Looping script can save us time

Why we need looping script?

If we have 1000 genome files:

Genome_1.fna 
Genome_2.fna
...
Genome_999.fna
Genome_1000.fna

We want to use Mash to identify the species, and save the output file:

mash screen -w -p 8 RefSeqSketches.msh genome_1.fna > genome_1_mash.tab
mash screen -w -p 8 RefSeqSketches.msh genome_2.fna > genome_2_mash.tab
...
mash screen -w -p 8 RefSeqSketches.msh genome_999.fna > genome_999_mash.tab
mash screen -w -p 8 RefSeqSketches.msh genome_1000.fna > genome_1000_mash.tab

Do we type the command 1000 times?

Maybe wild cards?

mash screen -w -p 8 RefSeqSketches.msh genome_*.fna > genome_*mash.tab

there will be an error:

-bash: *.tab: ambiguous redirect

Loop in shell script

The most typical loop used in a shell script is the for loop. The for loop allows you to iterate over a sequence of items, such as a list of files or directories, and perform a certain action on each item in the sequence.

for item in list
do
    # commands to execute on each item
done

Use loop script in Mash

In current directory, I want to iterate over all the genome files with the same command:

for i in *.fna
do 
mash screen -w -p 8 RefSeqSketches.msh $i > $i.mash.tab
done

add the sorting command to our loop:

for i in *.fna
do
mash screen -w -p 8 RefSeqSketches.msh $i > $i.mash.tab
sort -gr $i.mash.tab > $i.mash_sorted.tab
done

Not confident about your commands?

We can use echo " " to print the command first:

for i in *.fna
do 
echo "mash screen -w -p 8 RefSeqSketches.msh $i > $i.mash.tab"
echo "sort -gr $i.mash.tab > $i.mash_sorted.tab"
done

Copy and paste the commands into a txt file, name it with '.sh', so system will recognize it as a shell script file.

And then execute it:

bash script.sh

Reference

https://github.com/marbl/Mash