Scasa

Scasa performs isoform-level quantification of single-cell RNA sequencing (scRNA-seq) data to resolve transcript-level expression and detect isoform-specific cellular heterogeneity in datasets affected by 3' bias such as Chromium Single Cell 3' (10× Genomics).

Key Features:

• Isoform-Level Quantification: Provides transcript-level expression estimates to capture isoform-specific variation that is not apparent in gene-level quantification.
• Addresses 3' Bias: Targets limitations of high-throughput 3' biased protocols, specifically Chromium Single Cell 3' from 10× Genomics, to enable isoform analysis from such data.
• Transcription Clusters and Isoform Paralogs: Leverages transcription clusters and isoform paralogs to differentiate and quantify distinct isoforms of the same gene.
• Performance Benchmarking: Demonstrates competitive performance in simulation studies against Alevin, Cellranger, Kallisto, Salmon, Terminus, and STARsolo at both isoform- and gene-level tasks.
• Reanalysis of CITE-Seq: Applied to a CITE-Seq dataset and revealed previously undetected subgroups within CD14 monocytes.

Scientific Applications:

• Isoform identification and abundance: Enables identification of specific isoforms and estimation of their relative abundances in single cells.
• Cellular subset discovery: Facilitates detection of isoform-defined cellular subpopulations, exemplified by subgroups within CD14 monocytes in a CITE-Seq reanalysis.
• Disease mechanism investigation: Supports studies of biological processes and disease mechanisms where isoform diversity is functionally relevant.
• Isoform-specific target discovery: Aids identification of potential therapeutic targets that are specific to particular isoforms.

Methodology:

Scasa quantifies isoform-level expression by leveraging transcription clusters and isoform paralogs; it was benchmarked in simulation studies against Alevin, Cellranger, Kallisto, Salmon, Terminus, and STARsolo and applied to a CITE-Seq dataset to identify subgroups within CD14 monocytes.