Privacy-preserving outsourced gene data search in encryption domain

Human genome project is a grand scale scientific work, which aims at measuring three billion base pairs in human chromosomes (haploid). It brings a great challenging task to store and utilize these gene sequences (GS) securely and effectively. With the development of the cloud computing, the storage of gene information can be out of consideration. However, their secure utilization still puzzles data owners and data users. One popular way is to encrypt these GS and construct searchable indexes for secure retrieval. In this paper, we first define and solve the problem of privacy-preserving outsourced gene data search in encryption domain. We transfer GS into numerical vectors by reasonable mapping for ease of similarity calculation. We employ secure KNN algorithm to encrypt the query, index, and gene data and compute relevance scores securely. We test our scheme through a real-world dataset: plant GS from National Center of Biotechnology Information. Extensive experiments are conducted to demonstrate the efficiency of the proposed scheme. Copyright (c) 2016 John Wiley & Sons, Ltd.