new evidence types

Ensembl · Feb 15, 2016 · f786c45 · f786c45
1 parent bbbd341
commit f786c45
Show file tree

Hide file tree

Showing 2 changed files with 53 additions and 8 deletions.
diff --git a/modules/Bio/EnsEMBL/Variation/Pipeline/VariantQC/CheckdbSNPImport.pm b/modules/Bio/EnsEMBL/Variation/Pipeline/VariantQC/CheckdbSNPImport.pm
@@ -41,7 +41,7 @@ use strict;
 use warnings;
 
 use base qw(Bio::EnsEMBL::Variation::Pipeline::BaseVariationProcess);
-use Bio::EnsEMBL::Variation::Utils::QCUtils qw( count_rows count_for_statement);
+use Bio::EnsEMBL::Variation::Utils::QCUtils qw( count_rows count_for_statement get_evidence_attribs);
 
 my $DEBUG = 0;
 
@@ -89,6 +89,7 @@ sub run {
     my $complimented_desc    = $self->check_complimented_desc();
     my $bad_position         = $self->check_bad_position();
 
+    my $attribs_loaded       = $self->check_attribs();
 
     print $report "Post-import preQC check
 
@@ -110,6 +111,9 @@ VariationFeature without seqregion: $varfeat_no_seqreg
 VariationFeature where end+1<start: $bad_position
 \n";  
 
+    print $report "\nERROR: missing evidence attribs\n\n" if $attribs_loaded == 0;
+
+
     print $report "ERROR: $complimented_desc complimented descriptions found - to be fixed manually\n\n" if $complimented_desc >0;
 
     if($var_no_ss_allele    > 0  || 
@@ -120,7 +124,8 @@ VariationFeature where end+1<start: $bad_position
        $geno_no_sample     >0    ||
        $varfeat_no_pos     >0    ||
        $varfeat_no_seqreg  >0    ||
-       $geno_no_subsnp     >0
+       $geno_no_subsnp     >0    ||
+       $attribs_loaded   == 0
        ){
 
         print $report "Exiting - missing data to import\n"; 
@@ -348,4 +353,25 @@ sub check_bad_position{
     return count_for_statement($var_dba , $data_ext_stat);
 }
 
+=head2 check_attribs
+
+  Check the expected evidence attribs are available.
+  Check early to avoid partial jobs - not future-proof but will catch some problems
+
+
+=cut
+sub check_attribs{
+
+    my $self = shift;
+
+    my $var_dba = $self->get_species_adaptor('variation');
+    my $attribs = get_evidence_attribs($var_dba);
+
+    my $found_everything = 1;
+    foreach my $ev ( "1000Genomes", "Cited", "ESP", "ExAC", "Frequency", "HapMap", "Multiple_observations","1000Bull_Genomes", "WTSI_MGP"){
+        $found_everything = 0 unless defined $attribs->{$ev};
+    }
+    return $found_everything;
+}
+
 1;
diff --git a/modules/Bio/EnsEMBL/Variation/Utils/QCUtils.pm b/modules/Bio/EnsEMBL/Variation/Utils/QCUtils.pm
@@ -274,7 +274,7 @@ sub summarise_evidence{
 
 
     ## summarise ss information
-    my $ss_variations =  get_ss_variations($var_dbh, $first, $last);
+    my $ss_variations =  get_ss_variations($var_dbh, $first, $last, $species);
 
 
     ## extract list of variants with pubmed citations
@@ -291,7 +291,8 @@ sub summarise_evidence{
 
 	## dbSNP ss submissions
 	push @{$evidence{$var}},  $evidence_ids->{Multiple_observations}  
-	   if defined $ss_variations->{$var}->{count} && $ss_variations->{$var}->{count} > 1;
+	   if defined $ss_variations->{$var}->{count} && $ss_variations->{$var}->{count} > 1
+           &&  $species !~/Homo|Human/i ;
 
 	push @{$evidence{$var}}, $evidence_ids->{Frequency}            
            if defined $ss_variations->{$var}->{'freq'};
@@ -311,7 +312,16 @@ sub summarise_evidence{
       push @{$evidence{$var}}, $evidence_ids->{'1000Genomes'}
            if (defined $kg_variations->{$var} ||  defined $ss_variations->{$var}->{'KG'}) ;
 
-      ## pubmed citations
+      ## additional cow evidence
+      push @{$evidence{$var}}, $evidence_ids->{'1000Bull_Genomes'}
+           if (defined $kg_variations->{$var} ||  defined $ss_variations->{$var}->{'1000_BULL_GENOMES'}) ;
+
+      ## additional mouse evidence
+      push @{$evidence{$var}}, $evidence_ids->{'WTSI_MGP'}
+           if (defined $kg_variations->{$var} ||  defined $ss_variations->{$var}->{'SC_MOUSE_GENOMES'}) ;
+
+
+      ## pubmed citations - multi species
       push @{$evidence{$var}}, $evidence_ids->{Cited}               
            if defined $pubmed_variations->{$var}; 
 
@@ -363,7 +373,8 @@ sub get_ss_variations{
     my $var_dbh = shift;
     my $first   = shift;
     my $last    = shift;
-
+    my $species = shift;
+
     my %evidence;
 
     my $obs_var_ext_sth  = $var_dbh->prepare(qq[ select al.variation_id,
@@ -389,12 +400,20 @@ sub get_ss_variations{
 
         $l->[2] = "N" unless defined $l->[2];
 
+        ## human specific
         $evidence{$l->[0]}{'KG'}    = 1 if $l->[1] =~/1000GENOMES/;
         $evidence{$l->[0]}{'freq'}  = 1 if $l->[1] =~/1000GENOMES/;
 
-        #save  submitter handle, population and ss id to try to discern independent submissions
-        push  @{$save_by_var{$l->[0]}}, [  $l->[1], $l->[2], $l->[5] ];
+        ## cow specific
+        $evidence{$l->[0]}{'1000_BULL_GENOMES'} = 1 if $l->[1] =~/1000_BULL_GENOMES/;
+        ## mouse specific
+        $evidence{$l->[0]}{'SC_MOUSE_GENOMES'}  = 1 if $l->[1] =~/SC_MOUSE_GENOMES/;
 
+        #save submitter handle, population and ss id to try to discern independent submissions
+        ## no longer useful for Human 2016/01 
+        unless ($species =~/Homo|Human/i){
+          push  @{$save_by_var{$l->[0]}}, [  $l->[1], $l->[2], $l->[5] ];
+        }
 
        ## Save frequency evidence for variant by variant id - ensure at least 2 chromosomes assayed and variant poly
        ## only assign frequency status if population has more than 1 member.