SCH: EXP: Collaborative Research: Preserving Privacy in Human Genomic Data

University Of Arkansas, Fayetteville AR

Investigators

Abstract

The proposed research will advance theoretical understanding of fundamental issues related to genomic data analysis, as well as the design and implementation of practical techniques to effectively protect privacy. The proposed research is central to preventing privacy breaches for both study participants and regular individuals due to released summary statistics and results. A primary outcome of this research will be a suite of novel tools and technologies and a Web-portal for privacy preserving analysis of genomic data, which will help researchers in their endeavor to meet growing expectations in protecting privacy and provide privacy assurance when regular individuals share their genetic profiles. The proposed research will contribute significantly and creatively to the limited base of knowledge in the area of preserving privacy in genetic analysis. This research will involve, through courses and thesis projects, graduate and undergraduate students to enhance their knowledge and skills in solving problems in data mining, statistical analysis, privacy, and bioinformatics. The PIs will collaborate with industrial partners at Baylor College of Medicine and MD Anderson Cancer Institute to validate the effectiveness of developed tools. The proposed work is tied to a key health problem (i.e. genetic privacy) and will make a fundamental contribution to computer and information sciences, and biomedical research.. The proposed research activities include the following three major tasks: 1) develop novel differential privacy preserving techniques to provide rigorous privacy guarantees for genetic participants when researchers publish genomic summary statistics and/or conduct advanced analysis; 2) systematically evaluate potential privacy breaches for regular individuals due to the released genomic statistics and analysis results; and 3) work with collaborators to evaluate the efficacy of developed methods using real genomic data (cancer and Alzheimer's disease) and build an integrated Web-portal environment to provide researchers secure, reliable, and privacy preserving access to (anonymized) genomic raw data, statistics, and analysis results.