Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: An elastic-net logistic regression approach to generate classifiers and gene signatures for types of immune cells and T helper cell subsets

Fig. 3

Immune cell annotation prediction based on scRNA-seq data against prior annotations reported in melanoma dataset. a The inner pie chart summarizes the cell annotations reported by Tirosh et al. [23] and includes 298 unannotated CD45-positive non-malignant cells (labeled as Unresolved) isolated from melanoma tissue samples. Unannotated samples were acquired following gating for CD45+ single cells and classified as non-malignant based on inferred copy number variation patterns. Using gene expression values reported for each scRNA-seq sample, a new cell annotation was determined based on the closest match with the alternative cell signatures determined using elastic-net logistic regression, which are summarized in outer pie chart. b The contour plot for the likelihood of a sample to be either an NK cell or CD8+ T cell based on gene expression stratified by cells previously annotated by [23] to be T cells, macrophages, B cells, or NK cells

Back to article page