[Gmod-schema-cmts] SF.net SVN: gmod:[25262] schema/trunk/chado/modules

[<scottcain@us...>]

Revision: 25262
          http://gmod.svn.sourceforge.net/gmod/?rev=25262&view=rev
Author:   scottcain
Date:     2012-03-29 16:23:06 +0000 (Thu, 29 Mar 2012)
Log Message:
-----------
commiting Dave's/FlyBase's interaction module

Added Paths:
-----------
    schema/trunk/chado/modules/interaction/
    schema/trunk/chado/modules/interaction/interaction.sql

Added: schema/trunk/chado/modules/interaction/interaction.sql
===================================================================
--- schema/trunk/chado/modules/interaction/interaction.sql	                        (rev 0)
+++ schema/trunk/chado/modules/interaction/interaction.sql	2012-03-29 16:23:06 UTC (rev 25262)
@@ -0,0 +1,316 @@
+
+
+-- ==========================================
+-- Chado interaction module
+--
+-- ==========================================
+-- NOTES:
+--
+-- Designed to represent various types of interactions 
+-- initially to be deployed for physical interactions (protein-protein)
+-- between n number of chado features
+--
+-- prop and pub tables follow usual chado conventions
+--
+-- table:interaction serves as a coalescing table for all aspects of the interaction
+--       a uniquename and link to a cvterm for the type of interaction are stored here
+--
+-- table:feature_interaction links features to the interaction
+--       a cvterm_id for the role that the feature is playing in the interaction
+--       must be specified (eg. prey, bait, evidence_for)
+--       rank may be used to order interacting features in an interaction
+--
+
+-- ================================================
+-- TABLE: interaction
+-- ================================================
+drop table interaction cascade;
+create table interaction (
+       	interaction_id serial NOT NULL,
+       	primary key (interaction_id),
+	uniquename text NOT NULL,
+        type_id int NOT NULL,
+        foreign key (type_id) references cvterm (cvterm_id)
+	on delete cascade INITIALLY DEFERRED,
+        description text,
+ 	is_obsolete boolean not null default false,
+	constraint interaction_c1 unique(uniquename,type_id)
+);
+create index interaction_idx1 on interaction (uniquename);
+create index interaction_idx2 on interaction (type_id);
+
+grant all on interaction to public;
+grant all on interaction_interaction_id_seq to public;
+
+-- ================================================
+-- TABLE: interactionprop
+-- ================================================
+
+drop table interactionprop cascade;
+create table interactionprop (
+    interactionprop_id serial not null,
+    primary key (interactionprop_id),
+    interaction_id int not null,
+    foreign key (interaction_id) references interaction (interaction_id) on delete cascade INITIALLY DEFERRED,
+    type_id int not null,
+    foreign key (type_id) references cvterm (cvterm_id) on delete cascade INITIALLY DEFERRED,
+    value text null,
+    rank int not null default 0,
+    constraint interactionprop_c1 unique (interaction_id,type_id,rank)
+);
+create index interactionprop_idx1 on interactionprop (interaction_id);
+create index interactionprop_idx2 on interactionprop (type_id);
+
+grant all on interactionprop to public;
+grant all on interactionprop_interactionprop_id_seq to public;
+
+-- ================================================
+-- TABLE: interactionprop_pub
+-- ================================================
+create table interactionprop_pub (
+       interactionprop_pub_id serial not null,
+       primary key (interactionprop_pub_id),
+       interactionprop_id int not null,
+          foreign key (interactionprop_id) references interactionprop (interactionprop_id) on delete cascade INITIALLY DEFERRED,
+     pub_id int not null,
+     foreign key (pub_id) references pub (pub_id) on delete cascade INITIALLY DEFERRED,
+     constraint interactionprop_pub_c1 unique (interactionprop_id,pub_id)
+);
+create index interactionprop_pub_idx1 on interactionprop_pub (interactionprop_id);
+create index interactionprop_pub_idx2 on interactionprop_pub (pub_id);
+
+
+grant all on interactionprop_pub to public;
+grant all on interactionprop_pub_interactionprop_pub_id_seq to public;
+
+-- ================================================
+-- TABLE: interaction_pub
+-- ================================================
+
+drop table interaction_pub cascade;
+create table interaction_pub (
+       interaction_pub_id serial not null,
+       primary key (interaction_pub_id),
+       interaction_id int not null,
+       foreign key (interaction_id) references interaction (interaction_id) on delete cascade INITIALLY DEFERRED,
+       pub_id int not null,
+       foreign key (pub_id) references pub (pub_id) on delete cascade INITIALLY DEFERRED,
+       constraint interaction_pub_c1 unique(interaction_id,pub_id)       
+);
+create index interaction_pub_idx1 on interaction_pub (interaction_id);
+create index interaction_pub_idx2 on interaction_pub (pub_id);
+
+grant all on interaction_pub to public;
