Single cell transcriptome sequencing on the Oxford Nanopore platform with ScNapBar

Christoph Dieterich (University Hospital Heidelberg) explained how, when they started their work, they realised that single-cell RNA sequencing platforms lacked robust solutions for full-length RNA sequencing. They therefore wanted to combine the nanopore sequencing platform and its long-read output with the 10x method of generating single-cell cDNA libraries. They designed the ScNapBar software for analysis, and Christoph stated that this tool is particularly helpful if sequencing saturation (UMI coverage) is low.

Christoph demonstrated how ScNapBar works ‘behind the scenes’, comparing its performance to another analysis method, Sicelore, on simulated nanopore sequencing data. He showed, for example, how ScNapBar has similar sensitivity, but higher specificity than Sicelore, and displayed graphs of CPU processing time (which highly depends on the number of barcodes) for both tools.

He next focused on their single-cell sequencing experiment performed on a 1:1 pool of cells: wildtype GFP+ HEK293 cells, and NMD- mutant GFP- HEK293 cells which had a deactivated nonsense-mediated RNA decay pathway. Sequencing was performed on GridION and a traditional short-read sequencing platform. Explaining that the GFP label itself was not used in cell barcode assignment, Christoph displayed plots showing very clear separation of the GFP label as a result of their accurate ScNapBar barcoding assignment workflow.

Christoph lastly discussed their research into splice variation and NMD in the HEK293 mutant and wildtype cells, comparing isoforms expressed in each cell line. He shared the example of how the wildtype isoforms of the well-established NMD target SRSF2 (serine/arginine-rich splicing factor 2) were expressed in both GFP+/- cells, although in GFP- (mutant) cells the premature-termination-containing isoforms were more abundant. Looking at the read alignment results, Christoph explained how you can clearly see that inclusion of the third exon (the so-called ‘poison’ exon) is favoured in the GFP- mutant cells. Christoph stated how these findings of SRSF2 isoform-specific expression differences in wildtype vs. mutant cells demonstrate the success of their approach.