http://qs321.pair.com?node_id=1020265


in reply to Reading values from one .csv, searching for closest values in second .csv, returning results in third .csv?

No need for a hash, an array should be enough. Try the following:
#!/usr/bin/perl use warnings; use strict; open my $GENES, '<', 'genes' or die $!; open my $LOCATIONS, '<', 'locations' or die $!; chomp(my @locations = map { (split ' ')[1] } <$LOCATIONS>); # If IDs are not already sorted, uncomment the following line: # @locations = sort { $a <=> $b } @locations; for (<$GENES>) { my ($chromosome, $start, $end) = split ' '; print "$chromosome\t$start\t$end"; my $idx = 0; # For $end, start searching where you left for + $start. my $correction = 0; # Needed for Start(-) == Start and End(+) == E +nd. for my $pos ($start, $end) { $idx++ while $locations[$idx] <= $pos - $correction and $idx <= $#locations; die "No numbers around $pos ($idx)\n" if $idx == 0 or $idx > $#locations; print "\t$locations[$idx-1]\t$locations[$idx]"; $correction = 1; } print "\n"; }
Update: Fixed border cases.
لսႽ† ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ

Replies are listed 'Best First'.
Re^2: Reading values from one .csv, searching for closest values in second .csv, returning results in third .csv?
by pickleswarlz (Initiate) on Feb 26, 2013 at 19:38 UTC

    Thank you very much! I am using the first suggestion, as I don't really understand the binary search yet, and so far it doesn't seem to have problems with searching the 500,000 rows.

    I am now running into the problem of the results being output on different lines in the resulting CSV file for each print command. Here is my code:

    #!/usr/bin/perl use warnings; use strict; open my $GENES, '<', 'chr1data.csv' or die $!; open my $LOCATIONS, '<', 'chr1snps.csv' or die $!; chomp(my @locations = map { (split ',')[2] } <$LOCATIONS>); # If IDs are not already sorted, uncomment the following line: # @locations = sort { $a <=> $b } @locations; for (<$GENES>) { my ($chromosome, $start, $end) = split ','; print "$chromosome,$start,$end"; my $idx = 0; # For $end, start searching where you left for + $start. my $correction = 0; # Needed for Start(-) == Start and End(+) == E +nd. for my $pos ($start, $end) { $idx++ while $locations[$idx] <= $pos - $correction and $idx <= $#locations; die "No numbers around $pos ($idx) \n" if $idx == 0 or $idx > $#locations; print ",$locations[$idx-1],$locations[$idx]"; $correction = 1; } print "\n"; }

    Printing print ",$locations[$idx-1],$locations[$idx]"; puts this information on a new line. I'd like it to come out on the same line as print "$chromosome,$start,$end"; for each search. Do I have a \n in the wrong place?

      Your $end probably contains newline (I split on ' ', which removed it, you split on a comma). Just
      chomp $end;
      before printing it.
      لսႽ† ᥲᥒ⚪⟊Ⴙᘓᖇ Ꮅᘓᖇ⎱ Ⴙᥲ𝇋ƙᘓᖇ
        Brilliant, works perfectly! Thanks!