Establishment of methodology for assessment of disclosure risk and confidentiality of individual genome by random partition

2014 Fiscal Year Final Research Report

• Project/Area Number: 24300109
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Statistical science
• Research Institution: The Institute of Statistical Mathematics
• Principal Investigator: MANO Shuhei 統計数理研究所, 数理・推論研究系, 准教授 (20372948)
• Co-Investigator(Kenkyū-buntansha): TSUNODA Tatsuhiko 理化学研究所, 統合生命医科学研究センター, 医科学数理研究グループリーダー (10273468)
• Co-Investigator(Renkei-kenkyūsha): OOTA Hiroki 北里大学, 医学部, 准教授 (40401228)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 統計数学 / ゲノム

Outline of Final Research Achievements

If an individual genome is disclosed and the holder is identified, the person's sensitive information is disclosed. This problem is similar to the statistical disclosure control problem. We need methods to suppress the probability that a sample unique genome with some concealment becomes population unique. The fundamental statistical model is a random partition. In this research, the purpose was establishment of methodology for assessment of disclosure risk and confidentiality of individual genome. A computational method for a random partition with accounting correlation among points on a genome is proposed, and asymptotic properties of Gibbs-type partitions, which are in a general class of random partitions, were discussed. The methods for assessment of disclosure risk and confidentiality were discussed with people who obtain individual genome data and use them.

Free Research Field

統計科学