Towards a comprehensive evaluation of dimension reduction methods for transcriptomic data visualization.
Towards a comprehensive evaluation of dimension reduction methods for transcriptomic data visualization.
复制标题
DOI：
10.1038/s42003-022-03628-x
复制
发表时间：
2022-07-19
影响因子：
5.9
通讯作者：
中科院分区：
文献类型：
--
作者：
降维（DR）算法将数据从高维投影到较低维，以实现感兴趣的高维结构的可视化。DR算法广泛用于单细胞转录组数据的分析。尽管广泛使用的DR算法，如t-SNE和UMAP，这些算法的特点，导致缺乏信任：他们不保留高维结构的重要方面，是敏感的任意用户的选择。考虑到从DR中获得见解的重要性，在信任DR方法的结果之前，应该仔细评估它们。在本文中，我们介绍并执行流行的DR方法，包括t-SNE，art-SNE，UMAP，PaCMAP，TriMap和ForceMap 2的系统评估。我们的评估考虑了五个组成部分：局部结构的保存，全局结构的保存，参数选择的敏感性，预处理选择的敏感性和计算效率。该评估可以帮助我们选择符合用户科学目标的DR工具。作者提供了一个评估框架的降维方法，阐明了不同算法的优点和缺点，并应用此框架来评估PCA，t-SNE，UMAP，TriMap，PaCMAP，Forcephalas 2和PHATE算法。
Dimension reduction (DR) algorithms project data from high dimensions to lower dimensions to enable visualization of interesting high-dimensional structure. DR algorithms are widely used for analysis of single-cell transcriptomic data. Despite widespread use of DR algorithms such as t-SNE and UMAP, these algorithms have characteristics that lead to lack of trust: they do not preserve important aspects of high-dimensional structure and are sensitive to arbitrary user choices. Given the importance of gaining insights from DR, DR methods should be evaluated carefully before trusting their results. In this paper, we introduce and perform a systematic evaluation of popular DR methods, including t-SNE, art-SNE, UMAP, PaCMAP, TriMap and ForceAtlas2. Our evaluation considers five components: preservation of local structure, preservation of global structure, sensitivity to parameter choices, sensitivity to preprocessing choices, and computational efficiency. This evaluation can help us to choose DR tools that align with the scientific goals of the user. The authors provide an evaluation framework for dimension reduction methods that illuminates the strengths and weaknesses of different algorithms, and applies this framework to evaluate the PCA, t-SNE, UMAP, TriMap, PaCMAP, ForceAtlas2, and PHATE algorithms.
登录
查看更多内容
影响因子：
64.8
作者：
Cao, Junyue;Spielmann, Malte;Shendure, Jay
通讯作者：
Shendure, Jay
影响因子：
12.3
作者：
Lun AT;Bach K;Marioni JC
通讯作者：
Marioni JC
影响因子：
46.9
作者：
Moon, Kevin R.;van Dijk, David;Krishnaswamy, Smita
通讯作者：
Krishnaswamy, Smita
影响因子：
12.3
作者：
Townes, F. William;Hicks, Stephanie C.;Irizarry, Rafael A.
通讯作者：
Irizarry, Rafael A.
影响因子：
16.6
作者：
Belkina, Anna C.;Ciccolella, Christopher O.;Snyder-Cappione, Jennifer E.
通讯作者：
Snyder-Cappione, Jennifer E.