Fig. 4: DOLPHIN enhances alternative splicing detection and analysis.
From: DOLPHIN advances single-cell transcriptomics beyond gene level by leveraging exon and junction reads
a–c Detection of alternative splicing (AS) events across three datasets: Top: full-length PBMC, Middle: 10X colon, and Bottom: 10X rectum. a DOLPHIN identifies significantly more AS events, including exon skipping (ES) and mutually exclusive exons (MXE), compared to the baseline Outrigger tool, demonstrating superior sensitivity in detecting splicing variations. b Scatter plots of Percent Spliced-In (PSI) values show that DOLPHIN achieves higher correlation with pseudo-bulk data (used as a proxy ground truth), indicating more accurate AS quantification than conventional approaches. c UMAP plots based on PSI values reveal that DOLPHIN captures distinct cell-type-specific splicing patterns with greater clarity and biological relevance, improving resolution of splicing events missed by baseline methods. d Sashimi plots for the AS event HsaEX0051104 in the full-length PBMC dataset show stronger junction read signals after DOLPHIN aggregation, enabling detection of splicing events overlooked by conventional methods. e Similarly, for the AS event HsaEX0013878 in the 10X colon dataset, DOLPHIN enhances junction read signals, uncovering AS events missed by the baseline approaches. P values from one-sided Student’s t-tests: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Exact P values are provided in the source data. Source data are provided as a Source Data file.
