After unboxing the Nanopore alpha-beta PromethION device some weeks ago and having our IT support/linux ninja help us integrate yet another box into the university network, we finally got around to run some PromethION sequencing again. We updated the PromethION software to Minknow 2 inserted a flowcell and it passed QC with 7000+ pores available even though it is more than 3 months ago it arrived in our lab. My colleague Rasmus Wollenberg prepared a sequencing library of three different metagenomic samples using a mixture of Oxford Nanopore protocols as barcoding is currently not officially supported on the PromethION.
— Rasmus Kirkegaard (@kirk3gaard) June 18, 2018
He primed the flowcell twice as described in the protocol and we can confirm that we are looking very much forward to the possibility to use single channel pipettes for loading flowcells in the future. Aligning the 4 tips and inserting them into the sample inlets is unnecessary tricky. Then finally he loaded his first PromethION flowcell with the sequencing library and we started the sequencing run.
We were extremely happy to see that more than 1800 pores turned the right shade of green and started sequencing!!! We followed the run closely and it kept going at a steady rate getting closer and closer to 100 Gbp of estimated bases for a single flowcell. However, it turned out that live basecalling was not working for this release of the software so we had no idea how much data would actually end up being basecalled. Oxford Nanopore released a software fix to provide live basecalling so we stopped the run, updated the software and restarted the sequencing run to squeeze whatever was left out of the flowcell and started basecalling.
It sure helps to have the 4 tesla GPUs in the basecalling unit running to translate the raw signal into basecalled data. After basecalling we now have
107 Gbp from a single PromethION flowcell
(Median read length 6473, Median read quality: 9.2)
And we received an email from ONT:
“We believe you may be the first customer to have hit 100 Gbases from a single flow cell!”
It came as quite a surprise that no one else managed that yet* so we of course got a photo with the first flowcell to break the 100 Gbp barrier outside their labs! This flowcell was 3+ months old so I hope to see more people crushing this with new flowells soon. Now we just have to wait for miniasm to provide us with some evidence that the data can be stitched together to form some nice circular bacterial genomes and whatever structure the eukaryotic genomes might have before we do some hybrid assemblies.
*Please correct us if you know of others who have managed this before us 😉