Illumina released a new NovaSeq flowcell today: S Prime (SP). This allows users to run a very small number of samples and is much more akin to a HiSeq flowcell; one SP flowcells generates 1.6 billion reads or about 4-5 lanes of a HiSeq. It was not immediately clear if this is a single lane flowcells – but I’d expect this to be so making the instrument easier to run for smaller experiments.
I wrote several posts when NovaSeq was launched (here, here & here) and Mick Watson (Opiniomics) wrote a very nice post highlighting the savings in experimental costs that the system was likely to bring.
2-colour chemistry: Mick highlighted the move from the original Solexa/HiSeq 4 colour, to the NextSeq’s 2 colour chemistry. He pointed to a SeqAnswers thread where Brian Bushnell highlighted the lower quality sequence data. Illumina claim 75% of bases will be above Q30 for 2x150bp reads.
I’ve not followed up on this but would have expected to hear unhappy users if NovaSeq was not delivering the data quality needed. Big labs have run lots of samples on NovaSeq now so I’m reasonably confident we can sleep safely where %Q30 is concerned.
Barcode-swapping: The SP flowcell may allow smaller numbers of samples to be run but you’ll still need to use unique-at-both-ends barcodes – fortunately these are now available from Illumina and other providers.
Where’s the single-end kit? One thing that is still missing is a single end sequencing kit. I’ve been a long-time believer that for many DGE experiments the 40% lower costs of SE vs PE mean most users would be better getting more replicates and more reads than using paired-end reads. Something not everyone agrees with…but I’d like to see a robust discussion with Illumina.
See NovaSeq specs for more detail (table reproduced below)