Genomic Privacy

Genetic information is unique among all laboratory data because it not only informs the current health of the specific person tested but may also be predictive of the future health of the individual and, to varying degrees, all biological relatives.

As DNA sequencing has become ubiquitous with decreasing cost, large repositories of genomic data have emerged from the domains of research, healthcare, law enforcement, international security, and recreational consumer interest (i.e., genealogy). Broadly shared genomic data are believed to be a key element for future discoveries in human disease. For example, the National Cancer Institute's Genomic Data Commons is designed to promote cancer research discoveries by providing free access to the genome data sets of 12000 cancer patients. However, in parallel with the promise of curing diseases, genomic data also have the potential for harm. Genomic data that are deidentified by standard healthcare practices (e.g., removal of name, date of birth) can be reidentified by methods that combine genomic software with publicly available demographic databases (e.g., phone book). Recent law enforcement cases (i.e., Bear Brook Murders, Golden State Killer) in the US have demonstrated the power of combining DNA profiles with genealogy databases.

We examine the current environment of genomic privacy and confidentiality in the US and describe current and future risks to genomic privacy. Reidentification and inference of genetic information of biological relatives will become more important as larger databases of clinical, criminal, and recreational genomic information are developed over the next decade.