During the setup of the run you can choose where you want to save the output files. Just click the magnifying glass symbol on the “Output location” box and select the destination folder.

Important note: MinKNOW stores raw signal files before they are converted into fast5 in the default partition where MinKNOW is installed. Ideally this has to be an internal SSD to avoid any data loss or corruption. This location is specified and can be changed in the minknow/conf/user_conf file however this would be at your own risk if the location is changed.

You can also change to a network drive through MinKNOW. On the “Output location” box select your network drive. This should work for both Windows and Ubuntu PCs.

In case this doesn’t work, the other alternative is to edit the user_conf file.

For Windows OS, you can access this file through C:\\Program Files\MinKnow\conf\user_conf. You need to change the value to the correct path to your network drive (if the path is wrong MinKNOW will error).
For Ubuntu systems, the user_conf is located in opt/ont/minknow/conf/user_conf. You can open it using a text editor (such as Notepad++) and you will need to edit the same value to the path of your network drive.

You should be aware that writing the output files directly to a network drive brings some risks. You will need a stable internet connection to do this. If the internet or network crashes or blips at all, this can lead to a run crash as MinKNOW will lose the output location. If that happens any raw/temporary data that MinKNOW is generating to move across to the network drive will be irretrievably lost. If you suffer an internet outage you will lose all of the raw data, resulting in no fast5 or fastq files generated at all. We always recommend customers write the data locally to “local drive://data” and then move the data off to a network drive etc. This allows for MinKNOW to dump the raw files to a folder locally if it suffers a run crash, which can be converted to fast5 and fastq files anytime. Internet blips or crashes will also not affect a run that is being written locally.