+grant all on interaction_pub_interaction_pub_id_seq to public;
+
+-- ================================================
+-- TABLE: interaction_expression
+-- ================================================
+
+drop table interaction_expression cascade;
+
+create table interaction_expression (
+       interaction_expression_id serial not null,
+       primary key (interaction_expression_id),
+       expression_id int not null,
+       foreign key (expression_id) references expression (expression_id) on delete cascade INITIALLY DEFERRED,
+       interaction_id int not null,
+       foreign key (interaction_id) references interaction (interaction_id) on delete cascade INITIALLY DEFERRED,
+       pub_id int not null,
+       foreign key (pub_id) references pub (pub_id) on delete cascade INITIALLY DEFERRED,
+       unique(expression_id,interaction_id,pub_id)       
+);
+create index interaction_expression_idx1 on interaction_expression (expression_id);
+create index interaction_expression_idx2 on interaction_expression (interaction_id);
+create index interaction_expression_idx3 on interaction_expression (pub_id);
+
+grant all on interaction_expression to public;
+grant all on interaction_expression_interaction_expression_id_seq to public;
+
+--================================================
+-- TABLE: interaction_expressionprop
+-- ================================================
+
+drop table interaction_expressionprop;
+create table interaction_expressionprop (
+    interaction_expressionprop_id serial not null,
+    primary key (interaction_expressionprop_id),
+    interaction_expression_id int not null,
+    foreign key (interaction_expression_id) references interaction_expression (interaction_expression_id) on delete cascade,
+    type_id int not null,
+    foreign key (type_id) references cvterm (cvterm_id) on delete cascade INITIALLY DEFERRED,
+    value text null,
+    rank int not null default 0,
+    unique (interaction_expression_id,type_id,rank)
+);
+
+
+create index interaction_expressionprop_idx1 on interaction_expressionprop (interaction_expression_id);
+create index interaction_expressionprop_idx2 on interaction_expressionprop (type_id);
+
+grant all on interaction_expressionprop to public;
+grant all on interaction_expressionprop_interaction_expressionprop_id_seq to public;
+
+-- ================================================
+-- TABLE: interaction_cvterm
+-- ================================================
+
+drop table interaction_cvterm cascade;
+
+create table interaction_cvterm (
+       interaction_cvterm_id serial not null,
+       primary key (interaction_cvterm_id),
+       interaction_id int not null,
+       foreign key (interaction_id) references interaction (interaction_id) on delete cascade INITIALLY DEFERRED,
+       cvterm_id int not null,
+       foreign key (cvterm_id) references cvterm (cvterm_id) on delete cascade INITIALLY DEFERRED,
+       unique(interaction_id,cvterm_id)
+);
+create index interaction_cvterm_idx1 on interaction_cvterm (interaction_id);
+create index interaction_cvterm_idx2 on interaction_cvterm (cvterm_id);
+create index interaction_cvterm_idx3 on interaction_cvterm (pub_id);
+
+grant all on interaction_cvterm to public;
+grant all on interaction_cvterm_interaction_cvterm_id_seq to public;
+
+--================================================
+-- TABLE: interaction_cvtermprop
+-- ================================================
+
+drop table interaction_cvtermprop;
+create table interaction_cvtermprop (
+    interaction_cvtermprop_id serial not null,
+    primary key (interaction_cvtermprop_id),
+    interaction_cvterm_id int not null,
+    foreign key (interaction_cvterm_id) references interaction_cvterm (interaction_cvterm_id) on delete cascade,
+    type_id int not null,
+    foreign key (type_id) references cvterm (cvterm_id) on delete cascade INITIALLY DEFERRED,
+    value text null,
+    rank int not null default 0,
+    unique (interaction_cvterm_id,type_id,rank)
+);
+
+
+create index interaction_cvtermprop_idx1 on interaction_cvtermprop (interaction_cvterm_id);
+create index interaction_cvtermprop_idx2 on interaction_cvtermprop (type_id);
+
+grant all on interaction_cvtermprop to public;
+grant all on interaction_cvtermprop_interaction_cvtermprop_id_seq to public;
+
+
+-- ================================================
+-- TABLE: library_interaction
+-- ================================================
+
+drop table library_interaction cascade;
+
+create table library_interaction (
+       library_interaction_id serial not null,
+       primary key (library_interaction_id),
+       interaction_id int not null,
+       foreign key (interaction_id) references interaction (interaction_id) on delete cascade INITIALLY DEFERRED,
+       library_id int not null,
+       foreign key (library_id) references library (library_id) on delete cascade INITIALLY DEFERRED,
+       pub_id int not null,
+       foreign key (pub_id) references pub (pub_id) on delete cascade INITIALLY DEFERRED,
