Skip to content

1. How to run

Jump to bottom
Marit Hetland edited this page Mar 19, 2021 · 15 revisions

Table of contents

Basic usage

python susCovONT.py --input_dir /path/to/<run_name> --sample_names /path/to/sample_names.csv

Where:

  • --input_dir: Input directory <run_name> must contain fast5_pass and fastq_pass folders and sequencing_summary*.txt, with the <run_name> corresponding to your run (e.g. 20210213_1359_X5_FAO88697_5cf6e6f0)
  • --sample_names: A CSV-file which connects barcodes with sample names, following the format:
barcode,sample_name
barcode01,NEGCONTROL
barcode02,E1234567_P1
NB03,V2345678_P1

The barcode column can take values following the format barcode[0-9][0-9] or NB[0-9][0-9] (as in the example above), and the sample_name column can be anything you'd like.

This pipeline takes as input a folder with name <run_name> which contains the folders fast5_pass and fastq_pass and sequencing_summary*.txt from Sars-CoV-2 ONT sequencing together with a CSV-file which links barcode and sample name, and it outputs consensus.fasta files along with <run_name>_report.csv which includes pangolin lineage, nextstrain clade, mutations and QC.

This will output a <run_name>_report.csv which includes pangolin lineage, nextclade strain and QC_status for each sample, and a directory 003_consensusFasta with consensus.fasta files for each sample both separate and as a concatenated file.

Configuration

Make sure to update the config file scripts/config.cfg in this repository before running the script so it works on your computer. If you installed dependencies with scripts/install.sh this file is already updated.

Usage

usage: susCovONT.py [-h] [-v] -i INPUT_DIR -s SAMPLE_NAMES
                    [-b {r9.4_fast,r9.4_hac,r10_fast,r10_hac}]
                    [-k {none,native_1-12,native_13-24,native_1-24,native_1-96}]
                    [--guppy_resume_basecalling] [--guppy_use_cpu] [--cpu CPU]
                    [--generate_report_only] [--offline] [--no_move_files]
                    [--no_artic] [--dry_run] [--seq_sum_file SEQ_SUM_FILE]

susCovONT

optional arguments:
  -h, --help            show this help message and exit
  -v, --version         show program's version number and exit

Input options (required):
  -i INPUT_DIR, --input_dir INPUT_DIR
                        Input directory, which should contain "fast5_pass"
                        directory fast5-files.
  -s SAMPLE_NAMES, --sample_names SAMPLE_NAMES
                        Provide a comma-separated list showing which barcode
                        corresponds to which sample (for final report)

Necessary flags only if performing guppy basecalling/demultiplexing:
  -b {r9.4_fast,r9.4_hac,r10_fast,r10_hac}, --basecalling_model {r9.4_fast,r9.4_hac,r10_fast,r10_hac}
                        Use flag to perform basecalling before running the
                        artic pipeline. Indicate which basecalling mode to
                        use. In most cases you want to use a HAC option.
  -k {none,native_1-12,native_13-24,native_1-24,native_1-96}, --barcode_kit {none,native_1-12,native_13-24,native_1-24,native_1-96}
                        Use flag to perform demultiplexing of basecalled data.
                        Indicate which barcode-kits were used (or none).

Options for basecalling command:
  --guppy_resume_basecalling
                        This flag can be used with --basecalling to resume an
                        interrupted basecalling run. Default: off.
  --guppy_use_cpu       This flag can be used with --basecalling to run on CPU
                        instead of GPU. Will use 4 threads and 6 callers.
                        Default: GPU -auto x.

Advanced options:
  --cpu CPU             Specify cpus to use. Default: 20
  --generate_report_only
                        Do not run any tools, just (re)generate output report
                        from already completed run. Default: off.
  --offline             The script downloads the newest primer schemes,
                        nextclade and pangolin each time it runs. Use this
                        flag if you want to run offline with already installed
                        versions.fault: off.
  --no_move_files       By default, the input fast5_pass and fastq_pass dirs
                        will be moved to subdir 001_rawData. Use this flag if
                        you do not want that
  --no_artic            Use this flag to run only pangolin and nextclade on an
                        already completed artic nextflow (with same folder
                        structure)
  --dry_run             Executes nothing. Prints the commands that would have
                        been run in a non-dry run.
  --seq_sum_file SEQ_SUM_FILE
                        If the pipeline does not find the sequence summary
                        file, you can specify it. Generally not needed.

Advanced options

Offline mode

By default, this pipeline will download the newest versions of pangolin, nextclade and primer schemes each time it is run. If you do not have an internet connection or want to skip this step, use the --offline flag to use existing versions of tools.

Basecalling and demultiplexing

  • There is an option to perform basecalling and demultiplexing in this script. However, this should ideally be performed on the GridION.
  • It is very important that basecalling and demultiplexing is performed as described in the artic protocol.
  • You should check the PDF report from GridION to see that you have used the correct settings. If you have not, either re-do the analysis on the GridION, or use this script with the flags --basecall and --demultiplex

Dry run

If you want to simply see what the commands would look like, use the --dry_run flag. This can be useful if you want to make sure your input is correct before starting the pipe or if you want to run any of the commands individually.

Only generate the final report

If you want to regnerate the <run_name>_report.csv file (and still have the same folder structure) you can do that by running the pipeline with flag --generate_report_only

Skip artic

Use the flag --no_artic if you want to re-run pangolin and nextclade and update the <run_name>_report.csv.

Don't move input folders

By default, the input folders fast5_pass and fastq_pass folders are moved to a new subdirectory 001_rawData. You can stop this from happening by using the --no_move_files flag.

Clone this wiki locally