sort out synonym and deleted accession entrys

misc-scripts/xref_mapping/compare_mapping.pl

@@ -6,22 +6,42 @@
 # to dump this from an ensembl datbase use something like
 # mysql -u ensro -h ecs2 -P 3365 -e "select x.dbprimary_acc, ox.ensembl_object_type, ox.ensembl_id from xref x, object_xref ox, external_db edb where edb.db_name like 'Uniprot/SWISSPROT' and x.external_db_id=edb.external_db_id and x.xref_id=ox.xref_id order by x.xref_id" homo_sapiens_core_25_34e > ensembl_swissprot_translations.txt
 
+
+# ADDITIONALLY
+# read in the synonym conversion. This is created with something like:-
+#mysql -hecs4 -P3350 -uensadmin -pensembl -Dianl_xref_test -e"select x.accession, s.accession from synonym a, xref x, xref s where a.xref_id = x.xref_id and a.synonym_xref_id = s.xref_id " > xref_synonym.txt
+
+#ALSO get list of deleted accesions no need to look at these.
+#down load wget http://us.expasy.org/txt/delac_tr.txt
+#remove html rubbish and save.
+
+
+
 use strict;
 
 use Getopt::Long;
 
-my ( $old, $new, $pass, $port, $dbname  );
+my ( $old, $new, $del, $syn, $pass, $port, $dbname  );
+
+my %deleted;
+my %synonym;
+
 
 GetOptions( "old=s", \$old,
-	    "new=s", \$new);
+	    "new=s", \$new,
+	    "del=s", \$del,
+	    "syn=s", \$syn);
 
-if( !($old && $new) ) {
+if( !($old && $new && $syn && $del) ) {
   usage();
   exit(1);
 }
 
 # read files into xref/object hashes
 # TODO - more than one mapping?
+get_deleted($del);
+get_synonym($syn);
+
 
 my ($new_x2e_r, $new_e2x_r) = read_mappings($new);
 my ($old_x2e_r, $old_e2x_r) = read_mappings($old);
@@ -31,6 +51,38 @@ check_non_translations($new_e2x_r);
 compare($old_x2e_r, $new_x2e_r, "xref");
 compare($old_e2x_r, $new_e2x_r, "ensembl_object");
 
+
+sub get_deleted {
+
+  my ($file) = @_;
+
+  open(FILE,"<".$file) || die "Could not open $file\n";
+
+  while(<FILE>){
+    chomp;
+    $deleted{$_} = 1;
+  }
+  close FILE;
+}
+
+sub get_synonym {
+
+  my ($file) = @_;
+
+  open(FILE,"<".$file) || die "Could not open $file\n";
+
+  <FILE>; #junk first line
+
+  while(<FILE>){
+    chomp;
+    my @arr = split;
+    $synonym{$arr[1]} = $arr[0];
+  }
+  close FILE;
+}
+    
+
+
 # ----------------------------------------
 # Compare mappings keyed on xref
 
@@ -117,6 +169,7 @@ sub usage {
 print << "EOF";
 
 Usage: compare_mapping.pl -old <old mapping file> -new <new mapping file>
+       -del <deleted acc list file>  -syn <xref synonym file>
 
 Mapping files should be in the following format:
 
@@ -134,6 +187,8 @@ sub read_mappings {
 
   my %xref_to_ensembl;
   my %ensembl_to_xref;
+  my $del=0;
+  my $syn=0;
 
   open (FILE, $filename) || die "Can't read $filename\n";
   my $dummy = <FILE>; # skip first line
@@ -142,9 +197,18 @@ sub read_mappings {
 
     my ($xref_id, $type, $ensembl_id) = split;
     # TODO - better way of handling type?
-    my $value = $ensembl_id . "." . $type;
-    push @{$xref_to_ensembl{$xref_id}}, $value;
-    push @{$ensembl_to_xref{$value}}, $xref_id;
+    if(!defined($deleted{$xref_id})){
+      if(defined($synonym{$xref_id})){
+	$xref_id = $synonym{$xref_id};
+	$syn++;
+      }
+      my $value = $ensembl_id . "." . $type;
+      push @{$xref_to_ensembl{$xref_id}}, $value;
+      push @{$ensembl_to_xref{$value}}, $xref_id;
+    }
+    else{
+      $del++;
+    }
 
   }
 
@@ -156,6 +220,8 @@ sub read_mappings {
   }
 
   print "Read $i mappings from $filename\n";
+  print "$del entries ignored due to deletions\n";
+  print "$syn entries changed to synonym values\n";
 
   return (\%xref_to_ensembl, \%ensembl_to_xref);