+       unique(interaction_id,library_id,pub_id)       
+);
+create index library_interaction_idx1 on library_interaction (interaction_id);
+create index library_interaction_idx2 on library_interaction (library_id);
+create index library_interaction_idx3 on library_interaction (pub_id);
+
+grant all on library_interaction to public;
+grant all on library_interaction_library_interaction_id_seq to public;
+
+-- ================================================
+-- TABLE: feature_interaction
+-- ================================================
+
+drop table feature_interaction cascade;
+create table feature_interaction (
+       feature_interaction_id serial not null,
+       primary key (feature_interaction_id),
+       feature_id int not null,
+       foreign key (feature_id) references feature (feature_id) on delete cascade INITIALLY DEFERRED,
+       interaction_id int not null,
+       foreign key (interaction_id) references interaction (interaction_id) on delete cascade INITIALLY DEFERRED,
+       role_id int not null,
+       foreign key (role_id) references cvterm (cvterm_id) on delete cascade INITIALLY DEFERRED,
+       rank int not null default 0,
+       unique(feature_id,interaction_id, role_id)
+);
+create index feature_interaction_idx1 on feature_interaction (feature_id);
+create index feature_interaction_idx2 on feature_interaction (interaction_id);
+create index feature_interaction_idx3 on feature_interaction (role_id);
+
+grant all on feature_interaction to public;
+grant all on feature_interaction_feature_interaction_id_seq to public;
+
+-- ? do we want to add rank to the unique key ? thinking stochiometry issues
+-- and might we have one form modified and not another ? may be too much
+
+-- ================================================
+-- TABLE: feature_interactionprop
+-- ================================================
+
+drop table feature_interactionprop cascade;
+create table feature_interactionprop (
+       feature_interactionprop_id serial not null,
+       primary key (feature_interactionprop_id),
+       feature_interaction_id int not null,
+       foreign key (feature_interaction_id) references feature_interaction (feature_interaction_id) on delete cascade INITIALLY DEFERRED,
+       type_id int not null,
+       foreign key (type_id) references cvterm (cvterm_id) on delete cascade INITIALLY DEFERRED,
+       value text null,
+       rank int not null default 0,
+       constraint feature_interactionprop_c1 unique (feature_interaction_id,type_id,rank)
+);
+create index feature_interactionprop_idx1 on feature_interactionprop (feature_interaction_id);
+create index feature_interactionprop_idx2 on feature_interactionprop (type_id);
+
+grant all on feature_interactionprop to public;
+grant all on feature_interactionprop_feature_interactionprop_id_seq to public;
+
+-- ================================================
+-- TABLE: feature_interaction_pub
+-- ================================================
+
+drop table feature_interaction_pub cascade;
+create table feature_interaction_pub (
+       feature_interaction_pub_id serial not null,
+       primary key (feature_interaction_pub_id),
+       feature_interaction_id int not null,
+       foreign key (feature_interaction_id) references feature_interaction (feature_interaction_id) on delete cascade INITIALLY DEFERRED,
+       pub_id int not null,
+       foreign key (pub_id) references pub (pub_id) on delete cascade INITIALLY DEFERRED,
+       constraint feature_interaction_pub_c1 unique(feature_interaction_id,pub_id)       
+);
+create index feature_interaction_pub_idx1 on feature_interaction_pub (feature_interaction_id);
+create index feature_interaction_pub_idx2 on feature_interaction_pub (pub_id);
+
+grant all on feature_interaction_pub to public;
+grant all on feature_interaction_pub_feature_interaction_pub_id_seq to public;
+
+-- ================================================
+-- TABLE: interaction_cell_line
+-- ================================================
+
+drop table interaction_cell_line cascade;
+
+create table interaction_cell_line (
+       interaction_cell_line_id serial not null,
+       primary key (interaction_cell_line_id),
+       cell_line_id int not null,
+       foreign key (cell_line_id) references cell_line (cell_line_id) on delete cascade INITIALLY DEFERRED,
+       interaction_id int not null,
+       foreign key (interaction_id) references interaction (interaction_id) on delete cascade INITIALLY DEFERRED,
+       pub_id int not null,
+       foreign key (pub_id) references pub (pub_id) on delete cascade INITIALLY DEFERRED,
+       unique(cell_line_id,interaction_id,pub_id)       
+);
+create index interaction_cell_line_idx1 on interaction_cell_line (cell_line_id);
+create index interaction_cell_line_idx2 on interaction_cell_line (interaction_id);
+create index interaction_cell_line_idx3 on interaction_cell_line (pub_id);
+
+grant all on interaction_cell_line to public;
+grant all on interaction_cell_line_interaction_cell_line_id_seq to public;

