A first look at Illumina’s new NextSeq 500

[AllSeq]

AllSeq was recently at Illumina HQ to get a sneak peek at their new NextSeq 500. We’ve already listed all of the specs and given our opinion on how we think this new platform fits into Illumina’s lineup and the broader market, but this was our first chance to see it in action. Check out our blog to see how it’s at once both tiny and huge.

[williamhorne]

I bought the first two that were sold by Illumina including Basespace onsite. If interested in details let me know!

[AllSeq]

We'd love to hear how it's working for you - actual outputs, how the data looks compared with previous Illumina machines. Basically anything you're willing to share! I'm sure lots of people are interested.

[williamhorne]

Using High output we are actually getting over 500 million reads per run. Unlike our GAII, and HighSeq, we actually have to pay very close attention to cluster density. The target cluster density for high quality samples is 1.75pM-2pM. Anything above and below will results in under/over clustering. So your samples need to be very exact with concentration.

These are solely made to be streamlined with the BaseSpace. Right now it only works with BaseSpace onsite, not in the cloud as they are having some majority broker issues that still are not resolved. Make sure you do your research in regards to output files and data in regards to basespace because it is not a visual machine. It gives you the output files and you must use 3rd party software on a different computer to view the results. Very annoying.

Overall very impressed with the NextSeq's, not so much BaseSapce.

[bryanbriney]

We've had much the same experience as williamhorne. High Output produced roughly 500M reads, total output from a PE150 run was just shy of 150GB. We clustered at the high end of the recommended range, but still had about 80% Q30.

We've used the NextSeq with BaseSpace and it has some quirks. Run setup and sample entry is awkward if you have more than a handful of samples, although there's an option to upload an Excel file with sample info that we haven't tried yet. Never thought I'd say it, but the new run setup makes me miss Sample Sheets. On the other hand, once the run was going, BaseSpace works wonderfully.

As of right now, the NextSeq doesn't support dual indexes or custom indexes. Dual indexing is reportedly in the pipeline, not sure about custom indexes. Also, you can't start a run that exceeds the stated capacity of the reagent kit. MiSeq throws an error if you try, but it can be bypassed; NextSeq won't let you continue.

Overall, we're very happy. If read lengths get a little longer (come on, PE250!) and the BaseSpace quirks are fixed, the NextSeq will be darn close to perfect.

[GenoMax]

Can someone here comment on the actual alignability/usability of the resulting data?

[bryanbriney]

We're not doing any sort of analysis that involves alignment to a reference, but data quality from our PE150 NextSeq runs has been essentially identical to PE150 Rapid Runs on HiSeq.

[TonyBrooks]

As of right now, the NextSeq doesn't support dual indexes or custom indexes. Dual indexing is reportedly in the pipeline, not sure about custom indexes. Also, you can't start a run that exceeds the stated capacity of the reagent kit. MiSeq throws an error if you try, but it can be bypassed; NextSeq won't let you continue.

Does this mean you can't do more than 75 cycles in a 75 cycle kit. I confirmed with tech support that there are 25 additional cycles for dual indexing and we were told we could use those 100 cycles as we wished. We were hoping to do 47|6|47 from the 75 cycle kit

[bryanbriney]

Does this mean you can't do more than 75 cycles in a 75 cycle kit. I confirmed with tech support that there are 25 additional cycles for dual indexing and we were told we could use those 100 cycles as we wished. We were hoping to do 47|6|47 from the 75 cycle kit

I've only tried with a 300 cycle kit and the maximum number of cycles that I could perform was 308, apparently since the earliest kits only support single indexing. The error that we couldn't get past was related to the total number of cycles, and it appears that you can use those cycles however you wish -- I tried designing a run with a single 308 cycle read, and that didn't raise an error (although I didn't actually do the run).

[TonyBrooks]

I can confirm that the most you can get out of a 75 cycle kit is currently 92 cycles (76|8|8). This means you can't use the dark cycles for sequencing like you can on the MiSeq. You can register the run in BaseSpace but you get an error message when you insert the cartridge and you can't bypass it.

[Brian Bushnell]

Do any of you have NextSeq data for something common (phiX, e.coli, mouse, etc) that you would be willing to share? Our NextSeq has consistently produced data of far lower quality than our HiSeq/MiSeq machines, and I'm trying to determine whether this is specific to the individual machine or not.

[nucacidhunter]

Following data are publically available in BaseSpace:

NextSeq 500: TruSeq PCR Free WGS_RTA2.1.3.0 (NA12878)
NextSeq 500: TruSeq Nano 2x151 (PhiX)
NextSeq 500: RNA-Seq (8plex)
NextSeq 500: TruSight One (CEPH Trio replicates)

[Brian Bushnell]

I already looked at this one: "NextSeq 500: TruSeq Nano 2x151 (PhiX)"
...and it's just as bad as ours. But thanks for the suggestion; I'll take a look at the others, as they may have used a different machine. Still, I'm kind of hoping for data from e.g. bryanbriney, as his machine seems to be producing data on-par with HiSeq machines.

[GenoMax]

@nucacidhunder: All those appear to be "standard" (gold?) samples.

Brian: If PhiX standard does not look good then that is worrisome.

[nucacidhunter]

@nucacidhunder: All those appear to be "standard" (gold?) samples.

You are right. I have access to data from various libraries run on both HiSeq and NextSeq, but unfortunately I am not allowed to discuss it externally.