Skip to content

6. Manual run

Jump to bottom
Marit Hetland edited this page Feb 19, 2021 · 6 revisions

Running it all manually

If for some reason the pipeline is not working or you want to run any of the steps manually, you can do so with:

Assuming your fast5 reads are in a directory which includes the folder fast5_pass:

  1. Basecalling
guppy_basecaller -c dna_r9.4.1_450bps_hac.cfg -i fast5_pass -s fastq_pass_notDemultiplexed -x auto -r
  1. Demultiplexing Make sure to specify the correct barcodes, in the example below is for 1-24.
guppy_barcoder --require_barcodes_both_ends -i fastq_pass_notDemultiplexed -s fastq_pass --arrangements_files "barcode_arrs_nb12.cfg barcode_arrs_nb24.cfg"

Artic

When you have your fast5_pass and demultiplexed fastq_pass directories, you can run artic guppyplex and artic minion. This can be done on one sample at a time on its own (steps 3 and 4) or you can input a whole run folder to run both guppyplex and minion in one go (step 5).

  1. Read filtering
conda activate artic
artic --version
artic guppyplex --min-length 400 --max-length 700 --directory fastq_pass/barcode03 --prefix 002_guppyplex/guppyplex
conda deactivate

This will perform a quality check. If you are only using “pass” reads you can speed up the process with:

conda activate artic
artic --version
artic guppyplex --skip-quality-check --min-length 400 --max-length 700 --directory fastq_pass/barcode03 --prefix 002_guppyplex/guppyplex
conda deactivate

(Note: The code and text in steps 1-3 are from https://artic.network/ncov-2019/ncov2019-bioinformatics-sop.html)

  1. Artic minion pipeline
conda activate artic
artic minion --normalise 500 --threads 4 --scheme-directory ~/Programs/artic-ncov2019/primer_schemes --read-file barcode03.fastq --fast5-directory fast5_pass --sequencing-summary sequencing_summary.txt nCoV-2019/V3 samplename
conda deactivate

Artic via nextflow pipeline

  1. Nextflow pipeline

The read filtering and the artic minion pipeline (steps 3 and 4) have to be run separately for each barcode. Matt Bull at PHW has created a nextflow pipeline which can take a whole sequencing run as input and runs these steps (3 and 4) on the entire run, leaving you with a consensus.fasta file for each barcode, and a qc.csv report.

nextflow run ~/Programs/ncov2019-artic-nf/  -profile conda --nanopolish --prefix 2021010124_1282_X4_FAO88984_3275a195 --basecalled_fastq ./fastq_pass/ --fast5_pass ./fast5_pass/ --sequencing_summary ./fastq_pass_notDemultiplexed/sequencing_summary.txt --outdir ./002_articPipeline/ #--cache ~/Programs/conda_for_covid/work/conda/

Pangolin lineage assignment

When you have your consensus.fasta file(s), you can run it/them through pangloin for lineage-assignment:

conda activate pangolin
pangolin --update
pangolin consensus_sequences.fasta
conda deactivate pangolin

Nextclade

https://hub.docker.com/r/nextstrain/nextclade

docker pull nextstrain/nextclade
docker run -it --rm -u 1001 --volume="{PWD}:/seq" neherlab/nextclade nextclade --input-fasta '/seq/sequences.fasta' --output-json '/seq/results.json'
Clone this wiki locally