NanoTS: a deep learning tool for accurate SNP calling in nanopore long-read transcriptome data.

NanoTS achieves over 0.980 F1 score in SNP detection from nanopore transcriptome data, significantly improving accuracy for both RNA and cDNA sequencing.

• Why it matters: Accurate variant detection in nanopore long-read transcriptome sequencing is difficult, limiting its clinical and research applications.
• What they did: NanoTS employs deep learning to identify SNPs from diverse nanopore transcriptome datasets, focusing on variants with at least five supporting reads.
• The result: NanoTS surpasses existing methods, especially for allelically imbalanced variants, enabling precise detection and genotyping of pathogenic variants relevant to Mendelian disorders.