Assemble preprocessed shotgun metagenomic data, using Megahit, Spades or both! Results are automatically evaluated with Quast for contiguity. Output of this pipeline easily fits into our Metagenomic Binning pipeline. Single, paired-end, or paired-end with orphans sequencing data can be used.
To run assembly, first copy the assembly/config_assembly.yaml file to a working directory. Edit the config file to chose the assembler you want, the output directory, and the location of a sample_table which defines the mapping from sample names to sequence read files. This file must be tab-delimited with two columns- sample names in the first column, a comma-separated list of paths to sequencing reads in the second column. Comment lines are specified with the # character. An input table is automatically generated at the end of the preprocessing pipeline and can be found at 01_processing/assembly_input.txt. An example is below.
#Sample Reads1.fq[.gz][,Reads2.fq[.gz][,orphans.fq[.gz]]]
sample_a a_1.fq,a_2.fq,a_orphans.fq
sample_b b_1.fq,b_2.fq,b_orphans.fq
Launch this like any other pipeline, with arguments to submit jobs to the SCG scheduler and use singularity containers.
snakemake -s GITHUB/PATH/assembly/assembly.snakefile --configfile config_assembly.yaml --profile scg --jobs 99 --use-singularity --singularity-args '--bind /labs/,/oak/,/home/'
If you are satisfied with the assembly pipeline after examining the results, please run the cleanup command to remove unnecessary files and save space on SCG.
snakemake cleanup -s GITHUB/PATH/assembly/assembly.snakefile --configfile config_assembly.yaml
You can use this command, but might have to add arguments to submit to your SLURM scheduler, or singularity bind arguments as necessary. Configure the number of jobs and cores to fit your available resources.
snakemake -s ~/PATH/TO/assembly.snakefile --configfile config_assembly.yaml --use-singularity --jobs 8 --cores 8