Research bodies remove personal genetic data from web

Scientists in the US have developed a new way of identifying an individual's DNA from a mixed sample. An individual can be picked out in a sample that contains up to 1000 people's DNA and where it makes up just 0.1 per cent of the total. The new method will assist forensic scientists, but has also raised concerns over the privacy of volunteers who have previously taken part in other genetic studies.

The research was led by Dr David W Craig, from the Translational Genomics Research Institute in Phoenix, Arizona, and was published in the August issue of the journal PloS Genetics. The team developed a statisitical method to identify an individual's DNA in a mixed sample by analysing up to 50,000 genetic variations, called SNPs (single nucleotide polymorphism). 'This is a potentially revolutionary advance in the field of forensics,' said Dr Craig. At present, it is difficult for forensic investigators to detect an individual's DNA if it is less than 10 per cent of a mixture, or if many other people's DNA is present.

SNPs are commonly used in genome-wide association studies, to identify genes that play a role in a particular disease. Summary statistics from such studies were published on the Internet, and list thousands of SNPs from groups of patients. There are concerns that with this new statistical method, and an individual's SNP profile, it may be possible to find out if that person took part in a study (i.e. they are among the summary statistics), and if they were in the group diagnosed as having the disease.

As a precautionary measure, the National Institutes of Health (NIH), the Broad Institute in Cambridge, Massachusetts (both in the US), and the Wellcome Trust in the UK have removed summary statistics from genome-wide association studies from the web. There is a strong demand for summary statistics from genome-wide association studies as they are a useful way for scientists to quickly find out if a genetic association they find has been borne out in other studies. Researchers will now have to apply directly to the NIH and adhere to strict confidentiality rules to see the data.

Some argue that the threat to the privacy of research volunteers would have been small, as no information about individuals could be discovered unless someone already had access to their SNP profile. Several researchers have criticised the move, saying that it could impede science, and hold back the sharing of information that facilitates many discoveries. Others disagree, arguing that it is vital to ensure absolute anonymity of study participants and the withdrawal of data from the public domain will not be restrictive.