Not greedy (although it is), global (meaning find all of them).

Update: Show me to read quickly.

One more question related to this. Is there a way to try and find a variable outside of a start and end tag and then replace this with something new? So the start sentence would look like this,

"The increase in sensitivity of HIV - infected cells to <GENE> Fas </GENE> killing mapped to <GENE> vpu </GENE> , while nef , <GENE> vif </GENE> , <GENE> vpr </GENE> , and second exon of <GENE> tat</GENE> did not appear to contribute"

And end up looking like this:

"The increase in sensitivity of HIV - infected cells to <GENE> Fas </GENE> killing mapped to <GENE> vpu </GENE> , while <PGENE> nef </PGENE> , <GENE> vif </GENE> , <GENE> vpr </GENE> , and second exon of <GENE> tat</GENE> did not appear to contribute"

Notice the addition of a PGENE tag following the match of nef (outside of the start and end gene tags) to a variable.

I appreciate this might be slightly confusing, but at the moment I'm racking my brain trying to figure out a way to join it all back together correctly if I do this using lots of splits.

Thanks

Lots of splits? Use just one:

my $s = 'vif The increase in sensitivity of HIV - infected cells to <G
+ENE> vif Fas </GENE> killing mapped to <GENE> vpu </GENE> , while nef
+ , vif, <GENE> vif </GENE> , <GENE> vpr </GENE> , and second exon of 
+<GENE> tat</GENE> did not appear to contribute';
my @ar = split m%(</?GENE>)%, $s;
for my $i (0 .. @ar/4) {
    $ar[4*$i]=~ s%vif%<PGENE>vif</PGENE>%;
}
$s = join q[],@ar;
print "$s\n";
[download]