Year: 2024
Type: dissertation
Abstract: Large genomic data sets are generated from various activities, including genealogical research, biomedical studies, and clinical applications, providing significant value but often restricted by privacy concerns. Beacon services have emerged to broaden accessibility to such data, enabling queries for specific alleles to inform clinical decisions. However, these services can inadvertently reveal individual participation in datasets. To address these challenges, the Beacon v2 Interface for Accessing Genomic Information, developed by the Global Alliance for Genomics and Health (GA4GH), aims to create a secure, user-friendly solution for querying population-level genomic data, emphasizing privacy and security. This project, built on Daniel Naro?s work for the GenClinLab Project [92], incorporates a proxy interface to enhance the security, privacy, and granularity of genomic data management. By using Differential Privacy (DP) techniques, the interface ensures the anonymity of individual data, which is vital for research in rare disease genetics and cancer. The innovative algorithmic framework introduced in this project will ensure privacy with minimal impact on data utility, addressing both batch and online query settings. Ultimately, this interface aims to be a cornerstone in the genomic data-sharing ecosystem, advancing scientific and clinical research while upholding the highest standards of privacy and security.
Author: Marmol Asenjo, Alberto
Institution: Universitat Politècnica de Catalunya
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7111522538
