AUTHOR=Xiang Ruizhi , Wang Wencan , Yang Lei , Wang Shiyuan , Xu Chaohan , Chen Xiaowen TITLE=A Comparison for Dimensionality Reduction Methods of Single-Cell RNA-seq Data JOURNAL=Frontiers in Genetics VOLUME=Volume 12 - 2021 YEAR=2021 URL=https://www.frontiersin.org/journals/genetics/articles/10.3389/fgene.2021.646936 DOI=10.3389/fgene.2021.646936 ISSN=1664-8021 ABSTRACT=Single-cell RNA sequencing (scRNA-seq) is a high-throughput sequencing technology performed at the level of an individual cell, which can have potential to understand cellular heterogeneity. However, scRNA-seq data are high-dimensional, noisy, and sparse data. Dimension reduction is an important step in down-stream analysis of scRNA-seq. Therefore, several dimension reduction methods have been developed. We compared the performance of ten dimensionality reduction methods on 25 sets of simulation data and 10 sets of real data were used for testing. We evaluated their performance regarding K-means based Adjusted Rand Index (ARI), Normalized Mutual Information (NMI) and silhouette coefficient. Finally, we assessed the running time and hyperparameter tuning of all the methods and gave the user appropriate suggestions. The evaluation shows, no single method dominated on all of the metrics and test datasets. VAE has always shown good robustness and is superior to other methods based on ARI and NMI, but not is good at separating the distinct cell types. UMAP well preserves the original cohesion and separation of cell populations. In addition, it is worth noting that users need to set the hyperparameters according to the specific situation before using the dimensionality reduction methods based on non-linear model and neural network.