This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

[Gmod-schema-cmts] SF.net SVN: gmod:[25261] schema/trunk/chado/lib/Bio/Chado/Builder.pm

[<scottcain@us...>]

Revision: 25261
          http://gmod.svn.sourceforge.net/gmod/?rev=25261&view=rev
Author:   scottcain
Date:     2012-03-14 02:33:59 +0000 (Wed, 14 Mar 2012)
Log Message:
-----------
quoting the password so special characters will be protected.  Thanks to Hans Kraus for pointing it out.

Modified Paths:
--------------
    schema/trunk/chado/lib/Bio/Chado/Builder.pm

Modified: schema/trunk/chado/lib/Bio/Chado/Builder.pm
===================================================================
--- schema/trunk/chado/lib/Bio/Chado/Builder.pm	2012-03-13 18:41:31 UTC (rev 25260)
+++ schema/trunk/chado/lib/Bio/Chado/Builder.pm	2012-03-14 02:33:59 UTC (rev 25261)
@@ -352,7 +352,7 @@
       # loading chadoxml
       my $stag_string = "stag-storenode.pl -d 'dbi:Pg:dbname=$db_name;host=$db_host;port=$db_port'";
       $stag_string .= " --user $db_user " if $db_user;
-      $stag_string .= " --password $db_pass " if $db_pass;
+      $stag_string .= " --password '$db_pass' " if $db_pass;
       $sys_call = join( ' ',
         $stag_string,
         catfile( $conf->{'path'}{'data'}, $file->{'local'}.'xml')

This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

[Gmod-schema-cmts] SF.net SVN: gmod:[25260] schema/trunk/chado/Changes

[<scottcain@us...>]

Revision: 25260
          http://gmod.svn.sourceforge.net/gmod/?rev=25260&view=rev
Author:   scottcain
Date:     2012-03-13 18:41:31 +0000 (Tue, 13 Mar 2012)
Log Message:
-----------
updating changes

Modified Paths:
--------------
    schema/trunk/chado/Changes

Modified: schema/trunk/chado/Changes
===================================================================
--- schema/trunk/chado/Changes	2012-03-13 18:26:05 UTC (rev 25259)
+++ schema/trunk/chado/Changes	2012-03-13 18:41:31 UTC (rev 25260)
@@ -3,6 +3,8 @@
 Bio::DB::Das::Chado it would be happy.
 * Changed all perl script #! lines to use /usr/bin/env perl instead of the
 mishmash that was being used in various scripts.
+* Fixed the GFF3 preprocessor so that if an insert into the sorting table fails,
+the script will die to avoid losing GFF lines in the resulting file.
 
 Version 1.22 Thu Nov 17 12:48:10 EST 2011
 * Created a script to automatically create the schemas dirs for a release and

This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

[Gmod-schema-cmts] SF.net SVN: gmod:[25259] schema/trunk/chado/lib/Bio/GMOD/DB/Adapter.pm

[<scottcain@us...>]

Revision: 25259
          http://gmod.svn.sourceforge.net/gmod/?rev=25259&view=rev
Author:   scottcain
Date:     2012-03-13 18:26:05 +0000 (Tue, 13 Mar 2012)
Log Message:
-----------
making the GFF sorter die if it fails to insert a line (so it won't be lost from the resulting sorted GFF).

Modified Paths:
--------------
    schema/trunk/chado/lib/Bio/GMOD/DB/Adapter.pm

Modified: schema/trunk/chado/lib/Bio/GMOD/DB/Adapter.pm
===================================================================
--- schema/trunk/chado/lib/Bio/GMOD/DB/Adapter.pm	2012-03-09 20:29:01 UTC (rev 25258)
+++ schema/trunk/chado/lib/Bio/GMOD/DB/Adapter.pm	2012-03-13 18:26:05 UTC (rev 25259)
@@ -4688,7 +4688,7 @@
     refseq   varchar(4000),
     id       varchar(4000),
     parent   varchar(4000),
-    gffline  varchar(4000),
+    gffline  varchar(8000),
     row_id   serial not null,
     primary key(row_id)
     ) "); 
@@ -4720,7 +4720,8 @@
     my $self = shift;
     my ($refseq, $id, $parent, $line) = @_;
     $self->{'queries'}{'insert_gff_sort_tmp'}->execute(
-                                               $refseq, $id, $parent, $line);
+                                               $refseq, $id, $parent, $line)
+        or die "Inserting into the sort table failed:$!,\nProbably due to this line: $line\n";
     return;
 }
 

This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

[Gmod-schema-cmts] SF.net SVN: gmod:[25258] schema/trunk/chado/bin/gmod_make_cvtermpath.pl

