how to remove characters before and after a search pattern in a string?

BhariD has asked for the wisdom of the Perl Monks concerning the following question:

I have the following example string:

   ACGTTGGCTATTGGGCCCGCT
[download]

I am searching for the pattern 'GTT' in the above string. As you will notice, pattern 'GTT' occurs in the start of the string. Once the pattern is matched, I need to remove any number of characters before (here, AC) and 3 characters (GGC) after the pattern ('GTT').

so the output will look something like this:

   TATTGGGCCCGCT
[download]

Any pointers how I can do this? Thanks!

Comment on how to remove characters before and after a search pattern in a string? Select or Download Code

Replies are listed 'Best First'.
Re: how to remove characters before and after a search pattern in a string? by ikegami (Patriarch) on Mar 03, 2010 at 17:25 UTC
First GTT: `s/^.?GTT.{3}//s` [download] Last GTT: `s/^.GTT.{3}//s` [download] Every GTT: `s/^.*GTT.{3}//s` [download]	[reply] [d/l] [select]
Re^2: how to remove characters before and after a search pattern in a string? by jrsimmon (Hermit) on Mar 03, 2010 at 18:15 UTC
I don't see the difference bewteen Last GTT: `s/^.GTT.{3}//s` [download] and Every GTT:* `s/^.GTT.{3}//s` [download] . Shouldn't it have been `s/^.GTT.{3}$//s` [download] ? Also, won't the last one remove everything preceding the last occurrence of GTT (assuming it's followed by 3 somethings), and not just every occurrence of 'GTT' within the string?	[reply] [d/l] [select]
Re^3: how to remove characters before and after a search pattern in a string? by ikegami (Patriarch) on Mar 03, 2010 at 18:23 UTC
There isn't any difference. Deleting everything before every instance and deleting everything before the last instances gives the same result. No, adding `$` makes it so it will only match if GTT is three characters from the end of the string (or 4 if the last character is \n).	[reply] [d/l]
Re: how to remove characters before and after a search pattern in a string? by thunders (Priest) on Mar 03, 2010 at 18:49 UTC
since your pattern is a simple string, you can use index and substr. `my $str = "ACGTTGGCTATTGGGCCCGCT"; my $pattern = "GTT"; my $offset = index($str,$pattern) + length($pattern) + 3; $str = substr($str,$offset); print $str;` [download]	[reply] [d/l]
Re^2: how to remove characters before and after a search pattern in a string? by jwkrahn (Abbot) on Mar 03, 2010 at 19:58 UTC
Probably better as: `my $str = "ACGTTGGCTATTGGGCCCGCT"; my $pattern = "GTT"; if ( ( my $offset = index $str, $pattern ) >= 0 ) { + substr $str, 0, $offset + length( $pattern ) + 3, ""; } print $str;` [download]	[reply] [d/l]