A Lightweight Workflow for Targeted Long-Read Transcriptomic Profiling Using Oxford Nanopore Sequencing

Long-read sequencing technologies provide portable and flexible service, making them attractive for small-scale sequencing studies. However, many existing RNA-sequencing analysis frameworks are designed for transcriptome-wide analyses and require substantial computational resources. Here we present a lightweight and reproducible computational pipeline for targeted long-read transcriptomic profiling using Oxford Nanopore Technologies (ONT) cDNA sequencing data. The pipeline was evaluated using targeted long-read transcriptomic datasets generated from formalin-fixed paraffin-embedded (FFPE) colorectal carcinoma samples previously classified as microsatellite instability-high (MSI-high) by PCR-based testing. Libraries were sequenced on the Oxford Nanopore MinION platform using R10.4.1 flow cells. Application of the workflow enabled rapid quantification of mismatch repair gene expression and detection of immune-related transcripts including CD8A, PDCD1, and HAVCR2 across multiplexed barcode samples. The pipeline performs targeted alignment of long-read sequencing data to a custom transcript reference panel using minimap2, followed by gene-level read counting and normalization using reads-per-million (RPM). Optional modules enable immune marker profiling, detection of reads aligning to multiple genes, exploratory variant analysis, and visualization of expression patterns. By combining simplicity, reproducibility, and minimal computational overhead, the present pipeline provides an accessible framework for targeted transcriptomic analysis of long-read sequencing data. It may facilitate adoption of ONT-based transcriptomic profiling in settings with restricted computational resources.