MiSeq i100 Is Fast... And Confusing.
The iSeq was always… surprisingly slow. With a minimal run time of 9.5 hours, this was significantly slower than the fastest MiSeq (Classic) run time of 5.5 hours.
This always seemed weird… because the cycle time was pretty impressive…
Cycle Time
The cycle time is the time it takes to readout a single base. If you’re doing a 150bp read, that means 150 cycles.
In spite of its 1 color chemistry the iSeq still had the fastest cycle time of any Illumina instrument.
You might therefore expect the new MiSeq i100 to have an even faster cycle time, as it uses the same CMOS technology, but no longer needs that intermediate chemistry step.
Let’s try and confirm that.
The Illumina Knowledge base lists the iSeq has having a 2.2m cycle time. But let’s confirm that using the official specs.
A 2x75bp iSeq run takes 14 hours, 2x150bp, 19 hours. Running the numbers gives us ~2 minutes, so in the right ballpark.1
We can then do the same for the MiSeq i100. A 2x150bp run takes 7 hours, 2x300, 15 hours. That’s a 1.6 minute cycle time2.
Cycle time has therefore decreased. But this isn’t enough to explain the massive speed increases on the MiSeq i100. An iSeq 2x150bp run took 19 hours, on the MiSeq i100 it only takes 7!
A 20% decrease in cycle time isn’t enough to explain the difference.
Cluster Generation Times
As previously noted, Illumina cluster generation times have been going crazy. The iSeq 100 sits around for 5 hours generating clusters.
They seem to have fixed this issue on the MiSeq i100. If we assume 1.6 minute cycle time we can figure out the “non-cycle” overhead for various MiSeq i100 runs:
That’s right… cluster generation time is negative 1 hour.
Wait… that can’t be right.