Take the 2-minute tour ×
Unix & Linux Stack Exchange is a question and answer site for users of Linux, FreeBSD and other Un*x-like operating systems.. It's 100% free, no registration required.

So, I have a string that looks like this:

AUGGCCAUGGCGCCCAGAACUGAGAUCAAUAGUACCCGUAUUAACGGGUGA

And I want to split the string into 3-character chunks delimited by a '+' sign.

AUG+GCC+AUG+GCG+CCC+AGA+ACU+GAG+AUC+AAU+AGU+ACC+CGU+AUU+AAC+GGG+UGA

And I want to do that with my good friend sed.

I tried

cat codons | sed -r 's/([A-Z]\{3\})/\1\+/g'

...with no success.

What sed command can I use?

share|improve this question
1  
Aren't this somehow connected to Rosalind? Just curious. –  m0nhawk Dec 21 '12 at 20:32
add comment

4 Answers

up vote 13 down vote accepted

Since you don't want a trailing +, you could do:

fold -w3 | paste -sd+ -

That is, fold the lines on 3 character width, and paste those 3 character lines with themselves with + as the delimiter which in effect is like changing every newline character but the last one into a +. If the input had more than one line, you'll end up with those lines joined with a + which may or may not be what you want.

If you do need it to be sed, you can remove the trailing + after:

sed 's/.../&+/g;s/+$//'
share|improve this answer
    
Would you mind adding a short explanation of how that works? –  N.N. Dec 21 '12 at 20:35
    
@N.N. It works because +$ matches a plus symbol immediately before the end of a line. –  Chris Down Dec 21 '12 at 21:14
    
fold -w3 breaks the string into 3 character lines. paste -sd+ - turns the newlines into +. –  bahamat Dec 22 '12 at 3:40
add comment
sed 's/.../&+/g'

to get your way working you don't need to escape {} symbols:

sed -r 's/([A-Z]{3})/\1+/g'
share|improve this answer
1  
who knew! i was so close yet so far... thanks... –  ixtmixilix Dec 21 '12 at 20:10
    
Both add a trailing '+'. Is this intended? –  N.N. Dec 21 '12 at 20:25
add comment

This might work for you (GNU sed):

sed 's/...\B/&+/g' file
share|improve this answer
add comment

If sed is not a must using Ruby might be an alternative. The Ruby interpreter, ruby, can be used like sed and awk by running it with the -n option which makes it iterate over its input. The interpreter can then be feed with a Ruby one-liner by adding it as an argument to the -e option (which tells the interpreter to interpret the argument of -e rather than looking for a script in a file).

For this particular problem you can use the following one-liner (adapted from http://stackoverflow.com/a/3184271/789593):

ruby -ne 'puts $_.scan(/.{3}|.+/).join("+")'

In plain language it

  • matches any 3 characters or at least one character, scan(/.{3}|.+/), in the input string, $_ (in this case the input is expected to come from standard in) and puts each match in an array,
  • joins the array into a string with a '+' connecting each element, join("+"),
  • and prints it terminated by a newline puts.

For example

echo "AUGGCCAUGGCGCCCAGAACUGAGAUCAAUAGUACCCGUAUUAACGGGUG" | ruby -ne 'puts $_.scan(/.{3}|.+/).join("+")'
AUG+GCC+AUG+GCG+CCC+AGA+ACU+GAG+AUC+AAU+AGU+ACC+CGU+AUU+AAC+GGG+UG

Note that it does not add any trailing '+'.

share|improve this answer
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.