The accurate characterisation of fusion transcripts is of high importance for clinical research into diseases including some forms of cancer. However, their identification via traditional short-read sequencing approaches requires transcripts to be sequenced in small fragments before being reassembled computationally, which can lead to multimapping and misassembly. With long-read nanopore sequencing, fusion transcripts can be sequenced end-to-end in single reads, enabling comprehensive characterisation of fusions and their precise splice junctions.
Full-length sequencing of fusion transcripts
Fusion transcripts, representing the product of a fusion gene or the splicing together of transcripts encoded by different genes, are significant in clinical research due to their association with many diseases, including some cancers. Sequencing of transcripts using traditional RNA-Seq approaches requires cDNA sequencing libraries to be fragmented and sequenced in reads of ~50–100 bases; the full transcript sequences are then deduced via computational assembly of the reads. However, such short reads may make precise breakpoint mapping of fusions difficult, and can result in incorrectly assembled transcripts and high levels of multimapping (Figure 1), whereby reads cannot be assigned to a single transcript. Furthermore, the requirement for PCR may mean that transcripts which are difficult to amplify may be poorly represented or missing from sequencing data. With nanopore sequencing, there is no upper read length limit, and fragmentation is not required: transcripts can be sequenced end-to-end in single long reads, enabling unambiguous identification of fusion transcripts.
Tailor your workflow to your experimental goals with versatile RNA and cDNA sequencing methods
With versatile options for library prep and sequencing available, nanopore workflows can be tailored to suit your needs. Oxford Nanopore has developed the only method of sequencing RNA molecules in their native form with direct RNA sequencing, eliminating PCR bias and enabling base modifications to be identified alongside the nucleotide sequence. Direct cDNA sequencing is also PCR-free and provides higher throughput for applications requiring greater coverage, whilst PCR-cDNA sequencing is optimised for highest throughput with very low PCR bias. Each approach provides both quantitative and qualitative data (Figure 2): fusion transcripts can be detected and counted with confidence from a single dataset. Rapid turnaround times can be achieved through real-time sequencing and analysis on the portable, cost-effective MinION or Flongle, the flexible GridION, or scaled up for maximum throughput on the powerful PromethION. Samples can also be sequenced in multiplex to batch samples as needed and further reduce cost per sample.
Enrich and characterise fusion transcripts with semi-specific RT-PCR
In some instances, fusion transcripts may be produced by a fusion gene resulting from one of several translocation events; the identity of both genes may not be known, meaning that primers cannot be designed to target both ends of the transcript. Here, semi-specific RT-PCR can be used to enrich for fusion transcripts where the fusion partner is unknown (Figure 3). First, the total RNA in a library is reverse-transcribed via a VN primer, which hybridises to any poly(A)-tailed RNA and is tailed with a primer site for subsequent amplification. Semi-specific PCR is then performed, amplifying transcripts using one primer complementary to the tailed cDNA molecules and one targeting the known end of the transcript. In this way, full-length wildtype and fusion transcripts can be enriched and sequenced with high depth of coverage, enabling accurate detection of fusions.
Rapid identification of oncogenic gene fusions with cDNA sequencing
‘Nanopore sequencing fits perfectly with the needs of clinical fusion detection’William Jeck, Nanopore Community Meeting 2018
For diseases caused by gene fusions, such as acute promyelocytic leukaemia, time to treatment is often crucial; however, identification of these fusions via traditional methods can involve turnaround times of a week or longer. In his talk at Nanopore Community Meeting 2018, William Jeck described how he and his team have used nanopore cDNA sequencing to develop a rapid method of oncogenic gene fusion detection from clinical research samples. The method uses Anchored Multiplex PCR (AMP) to generate targeted cDNA libraries; these are then sequenced in multiplex on the MinION. The pipeline was first used to successfully identify a BCR-ABL1 fusion in an erythroleukemia cell line, with the first fusion reads generated within seconds of starting the run; the long reads enabled resolution of long-range exon structures across the fusion. A PML-RARA fusion was identified in a leukemia sample, and an ELBR-FLI1 gene fusion was successfully identified via sequencing on a Flongle Flow Cell, demonstrating its future potential as a cost-effective fusion detection tool.
How do I detect fusion transcripts with nanopore sequencing?
Enrichment of fusion transcripts can be performed using sequence-specific (where both ends of the transcript is known) or semi-specific (where one end of the transcript is known) RT-PCR. Preparation of libraries using the PCR-cDNA Sequencing Kit, followed by sequencing on one MinION Flow Cell, delivers ~18 million cDNA reads, for very high depth of coverage. Sequencing can be scaled down further on smaller Flongle Flow Cells for cost-effective long-read sequencing. Both the MinION and the GridION are compatible with MinION and Flongle Flow Cells; the portable MinION device is ideal for sequencing at the point of sampling, whilst the GridION enables sequencing on up to five individually addressable flow cells for flexible, on-demand analysis. A number of robust tools are available for analysing full-length nanopore RNA sequencing reads, both from Oxford Nanopore and the Nanopore Community. Visit the Bioinformatics section of the Nanopore Community for data analysis tutorials.
Discover more about the advantages of full-length transcript sequencing with Oxford Nanopore in the RNA sequencing white paper.