Privacy Preserving Clustering for Distributed Homogeneous Gene Expression Data Sets
In this paper, the authors present a new approach to perform principal component analysis (PCA)-based gene clustering on genomic data distributed in multiple sites (horizontal partitions) with privacy protection. This approach allows data providers to collaborate together to identify gene profiles from a global viewpoint, and at the same time, protect the sensitive genomic data from possible privacy leaks. The authors developed a framework for privacy preserving PCA-based gene clustering, which includes two types of participants such as data providers and a trusted central site. Within this mechanism, distributed horizontal partitions of genomic data can be globally clustered with privacy preservation. Compared to results from centralized scenarios, the result generated from distributed partitions achieves 100% accuracy by using this approach. An experiment on a real genomic data set is conducted, and result shows that the proposed framework produces exactly the same cluster formation as that from the centralized data set.