NovaSeq Experience

[kcchan]

Originally Posted by Markiyan

This is really important for RNAseq & CHIPseq applications, where we want to keep the cross-talk bellow 10^-6...

How did you come up with that 10^-6 figure? In an RNASeq or CHIPseq experiment that's only about 20-30 reads.

[Brian Bushnell]

Originally Posted by pmiguel

Did you see much phiX index hopping? Searching the "undetermined" fastq for index hops involving one of i7/i5 GGGGGGGG/TCGTAGTG I tentatively identified a fair number. Of course they may just be the result of signal bleed.

--
Phillip

I have not looked into that yet. Actually, I don't even know if we are spiking PhiX into our Novaseq runs, but that rate is worth examining, after I find out whether there is actually any PhiX present.

[luc]

I am not sure that index hopping onto PhiX reads would be a good measure. As far as I remember the Illumina PhiX adapter sequences are quite different from TruSeq or Nextera sequences.

[pmiguel]

Originally Posted by luc

I am not sure that index hopping onto PhiX reads would be a good measure. As far as I remember the Illumina PhiX adapter sequences are quite different from TruSeq or Nextera sequences.

Well, not as different as you might think. Please see the attachment.

--
Phillip

[austinso]

For those with NovaSeqs, are you allowed to say how much the consumables are? Based on what people are charging for a lane, it looks like about $40K per run...

[gringer]

I was told by an Illumina rep at Queenstown Research week about a month ago that NovaSeq run cost would be similar to HiSeq, but produce about double the output.

[misterc]

List price on S4 reagent kits is about $30K. For NovaSeq sites with ≥5 NovaSeq instruments, they're able to get a discount large enough to achieve the "20% cheaper than HiSeq X" statement Francis deSouza made when they announced this platform. Something tells me this is factoring in how many samples could be run per year on NovaSeq than HiSeq X due to the throughput and runtime, so their discount is likely not more than 20% off of list. Everyone else likely gets <10% discount. HiSeq 4000 (300 cycle) kits are about $17K list price, so they're not the same price but S4 will give you about 5x the output (in about half the time).