LongTREC presents:
Quality assessment of long read data in multisample lrRNA-seq experiments using SQANTI-reads
Netanya Keil, Carolina Monzó, Lauren McIntyre, Ana Conesa
Long-read RNA sequencing lets scientists read entire RNA molecules and discover which versions of each gene’s transcripts are present. But, as in any experiment, data quality can vary between samples, and poor samples can lead to misleading conclusions. In studies with many samples, spotting these problems early is essential.
The authors present SQANTI-reads, which builds on the widely used SQANTI3 tool. It examines reads and their splicing patterns across all samples in an experiment, and shows quality metrics in visual summaries organised by experimental design, so unusual samples stand out. It also adds new metrics to flag genes that may be poorly annotated, possible new transcripts, and variation at the points where RNA pieces are joined. The tool was demonstrated on a fruit fly development study and on a multi-platform dataset from the international LRGASP benchmarking project.
SQANTI-reads reveals how sequencing depth affects data quality and helps distinguish strong from weak splice sites. It gives researchers a practical way to check their long-read data before analysis, increasing confidence in the results of multi-sample transcriptomics studies.
Read the full text here: https://doi.org/10.1101/gr.280021.124
