THE CHALLENGES
Five reasons why it’s incredibly hard to link people’s genetic information
Creating an Internet of DNA is far trickier than inventing the World Wide Web ever was — the inventor of HTML, for one, never had to worry about protecting patient privacy or ensuring responsible use of the data being shared. Some of the potholes on the genomic information superhighway: PRIVACY, PRIVACY, PRIVACY Expanding access to data increases the chance of a breakthrough — and for privacy to be breached. This is a central paradox for genomic data sharing and the stakes are high; your genetic code contains information not just about your health, disease risk and heritage, but that of your relatives, too. And there’s no telling how DNA might someday be used by everyone from insurance providers to law enforcement.
But perhaps the unavoidable reality is some privacy risk will always exist — and people will just have to accept that when they share their genetic info for the greater good. GOVERNING THE INTERNET OF DNA If a patient in Canada shares their genomic information and a researcher in Japan misuses it, what are the consequences? What oversight is needed and who is responsible? What kind of consent forms should be signed by people who volunteer their DNA — and do their relatives need to sign too?
These are among the many thorny problems being deliberated by groups like the Global Alliance for Genomics and Health. “This is so complex,” said McGill University’s Bartha Knoppers, chair of the Global Alliance’s regulatory and ethics working group. “Every time you say, ‘Oh good, we’ve band-aided this part of the problem’ . . . something else comes up.” HELPING EVERY COUNTRY PLUG IN During the first month of Sierra Leone’s Ebola outbreak, lab workers collected blood samples from 78 patients — but it was only after they were shipped to Massachusetts that the viral genomes could be analyzed.
In many countries, researchers lack the local capacity to sequence and analyze genetic data, let alone share it online. To ensure a truly global “Internet of DNA,” every nation needs the tools to contribute. “This requires a democratization of sequencing technology,” said Nathan Yozwiak with the Broad Institute. “Sequencing should occur as close to the sample collection point as possible; we really want to empower local capacity.” IN SEARCH OF STORAGE SPACE “Big data” is big by definition — but that in itself is a problem. “We’ve come to the point where it’s too big to store,” said Tom Hudson of the Ontario Institute for Cancer Research. Just one of his projects, sequencing 89 different cancer types, has already generated about a petabyte of data — this would take six months to download even on the fastest university networks.
The Internet of DNA “is not going to happen with today’s technology,” he said. “We need to come up with cheaper and cheaper ways of storing the data and exchanging the data. We’re facing huge bottlenecks.” CREATING A CULTURAL SHIFT The greatest barrier to more data sharing? Research culture. When funds are scarce — and scientific careers depend on being the first to publish — there are few incentives to share. “Professors don’t want to share because they’ll get scooped,” said U of T’s Aled Edwards, founder of the Structural Genomics Consortium. “And the institution’s been told to protect their intellectual property. That means excluding people from using it.”
Cultural barriers prevent not just data from being shared but everything from patient samples to chemical compounds. The solution? Those who hold the most power in science — universities, academic journals, funders — need to change the reward systems, Hudson says.