalia had obtained during his PhD work.

However, Imperial’s internal review, carried out by two senior academics from the division, concluded that there was no reason to doubt the validity of Ahluwalia’s work - although it noted the possibility of a minor typographical error in the paper and highlighted several inaccuracies in the CV he had shown Segal. Chief among these was Ahluwalia’s claim to have obtained his undergraduate degree from Imperial: in fact, it was from UEL.

Segal urged Imperial to look again and Maze told Times Higher Education that he had asked Ahluwalia’s thesis supervisor, Istvan Nagy, to repeat Ahluwalia’s experiments. However, this was not followed up after Maze left Imperial for a position at the University of California, San Francisco, which he took up in September 2009.

Segal’s lobbying of successive Imperial rectors eventually led, in June 2010, to a correction to the Journal of Neurochemistry paper stating that the published units of measurement were three orders of magnitude too high. He then noticed that the units were corrected to a value below accurately measurable amounts and drew this to Imperial’s attention.

But it was only after the revelation early in 2011 that, prior to beginning his Imperial PhD, Ahluwalia had been dismissed from the University of Cambridge’s doctoral programme in 1998 for suspected research misconduct that Imperial finally set about trying to replicate his results. The failure of these attempts led to the retraction last August of the Journal of Neurochemistry paper, and Imperial is now considering whether to rescind Ahluwalia’s PhD.

Details of Ahluwalia’s misadventures in Cambridge only came to light because his supervisor there, Martin Brand, happened to learn about the UCL investigation and contacted Segal - who then passed on the information to Imperial, UEL and the widely read Retraction Watch website.

“Internet blogging sites [such as] Retraction Watch are playing an increasingly important role in policing science,” Segal says. “In addition to drawing attention to retractions and making them more public, they put pressure upon the institutions involved to behave properly.

“In this case, it is highly probable that they had a major influence on the actions of UEL and Imperial in finally addressing the problems presented by Ahluwalia.”

UEL announced last July that it had parted company with Ahluwalia following an internal investigation. His current professional whereabouts are unknown.

Concern about research misconduct, particularly in the biomedical field, has been rising in recent years. A high-level meeting to address the issue, convened earlier this year by the British Medical Journal and the Committee on Publication Ethics, resulted in a series of recommendations that fed into the UK funding bodies’ Concordat to Support Research Integrity, published last month.

The principles make clear that the prime responsibility for investigating and punishing research misconduct lies with employers. The concordat requires them to establish “robust, transparent and fair processes” for doing so, and recommends that they nominate a senior staff member to oversee research integrity.

Funders such as the Higher Education Funding Council for England are contemplating formally requiring the institutions they finance to comply with the concordat.

Segal applauds this move and hopes that funders will impose financial penalties on universities that fail to deal “properly” with scientific misconduct. But he believes that institution-led investigations “will always be open to the accusation of bias”, particularly given the threat of legal action from those under investigation.

For this reason, he endorses the view of the Commons Science and Technology Committee - rejected by the government - that a quasi-judicial watchdog body along the lines of the US Office of Research Integrity should be established in this country with powers to oversee institutional investigations of alleged misconduct.

A spokesman for Imperial acknowledges that its investigation into the Ahluwalia case had been “unacceptably slow” and says the college has taken steps to ensure that any future allegations of scientific misconduct are dealt with more swiftly.

“We now have a more established approach for investigating concerns about fraud or scientific misconduct and have reviewed our policy on investigating scientific misconduct,” he says.

He adds that Imperial is also considering “where it would be appropriate to improve communication to research students and their supervisors about what constitutes research misconduct, and (about) how these cases are highlighted and handled”.

The spokesman says that the investigation into Ahluwalia’s PhD was hampered by the university’s inability to contact him or to obtain his lab books, which are held by the industrial partner that co-funded his studentship, the drug firm Novartis. A spokesman for the firm says that it is now “complying with Imperial’s request” to see the books.

Although there is no suggestion that Novartis acted improperly, Maze is concerned by the ability of industrial partners to keep doctoral students’ data “under lock and key”. But he insists he responded to Segal’s concerns in the way he regarded at the time as appropriate.

In retrospect, he agrees that the UCL investigation and the inaccuracies in Ahluwalia’s CV should have prompted more vigorous scrutiny of his findings, but adds: “There is a certain amount of trust implicit…and with graduate students we expect that the admission process will have evaluated them.”

Segal agrees that reliance on a level of trust is unavoidable in science and he insists that he could not have done more to detect Ahluwalia’s deceit earlier: “I only see raw data: I can’t see what goes into a test tube.”

