Efficient and Highly Accurate Differentially Private Statistical Genomic Analysis using Discrete Fourier Transform

• Item type: SIG Technical Reports(1)
• 公開日: 2023-02-27
• タイトル: Efficient and Highly Accurate Differentially Private Statistical Genomic Analysis using Discrete Fourier Transform
• タイトル / 言語: en
• タイトル: Efficient and Highly Accurate Differentially Private Statistical Genomic Analysis using Discrete Fourier Transform
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: プライバシー保護
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: The University of Tokyo
• 著者所属: The University of Tokyo
• 著者所属(英): en
• 著者所属(英): The University of Tokyo
• 著者所属(英): en
• 著者所属(英): The University of Tokyo
• 著者名: Akito, Yamamoto; Tetsuo, Shibuya
• 著者名(英): Akito, Yamamoto; Tetsuo, Shibuya
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: As the amount of data containing human genome information increases, these data will be further utilized in medicine. However, if the genomic statistics are released unchanged, there is a risk of identifying individuals. Although there are several privacy-preserving methods to release them, they have the problems of poor accuracy and intensive computational complexity. In this paper, we propose differentially private methods with both efficiency and high accuracy to release the top K significant SNPs. First, we propose an extended Fourier perturbation algorithm with more accurate privacy guarantees. Then, we present novel methods combining DFT with the Laplace and exponential mechanisms. These methods take only O(mlogm) time for a dataset containing m SNPs. We also theoretically guarantee that the value of sensitivity for these methods is smaller than that for existing methods. The experimental results indicate that our methods are advisable rather than existing methods.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: As the amount of data containing human genome information increases, these data will be further utilized in medicine. However, if the genomic statistics are released unchanged, there is a risk of identifying individuals. Although there are several privacy-preserving methods to release them, they have the problems of poor accuracy and intensive computational complexity. In this paper, we propose differentially private methods with both efficiency and high accuracy to release the top K significant SNPs. First, we propose an extended Fourier perturbation algorithm with more accurate privacy guarantees. Then, we present novel methods combining DFT with the Laplace and exponential mechanisms. These methods take only O(mlogm) time for a dataset containing m SNPs. We also theoretically guarantee that the value of sensitivity for these methods is smaller than that for existing methods. The experimental results indicate that our methods are advisable rather than existing methods.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AN10116224
• 書誌情報: 研究報告マルチメディア通信と分散処理（DPS）; 巻 2023-DPS-194, 号 11, p. 1-8, 発行日 2023-02-27
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8906
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会