Published online 4 September 2008 | Nature | doi:10.1038/news.2008.1083


Researchers criticize genetic data restrictions

Fears over privacy breaches are premature and will impede research, experts say.

DNACould an individual's health details be extracted from pooled genetic data?Getty

As fears over privacy prompt genetic databases in the United States and Britain to close public access to some of their data, scientists working in the field are complaining that the moves are premature and will impede research.

The US National Institutes of Health (NIH), the Broad Institute in Cambridge, Massachusetts, and the Wellcome Trust in London all decided to restrict access to data from genome-wide association (GWA) studies — which contain collections of thousands of people's DNA — after research suggested that it is possible to identify an individual from their genetic fingerprint even when their DNA is mixed together with that of many other people.

The research, published in the August issue of PLoS Genetics, describes a technique that can pick out an individual's genetic fingerprint from a mixture of many DNAs, even if the individual's DNA is only 0.1% of the total1. The technique could help police investigators to identify possible suspects when a crime scene contains DNA from dozens of people. At present it is difficult for forensic investigators to detect an individual if their DNA comprises less than 10% of a mixture.

But data-holding institutions like the NIH are concerned that this technique could pose a threat to the confidentiality and privacy of patients involved in studies to understand the genetic basis of disease. GWA studies list tens of thousands of genetic variants called single nucleotide polymorphisms (SNPs) from groups of patients.

Fears premature

“I understand their concerns ... but it could hamper data sharing, which has facilitated so many discoveries.”

David Craig
Translational Genomics Research Institute, Phoenix, Arizona

To counter any potential privacy breaches, the institutions have said that researchers will now have to apply for access to the genetic data, which had previously been freely available online. Alan Schafer, head of molecular and physiological sciences at the Wellcome Trust, anticipates that it will take "less than a week" for researchers to apply for access.

But David Craig of the Translational Genomics Research Institute in Phoenix, Arizona, and lead author of the PLoS study, told Nature that the restrictions were "a very strong reaction and a bit premature". To identify an individual from a large DNA mixture using his technique, their genome would already need to be known, which limits the risk of privacy violations or misuse. "There would have to be another leap to identify someone," he adds. "I understand their concerns, they are just being safe. But it could hamper data sharing, which has facilitated so many discoveries."

David Balding, a statistical geneticist at Imperial College London, UK, agrees. He says that in practice, it will likely take much longer than a week to receive approval to access the databases.

"I was shocked that the NIH has removed GWA summary data from public access and changed its policy because of the recent PLoS Genetics paper," he says. "The actions of NIH will impede science. Although scientists will still be able to access data eventually, in practice getting all the relevant bureaucratic approvals can take months and this can be a huge deterrent to researchers and a restriction on scientific progress."

Schafer agrees that there is no real danger of abuse, adding that the Wellcome Trust had not decided if the restrictions would be permanent. He disagreed that the restrictions would impede research. "You do need the high-density genome of an individual to be able to compare it to the pool, so there is no real danger of abuse. But because we have given assurance of absolute anonymity to those involved in the studies, we don't want to step back from that," he says. "If it remains restrictive we will have to see. It is part of our learning on how to protect information." 

  • References

    1. Homer, N.. et al. PLoS Genet., 4, e1000167 (2008). doi:10.1371/journal.pgen.1000167 | PubMed |
Commenting is now closed.