Fig. 5: Applying InPACT on a human FBM scRNA-seq dataset identifies cell type-specific IPA events. | Nature Communications

Fig. 5: Applying InPACT on a human FBM scRNA-seq dataset identifies cell type-specific IPA events.

From: InPACT: a computational method for accurate characterization of intronic polyadenylation from RNA sequencing data

Fig. 5

a The t-distribute stochastic neighbor embedding (t-SNE) plot of 12 different cell types in human FBM on gene expression level. The HSC represents hematopoietic stem cell, the PMN represents polymorphonuclear leukocytes, the DC represents dendritic cells and the pDC represents plasmacytoid dendritic cells. b The bar plot shows the number of genes with novel identified IPA events for each cell type. c, d Two examples of InPACT-identified novel IPA events in human FBM scRNA-seq data. The plots show the read coverage in the locus for SRP68 and HMGCL, with the annotation of Refseq isoforms and InPACT-identified novel IPA isoforms. The read coverage is shown for each cell type in different colors. e The heatmap depicts cell type-specific IPA events for each cell type. The enriched GO terms (biological process) are listed on the right panel with Benjamini-Hochberg adjusted P-values estimated by hypergeometric test (two-sided). f The stacked violin plots depict examples of cell type-specific IPA events. The Y-axes represent the relative usage of cell type-specific IPA events.

Back to article page