Safeguarding Privacy in Genome Research: A Comprehensive Framework for Authors

Area of Science:

• Genomics
• Bioinformatics
• Data Privacy

Background:

• Genomic research relies on data sharing for collaboration and discovery.
• Protecting individual genetic information privacy is crucial during data sharing.
• Re-identification risks, like membership inference attacks (MIA), threaten genomic data privacy.

Purpose of the Study:

• To present a bidirectional framework for evaluating privacy risks in genomic research data sharing.
• To assess re-identification risks in shared genomic data, including summary statistics and research datasets.
• To ensure compliance with policies like the National Institutes of Health (NIH) genomic data sharing policy.

Main Methods:

• A structured workflow involving a researcher self-assessment questionnaire.
• Evaluation of data sharing practices and privacy protection measures.
• Calculation of re-identification risk by comparing practices against the NIH genomic data sharing policy.

Main Results:

• Identified potential vulnerabilities in genomic data sharing practices.
• Provided a method to calculate re-identification risk for genomic research papers.
• Demonstrated framework application using published genomic research case studies.

Conclusions:

• Bidirectional frameworks are essential for trustworthy open science and genomic data sharing.
• The framework aids in enhancing privacy measures before research publication.
• Promotes a balance between data sharing and individual privacy protection in genomics.