Cloud-assisted privacy-preserving genetic paternity test

Abstract

The rapid development of the DNA sequencing has greatly decreased the costs in genome sequencing, leading to a high availability of genomic data in the near future. Based on these data, we can realize genomic medical treatment, disease prevention, personalized medicine, social network applications, and etc., which greatly facilitate our daily life. However, since one's genomic data is the ultimate identifier of an individual and contains huge amount of sensitive private information, such as disease predispositions and ancestry information, the use of genomic data may raise serious privacy concerns. In this paper, we focus on one important genomic data application: Paternity Test, which involves the whole DNA sequence (about 3 · 109 letters) and needs huge storage and computation. Considering genomic data user's privacy and the great storage and computing power of the public cloud, it is necessary to find ways of using genomic data in the cloud without disclosure and abuse of the individual's privacy. We first provide a general framework for storing and processing genomic data both in the cloud, then introduce the proposed privacy-preserving scheme for genetic paternity test with RSA's commutative encryption properties in public clouds without disclosure of users' genomic privacy. Our proposed scheme can achieve a higher privacy preserving level and outsource majority of computation to the public cloud. Performance evaluation shows the effectiveness and efficiency of our proposed scheme.