Chapter 7 - Game theory for privacy-preserving sharing of genomic data

Abstract

The coalescence of efficient genome sequencing technologies, low-cost computing, and academic and commercial incentives has led to the establishment of genomic data repositories on millions of individuals across the world. Making data available from such repositories may enable important discoveries, but high-profile demonstrations have illustrated that the privacy of individuals could be infringed upon if appropriate safeguards are not set in place. A wide array of deterrents have been designed to protect genomic data privacy, some of which are technical (e.g., obfuscating data), while others are more social (e.g., legal contracts). However, it has not been obvious which protections need to be adopted. Game-theoretic approaches, which model rational interactions among the parties involved, have the ability to optimally balance utility and risks in genomic data-sharing scenarios. In this chapter, we review recent advancements in genomic data privacy that rely on game-theoretic approaches, while characterizing their core contributions, as well as their limitations.