nimagen_snakemake

A nimagen workflow implemented in snakemake.

Install

git clone git@github.com:TriassicSalamander/nimagen_snakemake.git
cd nimagen_snakemake
mamba env create -f environment.yml
mamba create --name pangolin --channel conda-forge pangolin
conda activate nimagen_snakemake

Creating Environments

It is recommended to use mamba (https://github.com/mamba-org/mamba) to create the environment, rather than conda.
Mamba can be installed by:

conda install mamba -n base -c conda-forge

Processing Waste Water Samples

Currently, whether or not freyja is used in the pipeline is controlled by commenting/uncommenting the 'Freyja Outputs' in 'rule all' in the Snakefile.
See comments in Snakefile for more detail.

Using autoNimagen

autoNimagen.sh is a wrapper which automatically executes the snakemake pipeline once the sequencing run is complete. This is useful when you know the output path of a given sequencing run, but don't know exactly when the run will finish.
Usage:

bash autoNimagen.sh -i <path/to/run/directory> -c <number of cores>

Pipeline Overview

Read Alignment

The pipeline begins by demultiplexing the samples using bcl-convert, generating fastqs.
These fastqs are then quality trimmed using TrimGalore.
Prinseq is used to trim bases with quality <30 from the 3' end.
The trimmed reads are aligned to a reference using bwa mem.
The primers are trimmed from the aligned reads using ivar trim.
The consensus is generated from the aligned reads using ivar consensus.

Config Notes

Input Paths

-sample_sheet: Path to sample sheet generated as part of the illumina sequencing run. Used by bcl-convert for demultiplexing and in the Snakefile for getting sample names. Normally found within the run directory.
-run_directory: Path to the output directory generated from the sequencing run. Contains multiplexed basecall files.
-fastq_directory: Path to directory which will contain demulitplexed fastq files. Demultiplexing is done as the first step in the pipeline.

Output Paths

-output_path: Output path which should contain other output directories.
-batch_dir: The name of your run. Recommended to be contained within output_path, e.g. output/batch_1 (assuming output_path is 'output').
-samples_dir: The name of the directory containing your samples. Recommended to be contained within batch_dir, e.g. output/batch_1/Samples
-summary_dir: The name of the directory containing your summary and statistics files. Recommended to be contained within batch_dir, e.g. output/batch_1/Summary

Other Parameters

-threads: Number of threads to use in rules where multithreading is possible. Deafult: 7.
-bcl-convert_threads: Number of threads to be used by bcl-convert for demultiplexing. Default 26.

Resources

-ref_genome: Path to reference genome used for alignment. Default: resources/ref/MN908947.3.fa
-primer_bed: Path to primer bed file. Default: resources/primers/nCoV-2019-NimaGen-V402.bed (for primer version 4.02).
-PE_primer_bed: Path to paired end primer bed file. Default: resources/primers/nCoV-2019-NimaGen-PE-V402.bed (for primer version 4.02).
-ambig_regions: Path to mask file specifying which regions to mask in consensus sequences. Format should be same as primer_bed. See resources/mask/primer_V3_ambig_regions for example. Default: resources/mask/blank_mask. Blank mask is used as default as ambiguous nucleotides was mainly an issue while V3 primers were being used.

Utilities

-scripts: Path to scripts used by pipeline. Default: utilities/scripts

Rulegraph