Adding parallel capabilities to a FOR task - parallel-processing

Question

I want to execute the following command:
for i in {0-months,3-months,6-months,9-months,12-months,EC1,EC2_CZ,EC2,EC3}; do freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/${i}.vcf ${i}.sort.grp.bam; done
But these task are independent of each other, and can be run in parallel. I was wondering if there is a way to do this with gnu parallel.
Usually when using parallel, I would have a file listing all commands needed to run, in this case it would look like this:
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/0-months.vcf 0-months.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/3-months.vcf 3-months.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/6-months.vcf 6-months.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/9-months.vcf 9-months.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/12-months.vcf 12-months.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/EC1.vcf EC1.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/EC2_CZ.vcf EC2_CZ.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/EC2.vcf EC2.sort.grp.bam
freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/EC3.vcf EC3.sort.grp.bam
So when the file is ready, I could simply run:
parallel -j 4 -a FILE freebayes
But that requires writing the commands out to a file, then invoking parallel, there must be a simpler way.
This seems to work:
parallel -j 4 -a \
<(for i in {0-months,3-months,6-months,9-months,12-months,EC1,EC2_CZ,EC2,EC3}; do echo "freebayes --fasta-reference ../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous --min-coverage 10 -F 0.01 -C 2 --vcf vcf/${i}.vcf ${i}.sort.grp.bam"; done)
freebayes
But it just looks silly... any easier way to do this?
Thanks!

Answer 1

I am really puzzled how you came up with your extremely convoluted (but working) way to do this:
parallel -j 4 freebayes --fasta-reference \
../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous \
--min-coverage 10 -F 0.01 -C 2 --vcf vcf/{}.vcf {}.sort.grp.bam \
::: 0-months 3-months 6-months 9-months 12-months EC1 EC2_CZ EC2 EC3
If these are all the vcf files in the vcf-dir and it is a 4-core machine, you can even do:
parallel freebayes --fasta-reference \
../Genome/ECIII_Lemming_assembly_masked.fasta --pooled-continuous \
--min-coverage 10 -F 0.01 -C 2 --vcf {} {/.}.sort.grp.bam \
::: vcf/*.vcf
Have you walked through the tutorial? man parallel_tutorial
Have you watched the intro videos: https://www.youtube.com/playlist?list=PL284C9FF2488BC6D1
Have you looked at the examples: LESS=+/EXAMPLE: man parallel