Edit all the fasta headers using awk

Question

kindly your comment highly appreciated

have more than 4k header sequences look like:

>LTR_retrotransposon100_Gypsy? 
cgtgcaccccaatgttcatagcagcactatttacaatagccaagacatggaaacaaccta aatgtcctttgacaggtgactggataaagaagctgtggtatatttatacaatggaatgct actcagccataaaaaagaataagataatgccatttacagcaacatggatagacctggaga atgttattctaagtgaagtaagccagaaagaaaaagaaaaataccatatgctatcactta

>LTR_retrotransposon10011_ERVK 
taatttgaaaagatacatgcaacccaatgttcatagcagcattatttacaattgccaaga tatggaagcaatctaagtgtccatcaacagatgaatggagagagaagatgtggtatatat acaataaaatactattctgtcataaaaatgaataaaattctgccatttgcaacaacatgg atagacctggagggtattatgctatgtgaaataagtcagacagagaaagacatatactat

>LTR_retrotransposon10011_ERVL-MaLR 
    taatttgaaaagatacatgcaacccaatgttcatagcagcattatttacaattgccaaga tatggaagcaatctaagtgtccatcaacagatgaatggagagagaagatgtggtatatat acaataaaatactattctgtcataaaaatgaataaaattctgccatttgcaacaacatgg atagacctggagggtattatgctatgtgaaataagtcagacagagaaagacatatactat

>LTR_retrotransposon100_Copia 
    cgtgcaccccaatgttcatagcagcactatttacaatagccaagacatggaaacaaccta aatgtcctttgacaggtgactggataaagaagctgtggtatatttatacaatggaatgct actcagccataaaaaagaataagataatgccatttacagcaacatggatagacctggaga atgttattctaagtgaagtaagccagaaagaaaaagaaaaataccatatgctatcactta

I would like to use awk or sed to remove everything between the underscore, including the underscore and replace it with / plus add sequential RNA followed by #.

and the target output could be:

>RNA1#LTR/Gypsy?
ctcagcagcactatttacaatagccaagacatggaaacaacctaaatgtcttatcaatag atgactggataaaggagctgtggtatatctatacaatggaataccatcagccataaaaaa gaataaaatattgccatttgcagcaacatggatggacctggagattatcattctaaggga agtaagccagaaagagaaagaaaaataccatatgatatcacttatatgtggaggtaaaaa aaaaaaaaaagacacaaatcaatttatttgcaaaacatacaTGGActttcagacatagaa

>RNA2#LTR/ERVL-MaLR
taatttgaaaagatacatgcaacccaatgttcatagcagcattatttacaattgccaaga tatggaagcaatctaagtgtccatcaacagatgaatggagagagaagatgtggtatatat acaataaaatactattctgtcataaaaatgaataaaattctgccatttgcaacaacatgg atagacctggagggtattatgctatgtgaaataagtcagacagagaaagacatatactat



>RNA3#LTR/ERVK
    taatttgaaaagatacatgcaacccaatgttcatagcagcattatttacaattgccaaga tatggaagcaatctaagtgtccatcaacagatgaatggagagagaagatgtggtatatat acaataaaatactattctgtcataaaaatgaataaaattctgccatttgcaacaacatgg atagacctggagggtattatgctatgtgaaataagtcagacagagaaagacatatactat

>RNA4#LTR/Copia 
    cgtgcaccccaatgttcatagcagcactatttacaatagccaagacatggaaacaaccta aatgtcctttgacaggtgactggataaagaagctgtggtatatttatacaatggaatgct actcagccataaaaaagaataagataatgccatttacagcaacatggatagacctggaga atgttattctaagtgaagtaagccagaaagaaaaagaaaaataccatatgctatcactta

Answer 1

EDIT after OP's update:

Try this Perl one-liner:

perl -pe 'BEGIN { $i = 1 } $i += s{>([^_]+)_.*_}{>RNA${i}#${1}/}' input_file > output_file

Here, the command line flags are:
-e: tells the perl interpreter to look for the code inline, rather than in a file with the script,
-p: loop over the input one line at a time, execute the code in the one-liner for each iteration of the loop, and print the result into STDOUT.
The code has regex replacement (see perlre for more details):
s{>([^_]+)_.*_}{>RNA${i}#${1}/}:
replace an underscore, followed by zero or more occurrences (* quantifier) of any character (.), followed by an underscore, with slash (/).
Keep > and the original header part before the underscore, captured by parens into $1 variable.
Add string RNA${i}, where $i is the counter of the fasta headers. The counter is incremented by 1 when the substitution occurs, that is, when the next header occurs.
The variables are used like so in the substitution: ${i}, ${1} to prevent wrong interpolation.
I am also using s{}{} as delimiters for readability (as in most cases) instead of the default s///.

EXAMPLE:

echo ">LTR_retrotransposon100_Gypsy?\nACGT\n>LTR_retrotransposon10011_ERVK\nTCGA" | perl -pe 'BEGIN { $i = 1 } $i += s{>([^_]+)_.*_}{>RNA${i}#${1}/}'

Prints:

>RNA1#LTR/Gypsy?
ACGT
>RNA2#LTR/ERVK
TCGA

Answer 2

Your own answer doesn't seem to do what you ask as it doesn't add the sequential identifier to each sequence. Maybe try this?

awk -v rec=0 '{
    if($0 ~ "^>"){
        rec++
        sub(">", ">RNA" rec "#", $0)
        sub("_.*_", "/", $0);
    } 
    print $0
}' yourfile.fa

Answer 3

sed -e 's:^\(.*\)_.*_\(.*\)$:\1/\2:' < input > output