The default MinKNOW configuration is suboptimal in detecting reads originating from short molecules such as tRNA. We optimised the configuration to capture ~10x more reads from tRNA experiments. Below, you can see the reads (in red) identified by default (top) and our custom (bottom) configuration.
Here we describe how to set up alternative MinKNOW configurations for enhanced capture of tRNA reads from direct RNA nanopore sequencing runs, which are otherwise discarded when using default MinKNOW settings.
Note, this tutorial is meant for Linux users and was tested with Ubuntu 20.04 and 22.04.
For MacOS, you should replace all MinKNOW paths
/opt/ont/minknow/
with
/Applications/MinKNOW.app/Contents/Resources/
(see #3).
To enable those, simply rsync conf files (root priviledges needed):
rsync -a conf/package/flow_cells.toml /opt/ont/minknow/conf/package
rsync -a conf/package/sequencing/*.toml /opt/ont/minknow/conf/package/sequencing
After, reloading scripts in MinKNOW
you should see alternative configurations in the flowcell type field
You can now run MinKNOW with alternative configurations (named "FLO-MIN106_short"), which will allow you to capture short tRNAs during the sequencing run, without having to save the bulk FAST5 file.
Everytime the scripts are reloaded, MinKNOW reads configuration of:
flowcells.toml
--> from/opt/ont/minknow/conf/package/flow_cells.toml
sequencing*.toml
--> from/opt/ont/minknow/conf/package/sequencing/sequencing_*.toml
Depending which flowcell is selected by the user and the experiment type, MinKNOW will use a matching configuration file. For example, if the user selects FLO-MIN106 and direct RNA sequencing kit, MinKNOW will match it to sequencing_MIN106_RNA.toml.
The toml files that have been made available were generated for MinKNOW version defined in the release information.
You may need to tweak the toml
files to make them suitable to your MinKNOW versions.
Note, currently the custom configurations can be used only with sequencing simulations. If you try to use them with real sequencing, it'll likely fail during QC step. In order to start simulation from bulk file, you'll need to:
- run your sequencing saving bulk file
Here, make sure you have lots of free space. Bulk file for 72h dRNAseq run from MinION may reach 250-300 GB. Of course, you'll need additional space for the Fast5 files.
- specify bulk file in the configuration file under
custom_settings
Thecustom_settings
part of your configuration file/opt/ont/minknow/conf/package/sequencing/sequencing_MIN106_RNA_short.toml
should look something like this:
###############################
# Sequencing Feature Settings #
###############################
# basic_settings #
[custom_settings]
enable_relative_unblock_voltage = true
unblock_voltage_gap = 480
run_time = 7200 #172800 # (seconds) 1hr=3600
start_bias_voltage = -180
# UI parameters
translocation_speed_min = 50
translocation_speed_max = 75
q_score_min = 7
simulation="/path/to/bulk_file.fast5"
- and start sequencing simulation using
FLO-MIN106_short
flowcell (alternative cofiguration).
In order to run simulation, you should connect a MinION device with a flowcell and select this device for sequencing (simulation). You can use a dummy flowcell that was provided with the device (the one you use for hardware testing).