KnuttReads2Bins
This is the version of the pipeline as it was used for the paper. Some data outputs weren't used and have been disabled.
Setup
The memory requirement depends on the selected Kaiju database in the config.yml
file. If you don't have at least 170 GB available, replace "nr_euk" with "refseq" (44GB).
- Install the latest version of a conda distribution, sra-tools (SRA Toolkit) and Snakemake (>=5.10)
- Clone this repository (paper branch)
- Download the read data into the
input
folder - Run the
paper
rule with conda enabled
conda create -y -n snake -c bioconda -c conda-forge snakemake>=5.10 sra-tools
conda activate snake
git clone -b paper https://github.com/KnuttPipeline/KnuttReads2Bins.git
cd KnuttReads2Bins
download_sra(){
fastq-dump --split-files -O input $1 --gzip
mv input/${1}_1.fastq.gz input/${2}_R1.fastq.gz
mv input/${1}_2.fastq.gz input/${2}_R2.fastq.gz
}
download_sra SRR11128853 ENR-Ac244days_meta
download_sra SRR11128854 ENR-Ac211days_meta
download_sra SRR11128855 PFL9_meta
snakemake -prj 16 --use-conda paper 2>&1 | tee run.log
# Run with 16 cores available