Olgun Gulden, Gopalan Vishaka, Hannenhalli Sridhar
Cancer Data Science Lab, National Cancer Institute, National Institutes of Health, Bethesda, MD 20892, USA.
iScience. 2022 Aug 17;25(9):104962. doi: 10.1016/j.isci.2022.104962. eCollection 2022 Sep 16.
Our understanding of miRNA activity at cellular resolution is thwarted by the inability of standard scRNA-seq protocols to capture miRNAs. We introduce a novel tool, miRSCAPE, to infer miRNA expression in a sample from its RNA-seq profile. We establish miRSCAPE's accuracy in 10 tumor and normal cohorts demonstrating its superiority over alternatives. miRSCAPE accurately infers cell type-specific miRNA activities (predicted versus observed fold-difference correlation ∼0.81) in two independent scRNA-seq datasets. We apply miRSCAPE to infer miRNA activities in scRNA clusters in pancreatic and lung adenocarcinomas, as well as in 56 cell types in the human cell landscape (HCL). In pancreatic and breast cancer scRNA-seq data, miRSCAPE recapitulates miRNAs associated with stemness and epithelial-mesenchymal transition (EMT) cell states, respectively. Overall, miRSCAPE recapitulates and refines miRNA biology at cellular resolution. miRSCAPE is freely available and is easily applicable to scRNA-seq data to infer miRNA activities at cellular resolution.
我们对miRNA在细胞分辨率下活性的理解受到标准单细胞RNA测序（scRNA-seq）方案无法捕获miRNA的阻碍。我们引入了一种新型工具miRSCAPE，用于从RNA测序谱推断样本中的miRNA表达。我们在10个肿瘤和正常队列中确定了miRSCAPE的准确性，证明了它相对于其他方法的优越性。在两个独立的scRNA-seq数据集中，miRSCAPE准确地推断出细胞类型特异性的miRNA活性（预测与观察到的倍数差异相关性约为0.81）。我们应用miRSCAPE推断胰腺和肺腺癌scRNA簇以及人类细胞图谱（HCL）中56种细胞类型的miRNA活性。在胰腺癌和乳腺癌scRNA-seq数据中，miRSCAPE分别概括了与干性和上皮-间质转化（EMT）细胞状态相关的miRNA。总体而言，miRSCAPE在细胞分辨率下概括并完善了miRNA生物学。miRSCAPE可免费获取，并且很容易应用于scRNA-seq数据，以推断细胞分辨率下的miRNA活性。
