| Summary: | The production of real-time sequence data from native, long single-molecule Deoxyribonucleic acid and Ribonucleic acid is unique to Oxford Nanopore Technologies sequencers. Real-time analysis can guide the application of both run-until and read-until, which increases the efficiency of sequencing runs. Run-until allows users to stop a sequencing run at certain thresholds, preserving flow cell health. Read-until,
or adaptive-sampling, can focus genome-wide sequencing onto specific regions of interest enriching coverage over these targets. In Chapter 1, we discuss the development and application of our fully-customisable nanopore simulator, which allowed us to test these tools whilst avoiding the costs of performing real sequencing.
In Chapter 2 we discuss a real-life application of real-time adaptive-sampling: viral pathogenomic surveillance. We describe a framework that optimises sequencing: it can provide adaptive-sampling targets for amplicon-based viral sequencing, increase the speed and efficiency of a run, and further preserve flow cell health via metrics that determine when all samples have produced enough data.
The highest throughput device that Oxford Nanopore Technologies currently offers is PromethION, designed at a scale that further optimises, and lowers the cost of, sequencing. As described in Chapter 3, by developing mappy-rs, a Rust-based extension to minimap2, we further developed and refactored readfish (a previously developed open source adaptive-sampling tool) to permit its use with PromethION scale sequencing. Mappy-rs increases the speed of alignments to enable processing of PromethION output.
To demonstrate the combination of ultra-long, high-throughput PromethION sequencing with adaptive-
sampling, we attempted to close the gaps in a de novo, nanopore-only genome assembly of the NA12878 cell line. This approach shows promise for the creation of haplotyped Telomere to Telomere assemblies using nanopore sequencing in the future.
|
|---|