[<nm249@us...>]

Revision: 25258
          http://gmod.svn.sourceforge.net/gmod/?rev=25258&view=rev
Author:   nm249
Date:     2012-03-09 20:29:01 +0000 (Fri, 09 Mar 2012)
Log Message:
-----------
wrapped code in a transaction 

Modified Paths:
--------------
    schema/trunk/chado/bin/gmod_make_cvtermpath.pl

Modified: schema/trunk/chado/bin/gmod_make_cvtermpath.pl
===================================================================
--- schema/trunk/chado/bin/gmod_make_cvtermpath.pl	2012-03-09 18:51:04 UTC (rev 25257)
+++ schema/trunk/chado/bin/gmod_make_cvtermpath.pl	2012-03-09 20:29:01 UTC (rev 25258)
@@ -99,6 +99,7 @@
 use Bio::GMOD::Config;
 use Bio::GMOD::DB::Config;
 use Getopt::Std;
+use Try::Tiny;
 
 our ($opt_H, $opt_D, $opt_v, $opt_t,  $opt_g, $opt_p, $opt_d, $opt_u, $opt_c, $opt_o);
 
@@ -146,7 +147,7 @@
 $dsn .= ";host=$dbhost";
 $dsn .= ";port=$port";
 
-my $schema= Bio::Chado::Schema->connect($dsn, $user, $pass||'', { AutoCommit=>0 });
+my $schema= Bio::Chado::Schema->connect($dsn, $user, $pass||'');
 
 my $db=$schema->storage->dbh();
 
@@ -219,9 +220,9 @@
 %sot = ();
 
 # this is a hash for storing the already-processed leaves for a given type term.
-our %seen ; 
+our %seen ;
 
- 
+
 while(keys %leaf){
     foreach my $l (keys %leaf){
 	foreach my $type (keys %type){
@@ -256,107 +257,106 @@
 
 sub recurse {
   my($subjects,$type,$dist) = @_;
-  
+
   # start with the last subject
   my $subject = $subjects->[-1];
   #get the parents for the subject with this type (defaults to IS_A)
   my @objects = objects($subject,$type);
-  
+
   #if there are no parents for this path, exit the loop (and the next leaf will be sent here again)
   if(!@objects){
       $leaf{$subject}++ ;
       return;
   }
   my $path;
-  
+
   # foreach parent construct a path with each child
   foreach my $object (@objects){
-      my $tdist = $dist;
-      # loop through the child terms
-      foreach my $s (@$subjects){
-	  #next if the path was seen (subject-object-type-distance)
-	  next if $sot{$s}{$object}{$type}{$tdist};
-	  if (exists($sot{$s}{$object}) && exists($sot{$object}{$s})) { 
-	      die " YOU HAVE A CYCLE IN YOUR ONTOLOGY for $s, $object ($type, $tdist)    C8-( \n" ;
-	  }
-	  $sot{$s}{$object}{$type}{$tdist}++;
-	  
-	  print $tdist,"\t"x$dist,"\t",$s,"\t" , $object,"\t" ,$type||'transitive',"\n";
-	  
-	  # if type is defined , create a path using it (see the first looping through %leaf keys) 
-	  if($type){
-	      
-	      $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
-		  {
-		      subject_id => $s,
-		      object_id  => $object,
-		      type_id    => $type,
-		      cv_id      => $cv_id,
-		      pathdistance => $tdist
-		  }, { key => 'cvtermpath_c1' } , );
-	      message( "Inserting ($s,$object,$type,$cv_id , $tdist) into cvtermpath...path_id = " . $path->cvtermpath_id(). "\n" );
-	      my $ttdist = -1 * $tdist;
+      my $coderef = sub {
+          my $tdist = $dist;
+          # loop through the child terms
+          foreach my $s (@$subjects){
+              #next if the path was seen (subject-object-type-distance)
+              next if $sot{$s}{$object}{$type}{$tdist};
+              if (exists($sot{$s}{$object}) && exists($sot{$object}{$s})) { 
+                  die " YOU HAVE A CYCLE IN YOUR ONTOLOGY for $s, $object ($type, $tdist)    C8-( \n" ;
+              }
+              $sot{$s}{$object}{$type}{$tdist}++;
+              print $tdist,"\t"x$dist,"\t",$s,"\t" , $object,"\t" ,$type||'transitive',"\n";
 
