Analysis: Science cheats

* A Turkish institution set up a committee of investigation to look into allegations of misconduct brought to its attention by a British journal. Some of the published paper's co-authors had contacted the editor to complain that they did not believe the claims of the first author. The committee established the extent of the misconduct and the paper was retracted.

* An author submitted a paper that was, in essence, copied from another researcher's work, published in the journal several years earlier. The original author raised the alarm. The journal editors tried to force a direct apology from the author, but only after a lengthy correspondence and the threat of legal action was a "veiled admission of fault" sent to the journal as a letter to the editor.

* In August 2000, three authors retracted a molecular biology paper in the journal Science after failing to replicate their results. The first author subsequently admitted altering data to validate the central conclusions of the work. The scientists informed the journal and institution and retracted a related paper. By this time, the original results had been evaluated by a distinguished scientist and probably prompted research by other scientists.

* A scientist was found to have plagiarised the work of another researcher. It was decided to suspend the person from submitting further papers to the journal during the tenure of the board of editors. The editor did not know how the miscreant's institution dealt with the incident.

* A decade ago, one professor at a leading British university claimed to have worked at the weekend on research mentioned in publications. Members of his laboratory were suspicious because no one had seen him in the building at the times he had claimed. Although the work merely confirmed a consensus view in an obscure field, fraud was uncovered. The scientist resigned, but the incident was hushed up.