PSQAN: a pipeline to prioritise novel and biologically relevant transcripts from long-read RNA sequencing

Motivation Long-read RNA sequencing has the potential to accurately quantify transcriptomes and reveal the isoform diversity of disease-causing genes. However, despite the recent advances in analysis tools for transcript discovery, long-read RNA sequencing data is still challenging to analyse, due to the detection of hundreds or even thousands of novel transcripts per gene. Results Here, we introduce PSQAN, a workflow to help researchers prioritise high-confidence and potentially biologically relevant transcripts associated with candidate genes and make transcript characterisation results more interpretable. PSQAN performs a gene-based analysis on characterised transcripts generated by SQANTI3 and TALON. PSQAN re-groups transcripts into easily interpretable categories to facilitate their prioritisation, allows transcript-level expression thresholds, and generates visualisations to determine optimal expression thresholds. Overall, we demonstrate that PSQAN is a useful tool which enables users to identify known and novel transcripts of potential biological importance. Availability and implementation PSQAN is an analysis workflow implemented in Snakemake and R and is licensed under the GNU General Public License version 3. The source code and documentation of this tool is available at https://github.com/sid-sethi/PSQAN.