Private Computation on Encrypted Genomic Data

Abstract

A number of databases around the world currently host a wealth of genomic data that is invaluable to researchers conducting a variety of genomic studies. However, patients who volunteer their genomic data run the risk of privacy invasion. In this work, we give a cryptographic solution to this problem: to maintain patient privacy, we propose encrypting all genomic data in the database. To allow meaningful computation on the encrypted data, we propose using a homomorphic encryption scheme.Specifically, we take basic genomic algorithms which are commonly used in genetic association studies and show how they can be made to work on encrypted genotype and phenotype data. In particular, we consider the Pearson Goodness-of-Fit test, the D and r 2 -measures of linkage disequilibrium, the Estimation Maximization (EM) algorithm for haplotyping, and the Cochran-Armitage Test for Trend. We also provide performance numbers for running these algorithms on encrypted data.

Stream Ciphers: A Practical Solution for Efficient Homomorphic-Ciphertext Compression

“…We chose 7 because, in practice, numerous applications use algorithms of multiplicative depth smaller than 7 (see e.g. [33,42]). In what follows we compare the results we obtain using Trivium, Kreyvium and also the LowMC cipher.…”

“…Because they allow arbitrary computations on encrypted data, FHE schemes suddenly opened the way to exciting new applications, in particular cloud-based services in several areas (see e.g. [46,33,42]). …”

Privacy-safe linkage analysis with homomorphic encryption

“…More recently, Lauter et al [6] conducted a study which demonstrates the application of homomorphic encryption in the analysis of genomic data. The study shows that statistical algorithms (Pearson Goodness-of-Fit Test, r 2 -measures of LD, Estimation Maximization (EM) algorithm for haplotyping, etc) peculiar to genetic studies can be replicated over encrypted data.…”

“…We adopt data packing technique to help manage the data expansion challenge that comes with encrypting the genes. Our encoded data storage and retrieval design makes it easier to dynamically compute the contingency table parameters necessary for computing the statistical measures, which shows a significant improvement from existing works [6], [7] that adopted homomorphic techniques.…”

“…Other studies have also suggested access control and security of databanks as the only measure, but that does not protect the privacy of the participants. The studies [6], [7] recommend cryptographic solutions like homomorphic encryption (HE), to encrypt the data and homomorphically compute the LD statistic measures. Deploying encryption and related cryptographic techniques are preferred because it is easy to mathematically prove the security of cryptosystems and equally extend proofs to the constructed solutions, therefore one is able to estimate possible information leakage.…”