Nanopore sequencing accurately resolves alternatively spliced transcripts

Short-read sequencing technologies struggle with transcripts of genes with multiple isoforms, due to difficulties in measuring exon connectivity where many exons are located further away from each other than the maximum read length¹, ². Bolisetty et al.¹ used nanopore sequencing and the Drosophila model system for a transcriptomic analysis of genes famous for having a large number of isoforms (Dscam1, Rdl, MRP, Mhc).

The team detected a total of 7,899 full-length isoforms for these four genes using a gene-specific RT-PCR approach. Dscam1 is the most abundantly alternatively spliced gene known and it has the potential to generate a total of 38,016 isoforms (115 exons, 95 of which are alternatively spliced, forming 4 clusters)³. For the purpose of the study, the authors focused on the region covering exons 3–10 (19,008 potential isoforms). They were able to reliably resolve 7,874 Dscam1 isoforms with an average identity of >90% for full-length alignments, despite the fact that the exons within each cluster shared 80% nucleotide identity with each other (Figure 1).

Bolisetty et al.¹ demonstrated that long nanopore reads are particularly suited for the characterisation of multipleisoform transcripts, even in cases where exons have significant similarity to one another.

A total of 7,874 Dscam1 isoforms, with an average identity of >90% for full-length alignments, were reliably resolved.