%0 Journal Article
%T 基于t-SNE与UMAP降维的细胞分类及差异化基因筛选研究<br>Cell Classification and Differential Gene Screening Based on t-SNE and UMAP Dimension Reduction
%A 李元夫
%J Advances in Applied Mathematics
%P 6951-6958
%@ 2324-8009
%D 2022
%I Hans Publishing
%R 10.12677/AAM.2022.1110737
%X 单细胞RNA测序技术已经广泛地应用于细胞异质性等关键生物学问题的研究中，与此同时该技术的发展也为基因数据分析提出了很大的挑战。本文基于t-SNE和UMAP两种非线性降维方法，对单细胞RNA数据进行降维、聚类并与线性主成分降维聚类结果进行对比，得出结论：UMAP方法针对单细胞RNA数据降维聚类的效果更为理想。最后以UMAP非线性降维聚类的结果为例筛选出不同细胞类别中的显著差异化基因。<br />
Single-cell RNA sequencing technology has been widely used in key biological problems such as cell heterogeneity, and at the same time, the development of this technology also poses great challenges in gene data analysis. In this paper, based on two nonlinear dimensionality reduction methods, t-SNE and UMAP, the dimensionality reduction and clustering of single-cell RNA data were carried out and compared with the results of linear principal component dimensionality reduction cluster-ing. The conclusion was drawn that the UMAP method was more ideal for the dimensionality reduc-tion clustering of single-cell RNA data. Finally, the results of UMAP nonlinear dimensionality reduc-tion clustering were taken as an example to screen out the significantly differentiated genes in dif-ferent cell categories.
%K 单细胞RNA测序，t-SNE，UMAP，显著差异化基因<br>Single-Cell RNA Sequencing
%K t-SNE
%K UMAP
%K Significantly Differentiated Genes
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=56610