Preserving Genotype Privacy Using AES and Partially Homomorphic Encryption

Abstract

     Increasingly, the availability of personal genomic data in cloud servers hosted by hospitals and research centers has incentivized researchers to turn to research that deals with analyzing genomic data. This is due to its importance in detecting diseases caused by genetic mutations, detecting genes that carry genetic diseases, and attempting to treat them in future generations.  Secure query execution on encrypted data is considered an active research area in which encryption is used to ensure the confidentiality of genomic data while restricting the ability to process such data without first decrypting it. To provide a secure framework and future insight into the potential contributions of homomorphic encryption to the field of genomic data, this paper proposes a framework for guaranteeing genomic data privacy using various partial homomorphic encryption techniques. By examining the characteristics of the three partial homomorphic encryptions based on different parameters. The framework has been online tested and compared based on different parameters. Three homomorphic encryption algorithms were adopted to ensure genomic data privacy by employing homomorphic operations in the query matching process. Experiments on real datasets, specifically MERS and SARSr-COV, showed that the proposed framework is efficient and improves query execution time by an average of 96% compared to existing work.