Comptute jaccard index from a k-mer abundance matrix

sampleTosampleJaccardIndex.pl

#! /usr/bin/perl
#
use strict;
use warnings;

my %jaccard_h;
while(<>) {
  chomp;
  my($k,@v) = split "\t", $_;
  for(my $i = 0; $i < scalar @v; $i++) {
    if($v[$i] == 0) {
      for(my $j = $i+1; $j < scalar @v; $j++) {
        my $key = $i.'@'.$j;
        if($v[$j] == 0) {
          next;
        } else {
          $jaccard_h{$key}{union}++;
        }
      }
    } else {
      for(my $j = $i+1; $j < scalar @v; $j++) {
        my $key = $i.'@'.$j;
        if($v[$j] == 0) {
          $jaccard_h{$key}{union}++;
        } else {
          $jaccard_h{$key}{inter}++;
          $jaccard_h{$key}{union}++;
        }
      }
    }
  }
}

print join("\t", "sample1", "sample2", "jaccard_index", "inter", "union"), "\n";
foreach my $key (keys %jaccard_h) {
  my($sample1, $sample2) = split "@", $key;
  print join("\t", $sample1, $sample2, $jaccard_h{$key}{inter} / $jaccard_h{$key}{union}, $jaccard_h{$key}{inter}, $jaccard_h{$key}{union}),"\n";
}