-	      $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
-		  {
-		      subject_id => $object,
-		      object_id  => $subject,
-		      type_id    => $type,
-		      cv_id      => $cv_id,
-		      pathdistance => $ttdist
-		  }, { key => 'cvtermpath_c1' } , );
-	      message( "Inserting ($object,$subject,$type,$cv_id , $ttdist) into cvtermpath...path_id = " . $path->cvtermpath_id() . "\n" );
-	  } else {  # if type exists (see second looping through %leaf keys) create a path using the is_a type
-              message("No type defined! Using default IS_A relationship\n");
-	      my $is_a = $schema->resultset("Cv::Cvterm")->search({ name => 'is_a' })->first();
+              # if type is defined , create a path using it (see the first looping through %leaf keys) 
+              if($type){
+                  $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
+                      {
+                          subject_id => $s,
+                          object_id  => $object,
+                          type_id    => $type,
+                          cv_id      => $cv_id,
+                          pathdistance => $tdist
+                      }, { key => 'cvtermpath_c1' } , );
+                  message( "Inserting ($s,$object,$type,$cv_id , $tdist) into cvtermpath...path_id = " . $path->cvtermpath_id(). "\n" );
+                  my $ttdist = -1 * $tdist;
 
-	      $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
-		  {
-		      subject_id => $s,
-		      object_id  => $object,
-		      type_id    => $is_a->cvterm_id(),
-		      cv_id      => $cv_id,
-		      pathdistance => $tdist
-		  }, { key => 'cvtermpath_c1' } , );
-	      message("Inserting ($s,$object, $type, " . $is_a->cv_id() . "  , $tdist) into cvtermpath...path_id = " . $path->cvtermpath_id() . "\n" );
+                  $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
+                      {
+                          subject_id => $object,
+                          object_id  => $subject,
+                          type_id    => $type,
+                          cv_id      => $cv_id,
+                          pathdistance => $ttdist
+                      }, { key => 'cvtermpath_c1' } , );
+                  message( "Inserting ($object,$subject,$type,$cv_id , $ttdist) into cvtermpath...path_id = " . $path->cvtermpath_id() . "\n" );
+              } else {  # if type exists (see second looping through %leaf keys) create a path using the is_a type
+                  message("No type defined! Using default IS_A relationship\n");
+                  my $is_a = $schema->resultset("Cv::Cvterm")->search({ name => 'is_a' })->first();
 
-	      $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
-		  {
-		      subject_id => $object,
-		      object_id  => $subject,
-		      type_id    => $is_a->cvterm_id(),
-		      cv_id      => $cv_id,
-		      pathdistance => -$tdist
-		  }, { key => 'cvtermpath_c1' } , );
-	      message( "Inserting ($object,$subject, " . $is_a->cvterm_id() . " ,$cv_id , -$tdist) into cvtermpath... path_id = " . $path->cvtermpath_id() . "\n" );
-	  }
-	  $tdist--;
-      }
-      $tdist = $dist;
-      # recurse with arrayref of subjects and the object, increment the pathdistance
-      recurse([@$subjects,$object],$type,$dist+1);
-      
-      $db->commit();
-  }
-  
-}
+                  $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
+                      {
+                          subject_id => $s,
+                          object_id  => $object,
+                          type_id    => $is_a->cvterm_id(),
+                          cv_id      => $cv_id,
+                          pathdistance => $tdist
+                      }, { key => 'cvtermpath_c1' } , );
+                  message("Inserting ($s,$object, $type, " . $is_a->cv_id() . "  , $tdist) into cvtermpath...path_id = " . $path->cvtermpath_id() . "\n" );
 
+                  $path = $schema->resultset("Cv::Cvtermpath")->find_or_create( 
+                      {
+                          subject_id => $object,
+                          object_id  => $subject,
+                          type_id    => $is_a->cvterm_id(),
+                          cv_id      => $cv_id,
+                          pathdistance => -$tdist
+                      }, { key => 'cvtermpath_c1' } , );
+                  message( "Inserting ($object,$subject, " . $is_a->cvterm_id() . " ,$cv_id , -$tdist) into cvtermpath... path_id = " . $path->cvtermpath_id() . "\n" );
+              }
+              $tdist--;
+          }
+          $tdist = $dist;
+          # recurse with arrayref of subjects and the object, increment the pathdistance
+          recurse([@$subjects,$object],$type,$dist+1);
+      };
+      try {
+          $schema->txn_do($coderef);
+      } catch {
+          die "An error occured. Rolling back! " . $_ . "\n\n";
+      };
+  } #object
+} #recurse
+
 #-------------------
 
 sub objects {
   my($subject,$type) = @_;
-  
   my @cvterm_rel;
   if($type){
-            
+
       @cvterm_rel = $schema->resultset("Cv::CvtermRelationship")->search(
 	  { subject_id  => $subject,
 	    type_id     => $type ,
 	  }
 	  );
-      
   } else {
-      
       @cvterm_rel = $schema->resultset("Cv::CvtermRelationship")->search(
 	  { subject_id  => $subject }
 	  );
@@ -368,7 +368,6 @@
 
 sub subjects {
   my($object,$type) = @_;
-  
   my @cvterm_rel;
   if($type){
 
@@ -379,7 +378,6 @@
 	  );
 
