Efficient and secure outsourcing of genomic data storage.

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Cloud computing offers efficient solutions for managing large genomic datasets.
• Storing and processing sensitive genomic data in the cloud raises significant privacy and security concerns.
• Advanced techniques are needed to protect genomic data from untrusted cloud providers while allowing data analysis.

Purpose of the Study:

• To develop a novel privacy-preserving algorithm for secure cloud-based storage and searching of genomic data.
• To ensure data and query confidentiality against potential breaches in public cloud environments.
• To enable efficient retrieval of specific genetic variants from encrypted datasets.

Main Methods:

• The study presents a new algorithm combining optimal encoding for genomic variants with homomorphic encryption and private information retrieval.
• The algorithm is designed for fully outsourcing the storage of large genomic data files to public clouds.
• Implementation was done in C++ and validated using real-world data from the 2016 iDash Genome Privacy-Protection Challenge.

Main Results:

• The proposed solution demonstrates superior performance compared to existing state-of-the-art methods.
• Researchers can search through millions of encrypted genomic variants in mere seconds.
• The algorithm effectively protects both data and query confidentiality.

Conclusions:

• Sophisticated privacy-enhancing technologies (PETs) are practical and efficient for real-world genomic data applications.
• The developed algorithm overcomes previous assumptions about the inoperability of advanced PETs in operational settings.
• This work facilitates secure and efficient cloud-based genomic data analysis.