| 1 |
#!/usr/bin/perl |
| 2 |
use strict; |
| 3 |
#use Getopt::Std; |
| 4 |
use FindBin;use lib $FindBin::Bin; |
| 5 |
|
| 6 |
my $usage = q{Usage: |
| 7 |
|
| 8 |
Download ftp://ftp.ncbi.nih.gov/pub/taxonomy/taxdump.tar.gz |
| 9 |
And unpack it in the current directory |
| 10 |
|
| 11 |
Then run $0 which will create |
| 12 |
the bcp files for loading into the database. |
| 13 |
}; |
| 14 |
umask 0002; |
| 15 |
my %merged; #insert "merged" entries too, even though their data is duplicate |
| 16 |
#getopts('o:') || die($usage."\n"); |
| 17 |
#my $outfile=$Getopt::Std::opt_o; |
| 18 |
die("$usage\n") if $ARGV[0]=~m/^\-[\-help]*$/; |
| 19 |
die("$usage Error: cannot find nodes.dmp!\n") unless -f 'nodes.dmp'; |
| 20 |
my $cmd='cut -f5 nodes.dmp | sort -u |'; |
| 21 |
open(TAXC, '>taxon_cat.bcp') || die "Error creating taxon_cat.bcp file!\n"; |
| 22 |
|
| 23 |
my %ranks; #rank_name => rank_id |
| 24 |
open(RANK, $cmd) || die "Error opening $cmd pipe!\n"; |
| 25 |
my ($rank_id, $rank_name); |
| 26 |
while (<RANK>) { |
| 27 |
chomp; |
| 28 |
s/^\s+//;s/\s+$//; |
| 29 |
next unless $_; |
| 30 |
$rank_id++; |
| 31 |
$rank_name=$_; |
| 32 |
$ranks{$rank_name}=$rank_id; |
| 33 |
print TAXC join("\t", 'R', $rank_id, '', $rank_name, '')."\n"; |
| 34 |
} |
| 35 |
close(RANK); |
| 36 |
|
| 37 |
open(MRG, 'merged.dmp') || die("Error opening merged.dmp!\n"); |
| 38 |
while (<MRG>) { |
| 39 |
chomp; |
| 40 |
next unless $_; |
| 41 |
s/\t\|$//; |
| 42 |
my @t=split(/\t\|\t/); |
| 43 |
push(@{$merged{$t[1]}},$t[0]); |
| 44 |
} |
| 45 |
close(MRG); |
| 46 |
|
| 47 |
open(DIV, 'division.dmp') || die("Error opening division.dmp!\n"); |
| 48 |
while (<DIV>) { |
| 49 |
chomp; |
| 50 |
next unless $_; |
| 51 |
s/\t\|$//; |
| 52 |
my @t=split(/\t\|\t/); |
| 53 |
print TAXC join("\t", 'D', @t[0..3])."\n"; |
| 54 |
} |
| 55 |
|
| 56 |
close(DIV); |
| 57 |
close(TAXC); |
| 58 |
open(NAMES, 'names.dmp') || die("Error opening names.dmp!\n"); |
| 59 |
open(TAXN, '>taxon_names.bcp') || die("Error creating taxon_names.bcp!\n"); |
| 60 |
my %names; # tax_id => [sci_name, com_name] |
| 61 |
while(<NAMES>) { |
| 62 |
chomp; |
| 63 |
next unless $_; |
| 64 |
s/\t\|$//; |
| 65 |
my @t=split(/\t\|\t/); |
| 66 |
print TAXN join("\t", @t)."\n"; |
| 67 |
if ($t[3]=~/\bcommon name/) { |
| 68 |
my $d=$names{$t[0]}; |
| 69 |
$names{$t[0]}= $d ? [$$d[0], $t[1]] : ['',$t[1]]; |
| 70 |
} |
| 71 |
elsif ($t[3]=~/\bscientific name/) { |
| 72 |
my $d=$names{$t[0]}; |
| 73 |
$names{$t[0]}= $d ? [$t[1], $$d[1]] : [$t[1]]; |
| 74 |
} |
| 75 |
} |
| 76 |
close(NAMES); |
| 77 |
close(TAXN); |
| 78 |
open(TAX, '>taxon.bcp') || die("Error creating taxon.bcp!\n"); |
| 79 |
open(NODES, 'nodes.dmp') || die("Error opening names.dmp!\n"); |
| 80 |
while(<NODES>) { |
| 81 |
chomp; |
| 82 |
next unless $_; |
| 83 |
s/\t\|$//; |
| 84 |
my @t=split(/\t\|\t/); |
| 85 |
my $rank_id=$ranks{$t[2]}; |
| 86 |
die("Error: invalid rank $t[2] from nodes.dmp line:\n$_\n") |
| 87 |
unless defined($rank_id); |
| 88 |
my $d=$names{$t[0]}; |
| 89 |
die("Error: names not found for taxon $t[0] at nodes.dmp line:\n$_\n") |
| 90 |
unless $d && $$d[0]; |
| 91 |
my $sci_name=$$d[0]; |
| 92 |
my $com_name=$$d[1] || ''; |
| 93 |
print TAX join("\t",$t[0],$t[1],$rank_id, $t[4], $t[5], |
| 94 |
$sci_name, $com_name)."\n"; |
| 95 |
if (my $md=$merged{$t[0]}) { |
| 96 |
foreach my $tid (@$md) { |
| 97 |
print TAX join("\t",$tid,$t[1],$rank_id, $t[4], $t[5], |
| 98 |
$sci_name, $com_name)."\n"; |
| 99 |
} |
| 100 |
} |
| 101 |
} |
| 102 |
close(NODES); |
| 103 |
close(TAX); |
