Utility and Privacy Assessment of Synthetic Microbiome Data.

The microbial communities of the human body are subject to extensive research efforts. The individual variations in the human microbiome reveal information about our diet, exercise habits and general well-being, and are useful for investigations on the prediction and therapy of diseases. On the other hand, these variations allow for microbiome-based identification of individuals, thus posing privacy risks in microbiome studies. Synthetic microbiome datasets hold the promise of reducing said risks while simultaneously keeping the utility of the data for research as high as possible. In this paper, we conduct an empirical evaluation of two open-source data synthetization tools on several publicly available microbiome datasets. In particular, we generate synthetic training data and investigate its performance for a variety of machine learning tasks on microbiome samples. Our findings indicate the suitability of synthetic microbiome data for analysis and for privacy protection.