|
| 1 | +#!/usr/bin/perl -w |
| 2 | + |
| 3 | +# Lookup the taxon ID associated with each gene in all CursDBs using |
| 4 | +# Config::get_species_taxon_of_strain_taxon() |
| 5 | +# If the lookup returns a result, change the Organism of the genes |
| 6 | +# to the result of the call |
| 7 | +# Then also update the organism of the corresponding genes in the TrackDB |
| 8 | + |
| 9 | +########################### |
| 10 | +# START OF BOILERPLATE CODE |
| 11 | + |
| 12 | +use strict; |
| 13 | +use warnings; |
| 14 | +use Carp; |
| 15 | +use feature ':5.10'; |
| 16 | + |
| 17 | +use File::Basename; |
| 18 | + |
| 19 | +BEGIN { |
| 20 | + my $script_name = basename $0; |
| 21 | + |
| 22 | + if (-f $script_name && -d "../etc") { |
| 23 | + # we're in the scripts directory - go up |
| 24 | + chdir ".."; |
| 25 | + } |
| 26 | +}; |
| 27 | + |
| 28 | +use lib qw(lib); |
| 29 | + |
| 30 | +use Canto::Config; |
| 31 | +use Canto::TrackDB; |
| 32 | +use Canto::Track; |
| 33 | +use Canto::Track::LoadUtil; |
| 34 | +use Canto::Meta::Util; |
| 35 | + |
| 36 | +my $app_name = Canto::Config::get_application_name(); |
| 37 | + |
| 38 | +$ENV{CANTO_CONFIG_LOCAL_SUFFIX} ||= 'deploy'; |
| 39 | + |
| 40 | +my $suffix = $ENV{CANTO_CONFIG_LOCAL_SUFFIX}; |
| 41 | + |
| 42 | +if (!Canto::Meta::Util::app_initialised($app_name, $suffix)) { |
| 43 | + die "The application is not yet initialised, try running the canto_start " . |
| 44 | + "script\n"; |
| 45 | +} |
| 46 | + |
| 47 | +my $config = Canto::Config::get_config(); |
| 48 | +my $schema = Canto::TrackDB->new(config => $config); |
| 49 | + |
| 50 | +my $track_schema = Canto::TrackDB->new(config => $config); |
| 51 | + |
| 52 | +# END OF BOILERPLATE CODE |
| 53 | +######################### |
| 54 | + |
| 55 | + |
| 56 | +# A collection of IDs of genes that need to have their organisms |
| 57 | +# updated in the TrackDB. We populate this map while iterating over |
| 58 | +# the CursDBs |
| 59 | +my %genes_to_update = (); |
| 60 | + |
| 61 | + |
| 62 | +my $proc = sub { |
| 63 | + my $curs = shift; |
| 64 | + my $curs_schema = shift; |
| 65 | + my $track_schema = shift; |
| 66 | + |
| 67 | + my $organism_rs = $curs_schema->resultset('Organism'); |
| 68 | + |
| 69 | + # orig taxon ID to new taxon ID map |
| 70 | + my %taxon_map = (); |
| 71 | + |
| 72 | + # first find all Organisms in this session that need updating, |
| 73 | + # capturing them into %taxon_map |
| 74 | + while (defined (my $organism = $organism_rs->next())) { |
| 75 | + my $orig_taxonid = $organism->taxonid(); |
| 76 | + my $lookup_taxonid = |
| 77 | + $config->get_species_taxon_of_strain_taxon($orig_taxonid); |
| 78 | + |
| 79 | + if (defined $lookup_taxonid && $orig_taxonid != $lookup_taxonid) { |
| 80 | + my $new_org = $curs_schema->resultset('Organism') |
| 81 | + ->find_or_create({ taxonid => $lookup_taxonid }); |
| 82 | + $taxon_map{$orig_taxonid} = $new_org; |
| 83 | + } |
| 84 | + } |
| 85 | + |
| 86 | + my $gene_rs = $curs_schema->resultset('Gene') |
| 87 | + ->search({}, { prefetch => 'organism' }); |
| 88 | + |
| 89 | + # Iterate over genes and update the Organism based on %taxon_map |
| 90 | + while (defined (my $gene = $gene_rs->next())) { |
| 91 | + my $gene_taxonid = $gene->organism()->taxonid(); |
| 92 | + |
| 93 | + my $new_org = $taxon_map{$gene_taxonid}; |
| 94 | + |
| 95 | + if (defined $new_org) { |
| 96 | + $genes_to_update{$gene->primary_identifier()} = 1; |
| 97 | + |
| 98 | + print "updating ", $gene->primary_identifier(), " in CursDB\n"; |
| 99 | + print " $gene_taxonid -> ", $new_org->taxonid(), "\n"; |
| 100 | + $gene->organism($new_org); |
| 101 | + $gene->update(); |
| 102 | + } |
| 103 | + } |
| 104 | +}; |
| 105 | + |
| 106 | +my $load_util = Canto::Track::LoadUtil->new(schema => $schema); |
| 107 | + |
| 108 | + |
| 109 | +my $txn_proc = sub { |
| 110 | + # iterate over CursDBs |
| 111 | + Canto::Track::curs_map($config, $track_schema, $proc); |
| 112 | + |
| 113 | + # update organisms of genes in the TrackDB |
| 114 | + for my $gene_primary_identifier (keys %genes_to_update) { |
| 115 | + my $gene = $track_schema->resultset('Gene') |
| 116 | + ->find({ primary_identifier => $gene_primary_identifier }, |
| 117 | + { prefetch => 'organism' }); |
| 118 | + if (defined $gene) { |
| 119 | + my $props_rs = $gene->organism()->organismprops()->search({}, { prefetch => 'type' }); |
| 120 | + my $orig_taxonid; |
| 121 | + while (defined (my $prop = $props_rs->next())) { |
| 122 | + if ($prop->type()->name() eq 'taxon_id') { |
| 123 | + $orig_taxonid = $prop->value(); |
| 124 | + last; |
| 125 | + } |
| 126 | + } |
| 127 | + if (!defined $orig_taxonid) { |
| 128 | + die "internal error: can't find taxon ID for $gene_primary_identifier\n"; |
| 129 | + } |
| 130 | + my $new_taxonid = |
| 131 | + $config->get_species_taxon_of_strain_taxon($orig_taxonid); |
| 132 | + my $new_organism = |
| 133 | + $load_util->find_organism_by_taxonid($new_taxonid); |
| 134 | + |
| 135 | + print "updating ", $gene->primary_identifier(), " in TrackDB\n"; |
| 136 | + $gene->organism($new_organism); |
| 137 | + $gene->update(); |
| 138 | + } |
| 139 | + } |
| 140 | +}; |
| 141 | + |
| 142 | +$track_schema->txn_do($txn_proc); |
| 143 | + |
| 144 | +exit 0; |
0 commit comments