Merge pull request #498 from Ensembl/release/107

Merged RR updates to 108
Ensembl · Sep 28, 2022 · e4ef765 · e4ef765
2 parents 16ba0b2 + afc3e55
commit e4ef765
Show file tree

Hide file tree

Showing 3 changed files with 42 additions and 9 deletions.
diff --git a/lib/Bio/EnsEMBL/DataCheck/Checks/CanonicalTranscripts.pm b/lib/Bio/EnsEMBL/DataCheck/Checks/CanonicalTranscripts.pm
@@ -66,7 +66,6 @@ sub tests {
   /;
   is_rows_zero($self->dba, $sql_2, $desc_2);
 
-  my $desc_3 = 'Canonical transcripts with translations belong to protein-coding genes';
   my $sql_3  = qq/
     SELECT g.stable_id FROM
       gene g INNER JOIN
@@ -81,7 +80,15 @@ sub tests {
       b.biotype_group <> 'coding' AND
       cs.species_id = $species_id
   /;
-  is_rows_zero($self->dba, $sql_3, $desc_3);
+  my $mca = $self->dba->get_adaptor('MetaContainer');
+  if($mca->single_value_by_key('genebuild.method') eq 'projection_build'){
+        skip "Canonical transcript checks are  not mandatory for projection builds", 1;
+	}
+	else{
+		 my $desc_3 = 'Canonical transcripts with translations belong to protein-coding genes';
+		 is_rows_zero($self->dba, $sql_3, $desc_3);
+	 }
+
 
   my $desc_4 = 'Canonical transcripts with translations are protein-coding';
   my $sql_4  = qq/

diff --git a/lib/Bio/EnsEMBL/DataCheck/Checks/GeneBiotypes.pm b/lib/Bio/EnsEMBL/DataCheck/Checks/GeneBiotypes.pm
@@ -136,14 +136,35 @@ sub biotype_groups {
     }
   }
 
-  my $desc_1 = 'genes have at least one transcript with a matching biotype group';
-  is(scalar(@group_mismatch), 0, $desc_1);
+  my $mca = $self->dba->get_adaptor('MetaContainer');
+  if($mca->single_value_by_key('genebuild.method') eq 'projection_build'){
+      if (scalar(@group_mismatch) > 100){
+	  my $desc_proj_1 = 'found no more that 100 genes with mismatched transcript biotype groups';
+	  is(scalar(@group_mismatch), 100, $desc_proj_1);
+      }
 
-  my $desc_2 = '"pseudogene" genes do not have coding transcripts';
-  is(scalar(@pseudogene_mismatch), 0, $desc_2);
+     if (scalar(@pseudogene_mismatch) > 5){
+        my $desc_proj_2 = 'found no more that 5 pseudogenes that have coding transcripts';
+        is(scalar(@polymorphic_mismatch), 5, $desc_proj_2);
+      }
+
+     if (scalar(@polymorphic_mismatch) > 5){
+        my $desc_proj_3 = 'found no more that 5 polymorphic_pseudogene with mismatched transcript biotype groups';
+        is(scalar(@polymorphic_mismatch), 5, $desc_proj_3);
+      }
+  }
+  else{
 
-  my $desc_3 = '"polymorphic_pseudogene" genes have at least one polymorphic_pseudogene transcript';
-  is(scalar(@polymorphic_mismatch), 0, $desc_3);
-}
+      my $desc_1 = 'genes have at least one transcript with a matching biotype group';
+      is(scalar(@group_mismatch), 0, $desc_1);
 
+      my $desc_2 = '"pseudogene" genes do not have coding transcripts';
+      is(scalar(@pseudogene_mismatch), 0, $desc_2);
+
+      my $desc_3 = '"polymorphic_pseudogene" genes have at least one polymorphic_pseudogene transcript';
+      is(scalar(@polymorphic_mismatch), 0, $desc_3);
+  }
+
+
+}
 1;
diff --git a/lib/Bio/EnsEMBL/DataCheck/Checks/RepeatFeatures.pm b/lib/Bio/EnsEMBL/DataCheck/Checks/RepeatFeatures.pm
@@ -49,6 +49,11 @@ sub tests {
       skip "Repeat features not mandatory for viruses", 1;
     }
 
+    if($mca->single_value_by_key('genebuild.method') eq 'projection_build') {
+      skip "Repeat features not mandatory for projection builds", 1;
+    }
+
+
     # Don't need to worry about species_id, because we don't do this
     # query for collection dbs...
     my @logic_names = ('dust', 'trf');