A study that delved into the genetics of ageing has come under the scanner after it claimed that it may be possible to ascertain whether people are likely to live to 100.
Other scientists have voiced technical doubts about the way the researchers came to their main conclusion that they could predict whether someone is likely to become a centenarian with an accuracy of 77 per cent.
The journal Science, which published the study in July, has now issued an unprecedented "editorial expression of concern" about the scientific paper by Boston's University's Paola Sebastiani, and Thomas Perls, associate professor of medicine.
Other researchers involved in similar studies where entire genomes are scanned by sophisticated "gene chips" have pointed out that one of the chips used in the study, called the 610-quad array, can under certain circumstances produce experimental artefacts that could result in skewed data.
"Science and the authors are taking these concerns seriously. Since learning of these potential problems, Sebastiani et al have been performing a thorough quality-control analysis of the raw data, as well as generating new data to compare the genotype calls from the 610-quad array and the other platforms within the same individuals.
These steps aim to eliminate biases between platforms," said Science.
In the study published in July, the scientists said they had identified 150 single-nucleotide polymorphisms, or mutations, that could be used to predict whether a person is likely to live to 100 with an accuracy of 77 per cent.
The researchers also reported that they had discovered that 45 per cent of the oldest centenarians, those over the age of 110, had a genetic signature with the highest proportion of longevity-associated mutations.
These figures are now likely to be revised.
"We've taken the letter of concern very seriously and done a lot of work to produce a clean, validated set of data, and we've been in the process of reperforming the analyses with this clean data set," the Independent quoted Perls as saying.