split one column by ',' and use the values in calculations - perl

Question

I'm writing a script in which I'm using a text file, where in one column there can be two letters (A,B,C or D) seperated by a ",". This column can also just contain one of those letters. I have to use both letters for further calculations in the rest of the script. This is a simplified example of my input file (here $variants):
C1 C2 C3 C4 C5 C6 ... C9
text 2 A D values and text in the other columns
text 4 B C values and text in the other columns
text 5 A B,D values and text in the other columns
So in line 3 of C4 there is a B and D. After C4 there are still a lot of columns, which cannot be changed since I need them in other parts of my script.
I have a second input file from which, based on the letters present in C3 and C4, some values are extracted. This is how this second input file looks like (here $frequency)
C1 C2 A a B b C c D d
text 1 0 1 0 0 0 0 0 0
text 2 1 0 5 4 0 0 0 0
text 3 0 0 0 0 10 11 3 6
text 4 1 0 9 4 0 2 0 0
text 5 5 3 0 0 6 7 4 0
This is how my output should look like:
C1 C2 C3 C4 C5 C6 C7 C8 C9 C10
text 2 A D 1 0 0 0 empty
text 4 B C 9 4 0 2 empty
text 5 A B,D 5 3 0 0 4 0
So for line 1, there is A in C3, then the script extracts the values for A and a from $frequency and puts them in C5 and C6. The values from C4 are then put in C7 and C8 from the output file. Now in the 3rd line there is B,D in C4. So what the script needs to do now is putting the corresponding values from B and b in C7 and C8 and the values for D and d in C9 and C10.
The only thing where I have still problems in my script is in splitting up this C4 when there is a ','. The rest is working.
This is how the problematic part of my script looks like
while(<$variants>){
next if /^\s*#/;
next if /^\s*"/;
chomp;
my ($chr, $pos, $refall, #altall) = split /\t/; # How should I specify here the C4, as an array? So that I don't know
my #ref_data = #{$frequency_data[$pos]}{$refall, lc($refall)};
my #alt_data = #{$frequency_data[$pos]}{$altall, lc($altall)}; # this works for C3 ($refall), but not for C4 when there are two letters
$pos = $#genes if $circular and $pos > $#genes; # adding annotation # this can be ignored here, since this line isn't part of my question
print join("\t","$_ ", $genes[$pos] // q(), #ref_data, #alt_data), "\n"; # printing annotation
}
So could someone help me with splitting of this C4 by ',' and still use the information for extracting values from $variants

Answer 1

I think the easiest would be treating columns 3 and 4 as lists from the get-go:
while(<$variants>){
next if /^\s*#/;
next if /^\s*"/;
chomp;
my ($chr, $pos, $refall_string, $altall_string, #other) = split /\t/;
my #refall = split(",", $refall_string);
my #altall = split(",", $altall_string);
my #ref_data_all = (); # Treat C3 as array just in case...
foreach my $refall (#refall) {
push #ref_data_all, #{$frequency_data[$pos]}{ $refall, lc($refall) };
}
my #alt_data_all = ();
foreach my $altall (#altall) {
push #alt_data_all, #{$frequency_data[$pos]}{ $altall, lc($altall) };
}
$pos = $#genes if $circular and $pos > $#genes;
print join("\t","$_ ", $genes[$pos] // q(),
#ref_data_all, #alt_data_all), "\n";
}
I didn't test this but the approach should be clear even if there's some minor bugs.

Answer 2

All you need is a couple of map calls.
If you write
map { $_, lc } split /,/, $refall
then you have split the field at any commas and duplicated each letter as upper case and lower case.
This is the complete loop (tested).
while (<$variants>) {
next if /^\s*#/;
next if /^\s*"/;
chomp;
my ($chr, $pos, $refall, $altall) = split /\t/;
my $entry = $frequency_data[$pos];
my #ref_data = map { $entry->{$_} } map { $_, lc } split /,/, $refall;
my #alt_data = map { $entry->{$_} } map { $_, lc } split /,/, $altall;
$pos = $#genes if $circular and $pos > $#genes;
print join("\t","$_ ", $genes[$pos] // q(), #ref_data, #alt_data), "\n";
}