   } else {
-      
       @cvterm_rel = $schema->resultset("Cv::CvtermRelationship")->search(
 	  { object_id  => $object }
 	  );

This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

[Gmod-schema-cmts] SF.net SVN: gmod:[25257] schema/trunk/chado/bin/gmod_load_cvterms.pl

[<nm249@us...>]

Revision: 25257
          http://gmod.svn.sourceforge.net/gmod/?rev=25257&view=rev
Author:   nm249
Date:     2012-03-09 18:51:04 +0000 (Fri, 09 Mar 2012)
Log Message:
-----------
disabling loading of cross-ontologies relationships. Temporary solution for loading PO which has cross references among anatomy and growth stages terms 

Modified Paths:
--------------
    schema/trunk/chado/bin/gmod_load_cvterms.pl

Modified: schema/trunk/chado/bin/gmod_load_cvterms.pl
===================================================================
--- schema/trunk/chado/bin/gmod_load_cvterms.pl	2012-03-06 18:54:12 UTC (rev 25256)
+++ schema/trunk/chado/bin/gmod_load_cvterms.pl	2012-03-09 18:51:04 UTC (rev 25257)
@@ -690,7 +690,7 @@
 		    next RELATIONSHIP;
 		}
 		############################################
-		push @novel_relationships, $r;
+                push @novel_relationships, $r;
 		my $predicate_term_name = $file_relationships{$r}->predicate_term()->name();
 
 		my $predicate_term;
@@ -729,7 +729,11 @@
 		}
 		if (!$opt_t) {
 		    message("Storing relationship $r. type cv_id=" . $predicate_term->cv_id() ."\n" ,1);
-		    my $new_rel = $schema->resultset('Cv::CvtermRelationship')->create(
+		    if ( $subject_term->cv_id != $object_term->cv_id ) {
+                        message("Wait!  subjcet term has cv namespace " . $subject_term->cv->name . " which is different from the namespace of the object term (" . $object_term->cv->name . "). Cross referencing relationships across namespaces is not supported (yet.. ) SKIPPING this relationship! \n");
+                        next RELATIONSHIP;
+                    }
+                    my $new_rel = $schema->resultset('Cv::CvtermRelationship')->create(
 			{ subject_id => $subject_term->cvterm_id(),
 			  object_id  => $object_term->cvterm_id(),
 			  type_id    => $predicate_term->cvterm_id(),
@@ -744,14 +748,14 @@
 	    die "TEST RUN! rolling back\n";
 	}
     };
-    
+
     try {
 	$schema->txn_do($coderef);
 	message("Committing! \n If you are using cvtermpath you should now run gmod_make_cvtermpath.pl . See the perldoc for more info. \n\n", 1);
     } catch {
 	# Transaction failed
 	die "An error occured! Rolling back! " . $_ . "\n";
-    }
+    };
 }
 
 sub recursive_children {

This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.

[Gmod-schema-cmts] SF.net SVN: gmod:[25256] schema/trunk/chado/bin/load_ncbi_taxonomy.pl

[<nm249@us...>]

Revision: 25256
          http://gmod.svn.sourceforge.net/gmod/?rev=25256&view=rev
Author:   nm249
Date:     2012-03-06 18:54:12 +0000 (Tue, 06 Mar 2012)
Log Message:
-----------
added lookup for organism by ncbi taxonomy id, since species names may change. This loader also assumes species names are unique, as used in NCBI taxonomy.

Modified Paths:
--------------
    schema/trunk/chado/bin/load_ncbi_taxonomy.pl

Modified: schema/trunk/chado/bin/load_ncbi_taxonomy.pl
===================================================================
--- schema/trunk/chado/bin/load_ncbi_taxonomy.pl	2012-02-28 21:26:20 UTC (rev 25255)
+++ schema/trunk/chado/bin/load_ncbi_taxonomy.pl	2012-03-06 18:54:12 UTC (rev 25256)
@@ -53,7 +53,7 @@
 http://www.ncbi.nlm.nih.gov/sites/entrez?db=Taxonomy 
 and search by taxid (e.g. txis4070[Subtree] )  
 
-=item -p
+=item -n
 
 phylotree name  [optional]
 
@@ -80,7 +80,7 @@
 
 driver. Override driver name in gmod_config
 
-=item -s 
+=item -p
 
 password. Override password in gmod_config
 
@@ -129,7 +129,7 @@
 
 =head1 VERISON AND DATE
 
-Version 2.0, October 2009.
+Version 2.1, March 2012.
 
