Fig. 5: CaSpER algorithm applied to single-cell GBM RNA-seq dataset. | Nature Communications

Fig. 5: CaSpER algorithm applied to single-cell GBM RNA-seq dataset.

From: CaSpER identifies and visualizes CNV events by integrative analysis of single-cell or bulk RNA-sequencing data

Fig. 5

a Heatmap of smoothed expression signal of all the genes across all samples is shown in the top panel. The color codes are explained on the right. b Smoothed BAF signal from the pooled patient-specific reads is shown in the plot. The smoothed patient-specific BAF signal shows shifts in deleted and amplified chromosomes. c The heatmap of summarized large-scale CNV events using the common events in all scale pairs is plotted. Columns represent chromosome arms whereas rows represent cells. The color codes represent the patient id. d MGH31 consists of two mutually exclusive sub-clones where one sub-clone contains chromosome 5q amplification whereas the other sub-clone contains chromosome 14q deletion. Additionally, one sub-clone contains 1p amplification and the other sub-clone contains 13q deletion. e Clustering of large-scale events generated by CaSpER in patient MGH31. Normal cells are clustered separately with a different CNV profile. Cells within the red rectangle correspond to normal cells. Rows correspond to cells whereas columns correspond to chromosome arms. Clustering separates cells harboring 1p and 5q amplification from cells harboring 13q and 14q deletion. f Mutually exclusive and co-occurring CNV events are plotted as a graph. Red colored events are amplified whereas blue colored events are deleted. The solid lines represent co-occurring events, whereas dashed lines represent mutually exclusive events. Edge width increases with event significance, which is assessed using Fisher's Exact test p-value (edge width = −log2(p-value)). The mutually exclusive 1p:13q and 5q:13q, 5q:14q event pairs for patient MGH31 is significant.

Back to article page