Creating UBA protein data file in shell

Question

[anabaena]

Hey all, I am trying to create a concatenated .faa file for analysis from the UBA genome data set, I downloaded the tar and unpacked it and a folder 'bacteria' was created with sub-folders labeled UBA1330, etc. and in those subfolders are .faa files. What I need to do is concatenate all the protein files into one major file and add the genome ID (e.g. UBA1330) to the >faa_id so I can locate that protein in the correct genome if it is a hit. I am new to shell and have done the following script

for GENOME in 'ls bacteria/';
do
sed "s|.*_|>${GENOME}_|" bacteria/${GENOME}/${GENOME}.faa | cat >> bacteria_proteins.faa; 
done

I recieve the following error:

sed: can't read bacteria/ls: No such file or directory

sed: can't read bacteria/ls: No such file or directory

sed: can't read bacteria.faa: No such file or directory

for some reason it isn't doing the 'ls bacteria/' command correctly and using bacteria as the {GENOME}, yet when I run:

$ls bacteria/

I get the correct output:

UBAXXXX UBAXXXXXX UBAXXXXXX etc.

I'm new to terminal and would love some input on what I am doing wrong. Thanks!

Answer 1

[lieven.sterck]

there is a syntax error in your script

this part

for GENOME in 'ls bacteria/';

needs to be:

for GENOME in `ls bacteria/`;

so using the backticks rather than normal single quotes

alternatively you could use the following:

for GENOME in $(ls bacteria/);

is you are in bash environment that is

what you originally had written was just a list of elements being 'ls' and 'bacteria' and what you need is that command to be executed and the result used in your loop.

[anabaena]

Awesome that worked, thanks!