Identification of stem cells from large cell populations with topological scoring.,Molecular Omics

当前位置： X-MOL 学术 › Mol. Omics › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Identification of stem cells from large cell populations with topological scoring.
Molecular Omics ( IF 3.0 ) Pub Date : 2020-8-13 , DOI: 10.1039/d0mo00039f
Mihaela E Sardiu ₁ , Andrew C Box , Jeffrey S Haug , Michael P Washburn

Affiliation

Machine learning and topological analysis methods are becoming increasingly used on various large-scale omics datasets. Modern high dimensional flow cytometry data sets share many features with other omics datasets like genomics and proteomics. For example, genomics or proteomics datasets can be sparse and have high dimensionality, and flow cytometry datasets can also share these features. This makes flow cytometry data potentially a suitable candidate for employing machine learning and topological scoring strategies, for example, to gain novel insights into patterns within the data. We have previously developed a Topological Score (TopS) and implemented it for the analysis of quantitative protein interaction network datasets. Here we show that TopS approach for large scale data analysis is applicable to the analysis of a previously described flow cytometry sorted human hematopoietic stem cell dataset. We demonstrate that TopS is capable of effectively sorting this dataset into cell populations and identify rare cell populations. We demonstrate the utility of TopS when coupled with multiple approaches including topological data analysis, X-shift clustering, and t-Distributed Stochastic Neighbor Embedding (t-SNE). Our results suggest that TopS could be effectively used to analyze large scale flow cytometry datasets to find rare cell populations.

中文翻译：

通过拓扑评分从大细胞群中鉴定干细胞。

机器学习和拓扑分析方法越来越多地用于各种大规模组学数据集。现代高维流式细胞仪数据集与其他组学数据集（如基因组学和蛋白质组学）共享许多特征。例如，基因组学或蛋白质组学数据集可以是稀疏的并且具有高维数，流式细胞术数据集也可以共享这些特征。这使得流式细胞术数据可能成为采用机器学习和拓扑评分策略的合适候选者，例如，以获得对数据中模式的新见解。我们之前开发了一个拓扑分数 (TopS) 并将其用于分析定量蛋白质相互作用网络数据集。在这里，我们展示了用于大规模数据分析的 TopS 方法适用于分析先前描述的流式细胞术分选的人类造血干细胞数据集。我们证明 TopS 能够有效地将这个数据集分类到细胞群中并识别稀有细胞群。我们展示了 TopS 在与多种方法（包括拓扑数据分析、X-shift 聚类和 t 分布随机邻域嵌入（t-SNE）相结合时）的效用。我们的研究结果表明，TopS 可以有效地用于分析大规模流式细胞仪数据集以寻找稀有细胞群。我们展示了 TopS 在与多种方法（包括拓扑数据分析、X-shift 聚类和 t 分布随机邻域嵌入（t-SNE）相结合时）的效用。我们的研究结果表明，TopS 可以有效地用于分析大规模流式细胞仪数据集以寻找稀有细胞群。我们展示了 TopS 在与多种方法（包括拓扑数据分析、X-shift 聚类和 t 分布随机邻域嵌入（t-SNE）相结合时）的效用。我们的研究结果表明，TopS 可以有效地用于分析大规模流式细胞仪数据集以寻找稀有细胞群。

更新日期：2020-08-13

点击分享查看原文

点击收藏

公开下载

阅读更多本刊最新论文本刊介绍/投稿指南11