Differentially Private Genome Data Dissemination Through Top-Down Specialization

We present a novel approach for disseminating genomic data while satisfying differential privacy. The proposed algorithm splits raw genome sequences into blocks, subdivides the blocks in a top-down fashion, and finally adds noise to counts to protect privacy. Preliminary experimental results suggest that the proposed algorithm can retain data utility that is higher than the baseline for a given privacy budget. The proposed algorithm can also be used to protect heterogeneous data, such as records consisting of both medical and genomic data. Further improvement is possible by refining the heuristic for splitting sequences and by introducing a scoring function in the data generalization process.