January 25, 2018 / by Rafał Małanij

First MinION experiment - Pichia fermentans

Once we have completed Lambda phage experiment following Nanopore’s advice, we started to plan our own NGS experiment. Our fellow researchers are working on a few yest strains so we decide to seqeuence one of them - especially that no one has not done this before.

We received an isolated DNA and planned the experiment. First of all we have searched for instructions on Nanopore’s website. We got the latest one and we faced the first surprise - we had totaly different reagents, other than stated in instructions. It turned out that this PDF was for the different version of the kit, the newest one. Seems like Nanopore is releasing new versions very often (we received ours 2-3 weeks before the experiment), so you have to carefully check what revision of the kit you have.

Finding a correct version of the instructions was not a challenge so we were able to start. The process of preparing a library is not very complicated, especially if you have a helping hand with experience in lab. The overal process takes around 20-30 minutes and you are ready to install the flow cell into your device.

We have pluged-in our MinION into USB 3.0 of the laptop, performed some quality checks and started collecting reads. Our first impression was - why it is working so slow? Performance monitor on the laptop showed almost 100% of utilization. We have paused the experiment and pluged MinION into regular desktop PC. On regular PC the utilization was around 50% so the conclusion is that regular hardware is much preferred by MinION. We planned the experiment for 24h, so we have left the device in the lab.

Our checkpoint after 6 hours showed that something wrong is happening with the experiment - we have collected very few reads. Brief internal discussion supported by browsing Nanopore Community and we concluded there is no point in continuing this experiment. We have to figure out what happeneded. The best would be to ask Nanopore support.

Nanopore’s answer was suggesting that we had something wrong with DNA library - only 5% of the pores on the flow cell were filled with DNA strains while there should be at least 50% of them. In fact we should have seen that in the software interface - active pores are coloured in light green while non-active ones are dark green on the status graphics. In fact if you do not know what to search for you won’t notice this slight difference. It is a bit puzzling that if the software is aware of this rate (active/non-active) why it does not inform user that the experiment has no chance to succeed. You may lose a day to find out that your library prep was not ideal, like we did.

We plan to repeat this experiment, but we have to work on the DNA first. For sure we will write it on our blog, so stay tuned!