Privacy Threats and Cryptographic Solutions to Genome Data Processing

Privacy Threats and Cryptographic Solutions to Genome Data Processing

Ugwuoke, C.I. (TU Delft Cyber Security)

Lagendijk, R.L. (promotor)
Erkin, Z. (copromotor)

Delft University of Technology

2021-05-21

The genome is the blueprint of life and has a detailed genotype and phenotype description of any organism. This in itself attributes sensitivity to genetic data, be it in the biological or electronic format. The possibility of sequencing the genome has opened doors to further probing of the data in its electronic form. Post sequencing of the biological genome sample, the electronic genome is stored, processed, and transmitted for variety of purposes including but not limited to Medicare, research, solving crimes and entertainment. However, due to the sensitivity of the genome data, security and privacy of the electronic data is considered to be imperative.
Owing to the privacy and security concerns associated with sharing genome data with third-party entities for processing, various secure and privacy-preserving solutions have been considered. Such scenarios include, a researcher obtains research data which includes genome of individuals, orwhen a healthcare institution outsources the genome of its patients to a cloud environment for storage and processing. In all of these scenarios, it is important that the utility (accuracy and efficiency) of the data is maintained while preserving privacy (confidentiality and unlinkability) simultaneously.
In this thesis,we focus on maintaining data utility when processing electronic genome data as well as preserving the privacy of the individuals whose data are analysed. We employ privacy enhancing techniques such as secure multi-party computation and homomorphic encryption to existing problems and develop provably secure cryptographic protocols that are fit for purpose for each scenario.

GWAS
genome privacy
privacy enhancing techniques
multi-party computation
machine learning

https://doi.org/10.4233/uuid:0c301dea-605b-47a1-a8f5-13b61c4e52ee

978-94-6421-363-8

Institutional Repository

doctoral thesis

© 2021 C.I. Ugwuoke