How to use gnu-parallel for processing a script with two inputs?

This is not really a Python question, it's a question about GNU parallel. You could try this if all files are prefixed with "KOG_":

seq 1 300 | parallel python stable.py KOG_{}.fan KOG_{}.fasta ">" KOG_{.}.stable

The quotes around the redirect (">") are important, unless you want all of the output in one file.

To handle generic prefixes:

ls *fan *fasta | parallel ---max-lines=2 python stable.py {1} {2} ">" {1.}.stable

This uses the -max-lines option to take 2 lines per command. Of course this works only if the *.fan and *.fasta files match up, i.e. there must be the same number of each, and the numbers need to match up, otherwise you'll end up pairing files that shouldn't be paired. If that is a problem, you can figure out a command that will more robustly feed pairs to parallel.

Try:

parallel python stable.py {} {.}.fasta '>' {.}.stable ::: *fan

I recommend you split this task in two steps:

1. Create a jobs file containing all commands you want to run with parallel. You need to create a text file jobs.txt that should be similar to the one presented bellow:

   python stable.py KOG_1.fan KOG_1.fasta > KOG_1.stable
   python stable.py KOG_2.fan KOG_2.fasta > KOG_2.stable
   python stable.py KOG_3.fan KOG_3.fasta > KOG_3.stable
   python stable.py KOG_4.fan KOG_4.fasta > KOG_4.stable
   ...
   python stable.py KOG_300.fan KOG_300.fasta > KOG_300.stable
   

   If all your files are prefixed with KOG, you can build up this file this way:

   for I in `seq 300`; do echo "python stable.py KOG_$I.fan KOG_$I.fasta > KOG_$I.stable" >> jobs.txt; done;
   

2. Run parallel using the jobs file

   Once you have the jobs file, you just need to run the following command:

   parallel -j4 < jobs.txt    
   

   Note that -j4 indicates that at most 4 commands from your jobs file will be running in parallel. You can adjust that according to the number of cores available on your computer.