Science 'express concern' over publication of 'flawed' longevity gene paper

Serious doubts have been raised over the validity of a study on the genetic basis of longevity published in the journal Science.

The article, published in the 1 July 2010 issue of Science, claimed to have identified a number of common genetic variants that were predictive of whether people would live to over 100. Now the editors of Science have published an 'expression of concern' in the 12 November issue, citing technical concerns from other researchers which call into question the validity of the published findings.

In the original paper by Dr Thomas Perls and Professor Paola Sebastiani and others from Boston University, the researchers conducted a genome-wide association study (GWAS) that compared genomic sequences from 1,055 individuals of 'exceptional longevity' (those who had lived to be over 100) with 1,267 control individuals.

They found that 'exceptional longevity' could be predicted with 77 percent accuracy in an independent set of centenarians, using a genetic model the researchers had developed that was based on 150 SNPs (single nucleotide polymorphisms) - minor alterations in the sequence of genetic material that occur within populations.

However, since the findings were published it has been suggested that the genotyping platforms used to produce the data in the study, and lack of quality-control measures in the data analysis, may have led to several false-positives being reported. One particular platform and 'gene chip' (610-quad array) used in the study, for example, was found to produce experimental errors in other unpublished studies.

Since these concerns were raised the researchers have re-analysed the data using improved quality-control measures, and obtained new results for the same set of individuals using more established platforms, as well as using an alternative method to validate the results from the original genotyping. The journal has agreed to review the study again, taking into account the new data and will then decide whether to support the original findings.

Dr Pearls a co-author of the original study told the Independent: '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 re-performing the analyses with this clean data set'.

The journal Science will then decide whether to accept the original paper, republish a corrected version, or retract it entirely, by the end of the year.