DP-DCAN: Differentially Private Deep Contrastive Autoencoder Network for Single-cell Clustering

TL;DR

This paper introduces DP-DCAN, a novel differentially private autoencoder for single-cell clustering that reduces privacy-induced performance loss by perturbing only part of the network, demonstrating superior results and robustness.

Contribution

The paper proposes a partial network perturbation approach in a deep autoencoder to improve privacy-utility trade-off in single-cell data analysis.

Findings

DP-DCAN outperforms traditional DP methods on six datasets.

DP-DCAN maintains high clustering accuracy with reduced noise.

The model shows strong robustness to adversarial attacks.

Abstract

Single-cell RNA sequencing (scRNA-seq) is important to transcriptomic analysis of gene expression. Recently, deep learning has facilitated the analysis of high-dimensional single-cell data. Unfortunately, deep learning models may leak sensitive information about users. As a result, Differential Privacy (DP) is increasingly used to protect privacy. However, existing DP methods usually perturb whole neural networks to achieve differential privacy, and hence result in great performance overheads. To address this challenge, in this paper, we take advantage of the uniqueness of the autoencoder that it outputs only the dimension-reduced vector in the middle of the network, and design a Differentially Private Deep Contrastive Autoencoder Network (DP-DCAN) by partial network perturbation for single-cell clustering. Since only partial network is added with noise, the performance improvement is obvious and twofold: one part of network is trained with less noise due to a bigger privacy budget, and the other part is trained without any noise. Experimental results of six datasets have verified that DP-DCAN is superior to the traditional DP scheme with whole network perturbation. Moreover, DP-DCAN demonstrates strong robustness to adversarial attacks.