Adaptive Differential Privacy of Character and Its Application for Genome Data Sharing

Genome sharing leads to privacy concerns of individual, because genome data can uniquely identify an individual and closely associate with inheritance, health, phenotype, and kinship. Existing differential privacy methods cannot achieve expected privacy preserving and expected data utility of genome data. To tackle the problem, we proposed privacy preserving model of genome data sharing satisfying expected privacy preserving and expected data utility. First, we proposed the definition of adaptive differential privacy of character achieving expected privacy preserving and expected data utility, and we gave the adaptive differential privacy mechanisms of character. And also, we theoretically analyzed the properties of definition of adaptive differential privacy of character and its mechanisms. Second, we constructed privacy preserving model of genome data sharing based on adaptive differential privacy mechanisms of character. Finally, our theoretical and experimental analysis showed that the proposed privacy preserving model of genome data sharing using adaptive differential privacy mechanisms of character satisfies expected privacy preserving and expected data utility. Our proposal makes a guideline on designing privacy preserving method of genome data sharing achieving expected privacy preserving and expected data utility