| tags |
|---|
ggg, ggg2024, ggg298 |
[toc]
Use the instructions here. It will be helpful to have an editor available, so RStudio Server is recommended.
For the srun command, use:
srun -p high2 --time=3:00:00 --nodes=1 --cpus-per-task 4 \
--mem 5GB --pty /bin/bash
which asks for 4 CPUs. We'll use these later!
Set things up:
cd ~/ggg298-lab-6
cp ~ctbrown/data/sulfo/* .
Load software:
module load mamba
mamba activate automation
(you may need to create the mamba environment first)
and put the following in a file in that directory named run-sourmash.sh:
set -e
set -x
for genome in *.fa.gz
do
name=$(basename $genome .fa.gz)
sourmash sketch dna $genome --name $name -o $name.sig.zip
done
sourmash compare *.sig.zip -o sulfo.cmp
sourmash plot sulfo.cmp
Let's go through this line by line.
The set -x says, print out each line as it's executed.
The set -e says, if any command fails, exit from the shell script.
The for loop says, run the command between do and done on every .fa.gz file in the directory.
The name= line removes the .fa.gz from the genome filename.
The sourmash sketch line runs sourmash to sketch the genome into a file it can use for later comparisons.
The sourmash compare command runs on all of the .sig.zip files and produces a comparison matrix.
The sourmash plot command generates a plot from the comparison.
What input files need to exist for each command? Let's start from the end.
What output file do we want at the very end?
What file(s) does sourmash plot need exist, in order to run?
What file(s) does sourmash compare need to exist, in order to run?
What file(s) does sourmash sketch need to exist, in order to run?
(whiteboarding)
input and output here are annotations that snakemake uses. They provide information about what the shell command is doing. They do not tell the shell command what to do!