Open source genomics raise enormous challenges when protecting patient identity
Over the past decade the genomic revolution has taken off in spectacular style ushering in the potential for truly personalized medicine. At the same time, genome sharing websites like openSNP now provide a platform for uploading -personal genomics. Companies including — Ancestry.com and 23andMe are offering kinship and limited diagnostic services based on DNA testing. Thousands of people have already made their genomes publicly available.
But the full potential of genomic data is still being discovered. What it could be capable of revealing about us and how it may be used in the future is still a big unknown. This has kickstarted a conversation about genetic privacy and what it might mean to open source your genome.
An unmistakable you
Genome sequencing identifies the order of nucleotides in a genome, the chemical building blocks that make up an organism’s DNA. The human genome contains around three billion base pairs or roughly 100 gigabytes of data. Although your genome is 99% identical with everyone else, it is unique to you (and in rare cases your identical twin) making it possible to identify individuals by only a few changes.
Historically, genome sequencing was a complex and expensive process that was mostly confined to large research laboratories. The first human genome costing nearly $3 Billion in the early 2000s, but the cost of sequencing has plummeted, with whole genome sequencing now reaching the $1,000 per genome goal. Cheap computing and high throughput technologies are making genomic data increasingly easy to collect, store and process.
“Genomic information will increasingly be part of people’s lives, either in a medical setting or otherwise. As a society, we need to be prepared for what it means,” says Prof. Giulio Superti-Furga, Scientific Director of the CeMM Research Center for Molecular Medicine and head of the Austrian Genome Project. He was the first person to be sequenced as part of the project, in which the genomes of 20 volunteers were published to help to open a dialog about genome sequencing.
It’s important to understand what genomic data is able to say; and it’s even more important to understand what it cannot say.
What’s the big deal?
Voluntarily sharing personal information has become part and parcel of the 21st century, forming the basis of most social media networking. So why should we care?
Health information such as temperature and blood pressure, like your relationship status on Facebook, are of relatively short-term value. Genetic information is different. It does not change over your lifetime and therefore has value that lasts for decades. If you’re healthy, you may not care who is able to see your genetic information, but you might care if you find out that you are at a high-risk for cancer, or at least think you are.
“Let’s assume you sequenced yourself and find out that you have a mutation that gives you a 27% greater chance of getting a specific disease, which on average appears in 25% of the population. Your susceptibility is above average.” It’s not a certainty, “it’s a probability, and what the numbers mean is not trivial,” says Prof. Stefan Katzenbeisser, head of the Security Engineering group at the TU Darmstadt.
Thus it’s important to understand what genomic data is able to say; even more important to understand is what it cannot say.
“Take something like intelligence. It’s a complex trait that is likely to be the result of the interactions of thousands of genes. Any one mutation cannot tell you if someone will be intelligent or not. In most cases, we don’t have a simple answer,” says Prof. Superti-Furga.
Spreading the risk
When you submit your sequencing data to any organization, it is only with consent. It’s also in an organization’s best interest to keep your information secure. International standards on how to handle genome security are still lacking.
Moreover, consumer based DNA testing increases the likelihood that genomic data will be made available in less regulated environments (like personal posts on social media). Who is responsible if your genetic information is leaked?
“It’s a major concern. Health care professionals are already having trouble maintaining privacy of standard clinical data, something that there are already solutions to. So if they are already struggling with things that we think are easy, how can they deal with things, like genomic data, that we think are hard?” wonders Prof. Katzenbeisser.
Steps have been taken already in the right direction. Laws are being put in place to avoid discrimination based on genetic information. The Charter of the Fundamental Rights of the European Union specifically states that persons cannot be discriminated against due to genetic characteristics. Educational programs are springing up to improve public understanding of genomic data. But more work is needed.
“It will be a mixture of legal, educational and economic changes in society that will spread out the risk to individuals, if their genomic data is made public,” says Prof. Superti-Furga.
It will not be an easy task.