ONT run stats - rrwick/Perfect-bacterial-genome-tutorial GitHub Wiki

This page shows some details about the ONT runs that generated the sample data.

This run was done on a previously used and washed flowcell. So while the flowcell didn't perform well by new-flowcell standards, it was more than enough for a single isolate with a 2.8 Mbp genome.

• 255545 reads
• 2.11 Gbp of total reads
• N50 read length of 22395 kbp
• Longest read with a single unbroken alignment: 172 kbp (d324a07f-60f5-436d-ba48-7503824176ae)

Here is the run's yield over time:

And here is a plot of the identity and length for the reads. Each dot is a read, and the marginal distributions are shown on the top and right sides:

A peak in the read-length distribution is visible between 20 and 21 kbp, corresponding to the 20.7 kbp plasmid in the genome. These are full-plasmid reads where the plasmid DNA was broken once (changed from circular to linear) and sequenced entirely.

This run was done using a new flowcell, so it generated more reads than the R9.4.1 run. However, since R10.4 pores run at a slower speed (~250 bp/sec) than R9.4.1 pores (~450 bp/sec), the yield is lower than a run would be on a new R9.4.1 flowcell.

• 1831719 reads
• 5.59 Gbp of total reads
• N50 read length of 4247 kbp
• Longest read with a single unbroken alignment: 140 kbp (fa31b3b1-1ddb-4cbc-819a-5a2883f1f514)

The N50 for the R10.4 run was relatively poor, but the high yield allowed for aggressive read QC in the tutorial to improve the post-QC N50.

Here is the run's yield over time:

And here is a plot of the identity and length for the reads. Each dot is a read, and the marginal distributions are shown on the top and right sides:

As was the case for the R9.4.1 run, the plasmid peak is present in this run's read-length distribution, though it's a bit harder to see here. There is also a small peak at 43 kbp, corresponding to a phage in this genome which was sometimes present in a separate circular form.