Tuesday, September 04, 2007

Investigating scientific misconduct: Divya Gandhi on the Kundu case

MSM takes a closer look at curious case of Gopal Kundu, with Divya Gandhi's column in The Hindu examining several different issues surrounding this case. Her column uses this case to explore some other questions as well, but I want to stick to the Kundu case here.

Here's the key point:

In an unusually public mark of dissent, the generally discreet Indian scientific community has voiced its concern over what could be the latest case of science gone astray. [...]

The unprecedented attention this case has received, however, has to do with more than the alleged malpractice. The disquiet in science circles comes instead from what is seen as the failure of institutional mechanisms in dealing rigorously and impartially with such cases — and also from the frustration at the absence of a central authority to bring closure to the growing incidence of misconduct. [bold emphasis added]

Given Rahul's analysis of the available evidence (using online documents at the Society for Scientific Values -- SSV), we are led to a pretty inescapable conclusion: identical figures -- in seven sets! -- have been passed off as arising from different experiments. Thus, there has been some closure in the minds of people who have taken a look Rahul's analysis.

However, it is also true that this case has not seen an 'official' closure. And that's because of two conflicting conclusions: one from an official body and the other from an 'almost official' body (the journal that published Kundu's papers). Specifically,

  • The official committee, headed by Prof. G. Padmanaban, an ex-Director of IISc, exonerated Gopal Kundu and his coworkers of any wrongdoing. This happened in August 2006.
  • However, in February 2007, the Journal of Biological Chemistry (JBC) withdrew one of Kundu's papers. On being queried by SSV (and also a Science reporter), the journal has admitted that the withdrawal of the paper followed the adverse conclusions of its own committee that went into the allegations against Kundu. More damaging was its admission that it was aware of the Padmanaban Committee's exoneration of Kundu.

JBC's withdrawal of the paper is clearly a slap in the face of the official investigative team, and it is this slap that's keeping this case alive (even though it has also been the subject of some sharp criticism in a Current Science editorial). Unless the Padmanaban Committee -- or NCCS, Kundu's employer, or the Department of Biotechnology, NCCS's official boss that the Padmanaban Committee also reports to -- gets the paper reinstated, the needle of suspicion will keep pointing at Kundu (and indirectly, at the investigative team as well).

Gandhi does a good job of highlighting the disconnect between the Padmanaban Committee's findings and what many others (including JBC) believe:

... Baffled by this verdict [of the Padmanaban Committee] and anxious for a resolution, many scientists have felt compelled to examine the papers for themselves. The two sets of photo-strips of the little protein bands have since been scrutinised keenly — as many as seven times — formally and voluntarily, by committees, individuals, and institutions, and have been discussed in the public domain.

The spotlight has shifted decisively from the authors of the contested paper ... to the Padmanabhan committee. Formed to bring an authoritative closure to the case, the committee has instead been embarrassingly contradicted by the JBC, which withdrew the paper, and finds itself increasingly isolated within the Indian scientific community, with its motives and investigation methodology brought under the scanner.

Here, then, is the current status of the Kundu case:

And so the NCCS saga continues. The Department of Biotechnology recently called for another report from the Padmanabhan committee, a vindication of sorts for SSV’s position. But the committee’s new report of 120 pages upholds its previous findings. The DBT will hand this one over to a set of three scientists for yet another review. The verdict of this eighth (and, with some luck, final) inquiry will signal more than the fate of this specific case. [bold emphasis added]

Clearly, there's something deeply wrong if a simple question -- Are two figures (bearing different labels) identical? -- needs an eighth inquiry (and more than a year) for a closure.

* * *

One final note. Gandhi quotes Satyajit Mayor (of the National Centre for Biological Science, Bangalore):

... [Others] in the scientific community are concerned that such cases could dent India’s credibility in the international sphere. Satyajit Mayor, a biologist at the National Centre for Biological Sciences, observes: “We need laws to protect the country from what happened to Korean science after the stem cell debacle. Now nothing they say is taken seriously, even though important research is being done.”

This quote (and it is possible that Mayor is being misquoted or quoted out of context) gives one the impression that Korean science was let down by their laws (or lack thereof). This impression cannot be more wrong. In the Korean case, when the allegations (which were doing the rounds in mailing lists and online forums for quite a while before they) finally broke through to the news media, the official investigation did its job impeccably well. And its conclusions and credibility have not been questioned! Acting on the conclusions of that inquiry, the papers with fabricated data were promptly withdrawn by the university. If at all any lesson needs to be learnt from the Korean episode, it is on how to act swiftly, impartially, dispassionately, and fairly.