Updated Illumina Specs Table

[AllSeq]

We've updated the AllSeq NGS Knowledge Bank to reflect the latest Illumina instruments - all the specs in one place!

[ECO]

Moving to Vendor Forum

[SNPsaurus]

Can the 2500 Rapid mode really produce 600M clusters in the two 'lanes'? I've seen results more along 150-180M clusters. Maybe I'm thinking of passing filters and this table is not, but I don't think I've seen the lanes loaded to make 300M clusters each either.

[AllSeq]

Can the 2500 Rapid mode really produce 600M clusters in the two 'lanes'? I've seen results more along 150-180M clusters. Maybe I'm thinking of passing filters and this table is not, but I don't think I've seen the lanes loaded to make 300M clusters each either.

I can't say for sure, but that's the spec for two flow cells (and the number of reads needed to generate the claimed 180Gb per run). Those could be pre-filtered reads, but I wouldn't think filtering would cut the # of reads by half. Hopefully others can chime in with their real world results.

[SNPsaurus]

Ahh, 600M X 300 bp = 180 Gb, so that would be 150M clusters per lane x 2 reads x 150 bp. I think you are counting clusters in some columns and total reads possible (2 per cluster) in others.

[kmcarr]

Can the 2500 Rapid mode really produce 600M clusters in the two 'lanes'? I've seen results more along 150-180M clusters. Maybe I'm thinking of passing filters and this table is not, but I don't think I've seen the lanes loaded to make 300M clusters each either.

Remember that Illumina stats are always in Run output assuming you are running with as many flow cells as possible. A HiSeq 2500 can output 600M clusters per Run which means (150M clusters per lane) X (2 lanes per flow cell) X (2 flow cells per run).

[AllSeq]

Ahh, 600M X 300 bp = 180 Gb, so that would be 150M clusters per lane x 2 reads x 150 bp. I think you are counting clusters in some columns and total reads possible (2 per cluster) in others.

In all cases the stated outputs should be for entire runs (both flow cells for two flow cell machines). All reads -should- be single reads/clusters. If you find examples where we screwed that up, please let me know and we'll fix it ASAP.

For the HiSeq 2500 Rapid Mode we have:
2 flow cells X 300M reads/flow cell X 2x150bp reads
2*300M*2*150b=180Gb

[SNPsaurus]

Right, sorry for the confusion. The single flow cell NextSeq got me messed up going back to the two-flow-cell HiSeq.

[AllSeq]

Right, sorry for the confusion. The single flow cell NextSeq got me messed up going back to the two-flow-cell HiSeq.

I agree that it gets kind of confusing - that's why we built the table