Naming of FastA sequences

Question

[Phil S.]

Hi there,

I have a usual looking fastA file like this:

>Translation: 2..112 (direct), 37 amino acids
ADTAQEFISTAVFGTSMSAHHILGLKPVPRVWLFAI*

>Translation: 1482..1790 (direct), 103 amino acids
MKKYTEQAKLSVVEDYCSGSAGHREVAHRHGVNANVIRKWLPIYRDKPVAPLPAFVPLQP
MPKRQADEAVVIALSLGDKSITVKWPISDPDGCARFIRSLSQ*
 
>Translation: 1787..2122 (direct), 112 amino acids
MIRIDAIWLATEPMDMRAGTETALVRVVAVFGAAKPHCAYLFANRRANRMKVLVHDGVGI
WLAARRLNQGKFHWPGTHRGLEVGLDAEQLQALVLGLPWQRVGANGAITMI*

now what I want to do is to kind of rename the sequences with a number which has to be 5 digits long. That means the three sequences above should be named like this:

>orf00001 2..112 (direct), 37 amino acids
ADTAQEFISTAVFGTSMSAHHILGLKPVPRVWLFAI*


>orf00002 1482..1790 (direct), 103 amino acids
MKKYTEQAKLSVVEDYCSGSAGHREVAHRHGVNANVIRKWLPIYRDKPVAPLPAFVPLQP
MPKRQADEAVVIALSLGDKSITVKWPISDPDGCARFIRSLSQ*
 
>orf00003 1787..2122 (direct), 112 amino acids
MIRIDAIWLATEPMDMRAGTETALVRVVAVFGAAKPHCAYLFANRRANRMKVLVHDGVGI
WLAARRLNQGKFHWPGTHRGLEVGLDAEQLQALVLGLPWQRVGANGAITMI*

So the 5 digits are fixed and I just need to count upwards seeing the '>' unfortunately I don't know how to make it a fixed length onto five digits.

Thanks for your help (once again ;) )

Best,

Phil

Answer 1

[Alex Reynolds]

Something like the following should get you close:

$ awk ' \
    BEGIN { idx = 0; } \
    if (/^>/) { \
        printf(">orf%05d %s\n", idx, substr($1, 2)); \
        idx++; \
    } \
    else { \
        print $0; \
    } \
​' mySeqs.fa > myRelabeledSeqs.fa

[Phil S.]

Thank you so much for the fast and correct answer. The only thing I had to adjust is a line brake...