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

作     者：Zhenlan Liang Min Li Ruiqing Zheng Yu Tian Xuhua Yan Jin Chen Fang-Xiang Wu Jianxin Wang Zhenlan Liang;Min Li;Ruiqing Zheng;Yu Tian;Xuhua Yan;Jin Chen;Fang-Xiang Wu;Jianxin Wang

作者机构：School of Computer Science and EngineeringCentral South UniversityChangsha 410083China College of MedicineUniversity of KentuckyLexingtonKY 40536USA Division of Biomedical EngineeringUniversity of SaskatchewanSaskatoonSK S7N 5A9Canada 

出 版 物：《Genomics, Proteomics & Bioinformatics》 (基因组蛋白质组与生物信息学报（英文版）)

年 卷 期：2021年第19卷第2期

页      面：282-291页

核心收录：

学科分类：0831[工学-生物医学工程（可授工学、理学、医学学位）] 0711[理学-系统科学] 07[理学] 08[工学] 

基　　金：the Natural Science Foundation of China(NSFC)-Zhejiang Joint Fund for the Integration of Industrialization and Information(Grant No.U1909208) the 111 Project,China(Grant No.B18059) the Hunan Provincial Science and Technology Program,China(Grant No.2019CB1007) the Fundamental Research Funds for the Central Universities-Freedom Explore Program of Central South University,China(Grant No.2019zzts592) the Natural Science Foundation,USA(Grant No.1716340) 

主　　题：Single-cell RNA sequencing Clustering Cell type Similarity learning Enhancement 

摘      要：Accurate identification of cell types from single-cell RNA sequencing(scRNA-seq)data plays a critical role in a variety of scRNA-seq analysis *** task corresponds to solving an unsupervised clustering problem,in which the similarity measurement between cells affects the result *** many approaches for cell type identification have been proposed,the accuracy still needs to be *** this study,we proposed a novel single-cell clustering framework based on similarity learning,called *** models the relationships between cells based on subspace assumption,and generates a sparse representation of the cell-to-cell *** sparse representation retains the most similar neighbors for each ***,three classical pairwise similarities are incorporated with a gene selection and enhancement strategy to further improve the effectiveness of *** on ten real scRNA-seq datasets and five simulated datasets,SSRE achieved the superior performance in most cases compared to several state-of-the-art single-cell clustering *** addition,SSRE can be extended to visualization of scRNA-seq data and identification of differentially expressed *** matlab and python implementations of SSRE are available at https://***/CSUBioGroup/SSRE.