Question

How can I replace any string that is not “A”, “C”, “G”, “T”, “N”, with sed?

For example I have the following data:

>AFCCCCC 1
cagktgagtgataaggc
>AFCGH22 1
cagntgagtgstaaggc

What I want to remove every character that is not [ACGTN] in line that do not start with ‘>’

Hence I hope to get this output:

 >AFCCCCC 1
 cagtgagtgataaggc
 >AFCGH22 1
 cagntgagtgtaaggc

Note that I removed ‘k’ and ‘s’ for first and second sequence.

Answer 1

Try this:

sed '/^>/!s/[^ACGTN]//ig' input.txt