But he believes that once caught, an example should be made of miscreants and that they could, and ideally should, be prosecuted. However, he admits that the cost and onerous nature of pursuing a prosecution makes it unpalatable.

“By the time this stage is reached, the academics and university have already wasted so much time and resource in identifying the problem and dealing with it at a local level that there is little appetite for further investment of both [in criminal proceedings],” he says.

Segal feels that a lot of time has been wasted on the affair during an important period of his career. “It has been a major diversion: do I really want several more years of witness statements and legal stuff?”

At the very least, he believes that wrongdoers should be barred for life from science. In an ideal world, he thinks, misconduct would result automatically in the revocation of doctorates, which would mean that universities looking to employ a scientist could simply check with the awarding institution whether their PhD was still valid.

“However, because the degree-conferring organisation is generally not the same as that in which misconduct has been perpetrated, there is little incentive, other than institutional solidarity and a belief in true academic principles, for them to get involved in the issue, with the administrative and legal resources that would require,” he notes.

Therefore, Segal believes that a central register of bona fide scientists needs to be created, plus a corresponding blacklist (on to which Ahluwalia could have been entered upon his expulsion from Cambridge). He suggests that within Europe, such lists could be maintained by the European Commission’s Directorate-General for Research and Innovation.

But above all, Segal thinks his case illustrates the need for major changes to a scientific culture that regards time spent on investigating misconduct as research time wasted, and which views the retraction of a paper as a major black mark against a lab head, regardless of the reasons for it.

“To withdraw a paper should not be a disgrace if you are not responsible for the misconduct,” he says. “If someone comes into your house and steals your family heirlooms, you shouldn’t be blamed. But many scientists hear ‘retraction’ and immediately think ‘misconduct’. Individuals that voluntarily subject themselves to these negative associations should be respected for their honesty and integrity, but the stigma may keep some researchers from coming forward to admit honest errors.”

He says the loss of a reputation as a “reliable and competent scientist” can lead to doubts about the veracity of other published work, which, in turn, has knock-on effects on a scientist’s ability to publish in major journals and to win further grants.

“There is also the suspicion of the possibility of complicity, if only through a lack of diligent supervision, or encouragement of the subservient scientist by the unquestioning acceptance of results that, unusually, are consistently positive,” he adds - although he stresses that Ahluwalia’s results also had negative ones interposed.

Segal says his lab has recently succeeded in identifying the potassium channel Ahluwalia was hired to find and is writing up the results. But he admits that his own reputation has suffered following his retraction, and notes that many senior colleagues advised him against looking too closely at Ahluwalia’s results.

“It was very difficult to catch this guy and very few would have spent two years trying to find out what was going on,” he says. “Most people wouldn’t have had an investigation: they would have waited for the paper to be covered by the sands of time and the mountain of new publications. At most, they would have withdrawn the paper quietly.”

But he says his “obsession” with exposing the truth made such options unacceptable.

“My scientific work is exemplary and I am not going to tolerate a blot on the landscape,” he says. “Setting the record straight has to be every self-respecting scientist’s first priority because the less accurate the record is, the less effective the scientific process can be. Individuals and institutions must act responsibly to eradicate the corrosive influence of [research] fraud.”

Secrets and lies: the timeline of a misconduct case

1996: Jatinder Ahluwalia begins PhD at the University of Cambridge

1998: Ahluwalia dismissed from Cambridge for suspected research misconduct

1999: Ahluwalia begins a PhD at Imperial College London

2002: Ahluwalia obtains doctorate from Imperial

2002: Tony Segal, Charles Dent professor of medicine at University College London, recruits Ahluwalia as a postdoc

2003: A paper is published in the Journal of Neurochemistry reporting results Ahluwalia obtained during his PhD

2004: Segal and Ahluwalia publish a paper in Nature setting out the latter’s ‘findings’ on the potassium channel in Segal’s lab

2006: Segal sets to work on repeating Ahluwalia’s potassium-channel experiments after a paper is published in the Journal of General Physiology contradicting his findings

2007: Ahluwalia leaves Segal’s lab

2008: Segal writes to Imperial to explain that UCL is about to begin a formal investigation into Ahluwalia’s research at UCL. He also reports strong suspicions about Ahluwalia’s 2003 paper

November 2010: UCL publishes the results of its formal investigation. The Nature paper is retracted

February 2011: Ahluwalia’s 1998 dismissal from Cambridge reported by Retraction Watch

July 2011: UEL announces that it has parted company with Ahluwalia following an internal investigation

August 2011: The 2003 Journal of Neurochemistry paper is retracted. Imperial announces it is considering revoking Ahluwalia’s PhD.