diff variant_effect_predictor/Bio/EnsEMBL/Variation/Pipeline/InitVariationClass.pm @ 0:1f6dce3d34e0

Uploaded

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/variant_effect_predictor/Bio/EnsEMBL/Variation/Pipeline/InitVariationClass.pm	Thu Apr 11 02:01:53 2013 -0400
@@ -0,0 +1,109 @@
+package Bio::EnsEMBL::Variation::Pipeline::InitVariationClass;
+
+use strict;
+use warnings;
+
+use base qw(Bio::EnsEMBL::Variation::Pipeline::BaseVariationProcess);
+
+use POSIX qw(ceil);
+
+my $DEBUG = 0;
+
+sub fetch_input {
+    
+    my $self = shift;
+
+    my $num_chunks = $self->required_param('num_chunks');
+    
+    my $var_dba = $self->get_species_adaptor('variation');
+        
+    my $aa = $var_dba->get_AttributeAdaptor;
+    
+    my $dbc = $var_dba->dbc();
+    
+    # first set everything in variation (except HGMDs) to 'sequence_alteration' by default
+    # because sometimes we miss them because there is no variation_feature
+    # or any alleles (though this should become unnecessary as we move to the
+    # new approach to failing for all species)
+
+    my $default_attrib_id = $aa->attrib_id_for_type_value('SO_term', 'sequence_alteration');
+
+    die "No attrib_id for 'sequence_alteration'" unless defined $default_attrib_id;
+
+    $dbc->do(qq{
+        UPDATE  variation v, source s
+        SET     v.class_attrib_id = $default_attrib_id
+        WHERE   v.source_id = s.source_id
+        AND     s.name != 'HGMD-PUBLIC'
+    });
+    
+    # now create some temp tables to store the class attribs
+
+    my $temp_var_table = 'temp_variation_class';
+    my $temp_var_feat_table = 'temp_variation_feature_class';
+    
+    $dbc->do(qq{DROP TABLE IF EXISTS $temp_var_table});
+    $dbc->do(qq{DROP TABLE IF EXISTS $temp_var_feat_table});
+    
+    $dbc->do(qq{CREATE TABLE $temp_var_table LIKE variation});
+    $dbc->do(qq{CREATE TABLE $temp_var_feat_table LIKE variation_feature});
+
+    $dbc->do(qq{ALTER TABLE $temp_var_table DISABLE KEYS});
+    $dbc->do(qq{ALTER TABLE $temp_var_feat_table DISABLE KEYS});
+
+    # now get an ordered list of all the variation_ids
+        
+    my $get_var_ids_sth = $dbc->prepare(qq{
+        SELECT variation_id FROM variation ORDER BY variation_id
+    });
+
+    $get_var_ids_sth->execute;
+
+    my @var_ids;
+
+    while (my ($var_id) = $get_var_ids_sth->fetchrow_array) {
+        push @var_ids, $var_id;
+    }
+
+    # and split them up into as many chunks as requested
+
+    my $num_vars = scalar @var_ids;
+
+    my $chunk_size = ceil($num_vars / $num_chunks);
+
+    my @output_ids;
+
+    while (@var_ids) {
+
+        my $start = $var_ids[0];
+        my $stop  = $chunk_size <= $#var_ids ? $var_ids[$chunk_size - 1] : $var_ids[$#var_ids];
+
+        splice(@var_ids, 0, $chunk_size);
+
+        push @output_ids, {
+            variation_id_start  => $start,
+            variation_id_stop   => $stop,
+            temp_var_table      => $temp_var_table,
+            temp_var_feat_table => $temp_var_feat_table,
+        };
+    }
+
+    $self->param('chunk_output_ids', \@output_ids);
+
+    $self->param(
+        'finish_var_class', [{
+            temp_var_table      => $temp_var_table,
+            temp_var_feat_table => $temp_var_feat_table,
+        }]
+    );
+}
+
+sub write_output {
+    my $self = shift;
+
+    $self->dataflow_output_id($self->param('finish_var_class'), 1);
+    $self->dataflow_output_id($self->param('chunk_output_ids'), 2);
+}
+
+1;
+