the new versions of MinKNOW allows to demultiplex with the option “requiring barcodes at both ends”, which is what we need for the artic pipeline.
However, recently not just the fastq files but also the fast5 files will be in subfolders devided by barcode.
e.g. path_to_fast5/barcode01/ .fast5; path_to_fast5/barcode02/ .fast5 etc.
In the artic pipeline in the following step:
artic minion --normalise 200 --threads 4 --scheme-directory ~/artic-ncov2019/primer_schemes --read-file run_name_barcode03.fastq --fast5-directory path_to_fast5 --sequencing-summary path_to_sequencing_summary.txt nCoV-2019/V3 samplename
Do we have to point to the fast5 folder or e.g. fast5/barcode01 folder. Or is there some kind of -r (recursive) option?
I don’t really get how nanopolish matches the fast5 and the fastq and I am worried polishing will fail if the path is not set the right way.