 =head1 TODO
 
@@ -147,19 +147,19 @@
 use Bio::GMOD::DB::Config;
 
 use Bio::Chado::Schema;
-
+use Try::Tiny;
 use Getopt::Std;
 
-our ($opt_H, $opt_D, $opt_v, $opt_t, $opt_i, $opt_p, $opt_g, $opt_u, $opt_s, $opt_d);
+our ($opt_H, $opt_D, $opt_v, $opt_t, $opt_i, $opt_p, $opt_g, $opt_u, $opt_n, $opt_d);
 
-getopts('H:D:i:p:g:u:s:d:tv');
+getopts('H:D:i:p:g:u:n:d:tv');
 
 my $dbhost = $opt_H;
 my $dbname = $opt_D;
 my $infile = $opt_i;
-my $phylotree_name= $opt_p || 'NCBI taxonomy tree';
+my $phylotree_name= $opt_n || 'NCBI taxonomy tree';
 my $user = $opt_u;
-my $pass = $opt_s;
+my $pass = $opt_p;
 my $driver = $opt_d;
 my $port;
 
@@ -196,7 +196,7 @@
 $dsn .= ";host=$dbhost";
 $dsn .= ";port=$port" if $port;
 
-$schema= Bio::Chado::Schema->connect( $dsn, $user, $pass, { AutoCommit=>0 });
+$schema= Bio::Chado::Schema->connect( $dsn, $user, $pass||'');
 $dbh=$schema->storage->dbh();
 
 
@@ -377,7 +377,7 @@
 my %phylonode=();
 my $node_count=0;
 
-eval {
+my $coderef = sub {
     my $root_id;
     my $organism_id = $maxval{'organism'};
   NODE: foreach my $id ( keys %node ) {
@@ -417,10 +417,13 @@
       if (!$genus || !$species) { die "NO GENUS OR SPECIES FOUND FOR tax_id $genbank_taxon_accession! Check your input file! \n" ; } 
       $organism = $schema->resultset('Organism::Organism')->search(
 	  {
-	      genus   => { 'ilike'=> $genus },
 	      species => {'ilike' => $species }
-	  })->single();
-      if (!$organism) {  #create a new empty row object 
+	  })->single; # lookup is by species only . NCBI species should be unique!
+      if (!$organism) {  #maybe the organism is already loaded with the ncbi taxonomy id, but the species name has changed?
+	  my $organism_dbxref = $dbxref->organism_dbxrefs->single;
+	  $organism= $organism_dbxref->organism if $organism_dbxref;
+      }
+      if (!$organism) { #create a new empty row object 
 	  $organism = $schema->resultset('Organism::Organism')->new({});
 	  $insert=1;
       } else { $update = 1; }
@@ -433,12 +436,12 @@
       if ($update) {
 	  $organism->update();
 	  message( "*Updating organism " . $organism->get_column('organism_id') . " (species=" . $organism->species . ")\n", 1); 
-	      }
+      }
       if ($insert) {
 	  $organism->insert();
-	    message("New organism " . $organism->get_column('organism_id') . " (species=" . $organism->species . ")\n", 1); 
+	  message("New organism " . $organism->get_column('organism_id') . " (species=" . $organism->species . ")\n", 1); 
       }
-      	my $organism_id= $organism->get_column('organism_id');
+      my $organism_id= $organism->get_column('organism_id');
       
       ###########################################
       #store the organism synonyms 
@@ -541,13 +544,16 @@
 	$setright->execute($ctr++, $phylonode_id);
 	message( "Setting right index= $ctr for phylonode id $phylonode_id\n\n",1);
     }
+    if ($opt_t) { die "TEST RUN! rolling back!\n"; }
 };
+try {
+    $schema->txn_do($coderef);
+    message( "Commiting!! \n");
+    message("Inserted $node_count phylonodes. \n",1 );
 
-if ($@ || $opt_t) { 
-    $dbh->rollback();
+} catch {
+    message( "An error occured! Rolling back! \n $_ \n Resetting database sequences...\n", 1);
     
-    message( "Rolling back! \n $@\n Resetting database sequences...\n", 1);
-    
     #reset sequences
     foreach my $key ( keys %seq ) { 
 	my $value= $seq{$key};
@@ -555,14 +561,8 @@
 	if ($maxvalue) { $dbh->do("SELECT setval ('$value', $maxvalue, true)") ;  }
 	else {  $dbh->do("SELECT setval ('$value', 1, false)");  }
     }
-}else {    
-    message( "Commiting!! \n");
-    message("Inserted $node_count phylonodes. \n",1 );
-    
-    $dbh->commit(); 
-}
+};
 
-
 sub set_maxval {
     my $key=shift;
     my $id_column= $key . "_id";

This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.