SSRE: Cell Type Detection Based on Sparse Subspace Representation and Similarity Enhancement [article]

Zhenlan Liang, Min Li, Ruiqing Zheng, Yu Tian, Xuhua Yan, Jin Chen, Fangxiang Wu, Jianxin Wang
2020 bioRxiv   pre-print
Accurate identification of cell types from single-cell RNA sequencing (scRNA-seq) data plays a critical role in a variety of scRNA-seq analysis studies. It corresponds to solving an unsupervised clustering problem, in which the similarity measurement between cells in a high dimensional space affects the result significantly. Although many approaches have been proposed recently, the accuracy of cell type identification still needs to be improved. In this study, we proposed a novel single-cell
more » ... stering framework based on similarity learning, called SSRE. In SSRE, we model the relationships between cells based on subspace assumption and generate a sparse representation of the cell-to-cell similarity, which retains the most similar neighbors for each cell. Besides, we adopt classical pairwise similarities incorporated with a gene selection and enhancement strategy to further improve the effectiveness of SSRE. For performance evaluation, we applied SSRE in clustering, visualization, and other exploratory data analysis processes on various scRNA-seq datasets. Experimental results show that SSRE achieves superior performance in most cases compared to several state-of-the-art methods.
doi:10.1101/2020.04.08.028779 fatcat:zj7du2nkp5arfah2jfkewj3jru