Protecting genomic privacy through phone apps

Police are increasingly using genomic databases in their investigations. Concerningly, they often do so without a warrant requirement. A new paper co-written by Illinois faculty presents possible technological solutions, such as phone applications, that will ensure Fourth Amendment protections of consumer’s genomic data.

Over the past few years, many individuals have become interested in analyzing their data through genomics companies, such as 23andMe or Nebula Genomics. The process involves submitting a DNA sample to the company, which sequences the DNA and makes the resulting data available online through a password-protected website. Through these services, the consumers can learn about underlying health conditions or their family history. However, in all these cases, the companies control the data and the analysis tools, decreasing the user’s privacy.  

“According to the third-party doctrine of the Fourth Amendment, if you disclose information to a third party, they can turn over that information to the government,” said Jacob S. Sherkow (GSP), a professor of law. “Genomic data is problematic because it can be inferred through familial sequencing. By our estimates, the genome of essentially every single person in the U.S. of European ancestry can be inferred because of these sequencing companies.”

While many of us may want to help police investigations, allowing them to use our genetic data is problematic, according to Sherkow. “Disclosing genomic information necessarily implicates your relatives in a way that does not apply to other information such as your bank records,” Sherkow said. “In addition to making you more susceptible to warrantless searches, it also contributes to a dystopian situation where everybody can be uniquely identified regardless of consent.”  

To better protect the security and privacy of genomic technology, the researchers are developing a computational system. “The idea is that if you upload your genome to a third-party website, it will not be considered private. However, if I send you a program that analyzes your data on your computer, your privacy is protected,” said Carl Gunter (GSP leader), a professor of computer science. “It's like a home pregnancy test—you take it home and find out the answer. Nobody else needs to know unless you choose to tell them.”

Cell phone apps are a perfect example of how the computational tools can be sent to the data: they are capable of computing large amounts of data and are private. Although these apps work well for testing whether the users have an underlying health condition, they do not work as well for familial searches.

“Apps can help people answer most of the genomics-related questions they may have. On the other hand, finding relatives requires access to a large database, creating privacy complications,” Gunter said. “However, people have been working on this problem and it can be solved using secure hardware, encryption, and using trusted third parties.”

The GSP theme is currently working on this problem through different projects. They are conducting surveys to understand whether the public is open to using these apps to learn about their genetic information. The researchers are hopeful that since most of us have used apps, people will be open to getting information through an app of their choice. The theme is also focused on developing these apps. So far, they have used artificial intelligence to develop apps that range from simple tests—like testing for sprinter’s gene, which is associated with power athletes—to complex tests, such as predicting the progress of macular degeneration.

The work was done in collaboration with Natalie Ram, a professor of law at the University of Maryland Carey School of Law. The paper “Familial Searches, the Fourth Amendment, and Genomic Control” will be published in 96 Southern California Law Review. The study was funded by the NIH.