||
第一作者:Kevin R. Moon
第一单位:美国犹他州立大学
通讯作者:Smita Krishnaswamy
Abstract
背景回顾:The high-dimensional data created by high-throughput technologies require visualization tools that reveal data structure and patterns in an intuitive form.
主要工作:We present PHATE, a visualization method that captures both local and global nonlinear structure using an information-geometric distance between data points. 横向比较:We compare PHATE to other tools on a variety of artificial and biological datasets, and find that it consistently preserves a range of patterns in data, including continual progressions, branches and clusters, better than other tools. Fig. 1b, Left, a 2D drawing of an artificial tree with color-coded branches. Right, comparison of PCA, t-SNE and the PHATE visualizations for the high-dimensional artificial tree data. Fig. 1c, Comparison of PCA, t-SNE and the PHATE visualizations for new EB data. 具体评价:We define a manifold preservation metric, which we call denoised embedding manifold preservation (DEMaP), and show that PHATE produces lower-dimensional embeddings that are quantitatively better denoised as compared to existing visualization methods. 实例演示:An analysis of a newly generated single-cell RNA sequencing dataset on human germ-layer differentiation demonstrates how PHATE reveals unique biological insight into the main developmental branches, including identification of three previously undescribed subpopulations. Fig. 6 | PHATE analysis of embryoid body scRNA-seq data with n = 16,825 cells. 应用拓展:We also show that PHATE is applicable to a wide variety of data types, including mass cytometry, single-cell RNA sequencing, Hi-C and gut microbiome data. Fig. 1d, PHATE applied to various datatypes. Left, PHATE on human microbiome data. Middle, PHATE on Hi-C chromatin conformation data. Right, PHATE on iPSC CyTOF data.
摘 要
由高通量技术所获取的高纬度数据需要通过可视化工具来直观地揭示数据结构与模式。本文中,作者开发了一个可视化的算法PHATE,可以利用数据点之间的信息-几何距离来捕获局部和整体的非线性结构。作者比较了PHATE与其它的工具在处理多个人工数据和生物学数据方面的能力。作者定义了一个叫做“去噪嵌入流形保存(DEMaP)”的流形保存指标,发现PHATE能够产生更低维度的嵌入,并且要比其它的几个工具去噪效果更好。作者通过对一个新出的人类胚层分化单细胞转录组测序数据进行了分析,显示了PHATE如何揭示主要发育分支的独特生物学见解,鉴定了三个先前未曾发现的亚群。另外,作者还演示了PHATE适用于其它类型的数据,包括质谱数据、scRNA数据、Hi-C数据以及肠道微生物组数据等。
通讯作者 **Smita Krishnaswamy** 个人简介: 密西根大学,博士。 研究方向:开发无监督的机器学习方法, 尤其是图形信号处理和深度学习方面, 通过对大数据、高通量以及高维度的生物医学数据 进行去噪、归因,可视化及提取结构、模式和关系。
doi: 10.1038/s41587-019-0336-3
Journal: Nature Biotechnology
Published date: December 03, 2019
Archiver|手机版|科学网 ( 京ICP备07017567号-12 )
GMT+8, 2024-9-21 07:52
Powered by ScienceNet.cn
Copyright © 2007- 中国科学报社