Data Sanitization to Reduce Private Information Leakage from Functional Genomics.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Functional genomics datasets are rapidly expanding, offering insights into gene regulation and organismal phenotypes.
• Next-generation sequencing in these studies raises privacy concerns, limiting broad data sharing.
• Existing controlled-access models hinder the sharing of raw reads crucial for statistical power and reproducibility.

Purpose of the Study:

• To develop a data-sanitization procedure for sharing raw functional genomics reads.
• To minimize privacy leakage while enabling principled privacy-utility trade-offs.
• To facilitate broader sharing of functional genomics data.

Main Methods:

• A novel data-sanitization protocol was developed for raw functional genomics reads.
• The procedure quantifies privacy leakage by statistically linking study participants to known individuals.
• Linkage analysis was performed using both high-accuracy reference genomes and environmental samples.

Main Results:

• The data-sanitization procedure effectively minimizes privacy leakage from functional genomics reads.
• The protocol is compatible with traditional Illumina assays and advanced 10x single-cell RNA sequencing.
• The method allows for balancing privacy protection with data utility for research purposes.

Conclusions:

• A new method for sanitizing functional genomics data allows for wider sharing.
• This approach addresses privacy concerns associated with next-generation sequencing data.
• The developed protocol supports enhanced research reproducibility and statistical power.