A snakemake-wrapper for creating de novo bacterial genome assemblies from paired-end Illumina sequencing data.
Currently included programs:
# Install
git clone https://github.com/pmenzel/illumina-assembly-snake.git
conda config --add channels bioconda
conda env create -n illumina-assembly-snake --file illumina-assembly-snake/environment.yaml
source activate illumina-assembly-snake
# Prepare data
mkdir fastq
cp /data/my_sample/reads_R1.fastq.gz fastq/mysample_R1.fastq.gz
cp /data/my_sample/reads_R2.fastq.gz fastq/mysample_R2.fastq.gz
# Declare desired assemblies and run workflow
mkdir -p assemblies/mysample_shovill
mkdir -p assemblies/mysample_spades
snakemake -s illumina-assembly-snake/Snakefile --cores 10
Clone repository, for example:
git clone https://github.com/pmenzel/illumina-assembly-snake.git /opt/software/illumina-assembly-snake
Install conda and then create a new environment containing all the programs:
conda config --add channels bioconda
conda env create -n illumina-assembly-snake --file /opt/software/illumina-assembly-snake/environment.yaml
and activate environment:
source activate illumina-assembly-snake
First, prepare a folder called fastq containing the sequencing reads as
two fastq files per sample, called samplename_R1.fastq.gz and samplename_R2.fastq.gz.
Next, create a folder assemblies and inside create empty folders specifying
the desired assemblies.
Sample name and assembler name need to be separated by an underscore. NB: This also means that sample names must not contain underscores.
This example contains two samples.
Sample 1 should be assembly with shovill, whereas sample2 should be assembly by both shovill and spades.
.
├── fastq
│ ├── sample1_R1.fastq.gz
│ ├── sample1_R2.fastq.gz
│ ├── sample2_R1.fastq.gz
│ └── sample2_R2.fastq.gz
│
└── assemblies
├── sample1_shovill
├── sample2_shovill
└── sample2_spades
Run workflow in that folder, e.g. with 10 threads:
snakemake -k -s /opt/software/illumina-assembly-snake/Snakefile --cores 10
The assemblies are contained in the files output.fa in each folder.