Privacy-Preserving Whole Genome Sequence Processing through Proxy-Aided ORAM

Abstract

Widespread use and low prices of genomic sequencing bring us into the area of personalized medicine and biostatistics of large cohorts. As the processed genomic data is highly sensitive, Privacy-Enhancing Technologies for genomic data need to be developed. In this work, we present a novel and flexible mechanism for the private processing of whole genomic sequences which is flexible enough to support any query. The basic underlying idea is to store DNA in several small encrypted blocks, use ORAM mechanisms to access the desired blocks in an oblivious manner, and finally run secure two-party protocols to privately compute the desired functionality on the retrieved encrypted blocks. Our construction keeps all sensitive information hidden and reveals only the end result to the legitimate party. Our main technical contribution is the design of a new ORAM that allows for access rights delegation while not requiring the data owner to be online to reshuffle the database. We validate the practicability of our approach through experimental studies.