diff --git a/modules/Bio/EnsEMBL/IdMapping/ExonScoreBuilder.pm b/modules/Bio/EnsEMBL/IdMapping/ExonScoreBuilder.pm
index c3908f6a74b100c15878a937f08f90c6c93b57c7..4101c280d9c0d5766ca5d212b2dbbb8ac5d2512d 100644
--- a/modules/Bio/EnsEMBL/IdMapping/ExonScoreBuilder.pm
+++ b/modules/Bio/EnsEMBL/IdMapping/ExonScoreBuilder.pm
@@ -678,6 +678,34 @@ sub non_mapped_transcript_rescore {
my @target_transcripts = @{ $self->cache->get_by_key(
'transcripts_by_exon_id', 'target', $entry->target) };
+ # EG reworking of logic to allow no source/target e.g. for new
+ # species in multispecies databases
+ my $st =
+ $self->cache()
+ ->get_by_key( 'transcripts_by_exon_id', 'source',
+ $entry->source() );
+ my @source_transcripts;
+ if ( !defined($st) ) {
+ $self->logger->warning(
+ "Can't find source transcipts by exon_id for "
+ . $entry->source() );
+ } else {
+ @source_transcripts = @{$st};
+ }
+
+ my $tt =
+ $self->cache()
+ ->get_by_key( 'transcripts_by_exon_id', 'target',
+ $entry->target() );
+ my @target_transcripts = ();
+ if ( !defined($tt) ) {
+ $self->logger->warning(
+ "Can't find target transcipts by exon_id for "
+ . $entry->target() );
+ } else {
+ @target_transcripts = @{$tt};
+ }
+
my $found_mapped = 0;
TR:
diff --git a/modules/Bio/EnsEMBL/IdMapping/ResultAnalyser.pm b/modules/Bio/EnsEMBL/IdMapping/ResultAnalyser.pm
index b6e7259c1e0f208a848f08946c9b6d3598338674..2c3476402967fd7a854720a003b475991e4d4222 100644
--- a/modules/Bio/EnsEMBL/IdMapping/ResultAnalyser.pm
+++ b/modules/Bio/EnsEMBL/IdMapping/ResultAnalyser.pm
@@ -733,28 +733,33 @@ sub create_summary_email {
print $fh "\n";
- #
- # clicklist of first 10 deleted genes
- #
- print $fh qq(\nFirst 10 deleted known genes:\n);
- print $fh qq(=============================\n\n);
+ # EG genes_lost.txt file may not exist if species is new
+ if ( $self->file_exists( 'genes_lost.txt', 'debug' ) ) {
+ #
+ # clicklist of first 10 deleted genes
+ #
- my $in_fh = $self->get_filehandle( 'genes_lost.txt', 'debug', '<' );
- my $prefix = $self->conf->param('urlprefix');
- my $i;
+ print $fh qq(\nFirst 10 deleted known genes:\n);
+ print $fh qq(=============================\n\n);
- while (<$in_fh>) {
- last if ( ++$i > 10 );
+ my $in_fh = $self->get_filehandle( 'genes_lost.txt', 'debug', '<' );
+ my $prefix = $self->conf->param('urlprefix');
+ my $i;
- chomp;
- my ( $stable_id, $type ) = split(/\s+/);
+ while (<$in_fh>) {
+ last if ( ++$i > 10 );
- next unless ( $type eq 'known' );
+ chomp;
+ my ( $stable_id, $type ) = split(/\s+/);
- print $fh sprintf( $fmt2, $stable_id, "${prefix}$stable_id" );
- }
+ next unless ( $type eq 'known' );
+
+ print $fh sprintf( $fmt2, $stable_id, "${prefix}$stable_id" );
+ }
+
+ close($in_fh);
+ } ## end if ( $self->file_exists...)
- close($in_fh);
close($fh);
}
diff --git a/modules/Bio/EnsEMBL/IdMapping/StableIdGenerator/EnsemblGeneric.pm b/modules/Bio/EnsEMBL/IdMapping/StableIdGenerator/EnsemblGeneric.pm
index aa475697813a2813afc111edcd42afe08dbd6b86..2ba059869b0ce45c317ecd4d6cd2e525a23bca12 100644
--- a/modules/Bio/EnsEMBL/IdMapping/StableIdGenerator/EnsemblGeneric.pm
+++ b/modules/Bio/EnsEMBL/IdMapping/StableIdGenerator/EnsemblGeneric.pm
@@ -120,30 +120,41 @@ sub initial_stable_id {
my $self = shift;
my $type = shift;
- my $init_stable_id;
-
- # use stable ID from configuration if set
- if ($init_stable_id = $self->conf->param("starting_${type}_stable_id")) {
- $self->logger->debug("Using pre-configured $init_stable_id as base for new $type stable IDs.\n");
- return $init_stable_id;
- }
-
- my $s_dba = $self->cache->get_DBAdaptor('source');
- my $s_dbh = $s_dba->dbc->db_handle;
-
- # look in the ${type}_stable_id table first
- my $sql = qq(SELECT MAX(stable_id) FROM ${type}_stable_id);
- $init_stable_id = $self->fetch_value_from_db($s_dbh, $sql);
+ # EG modifications to permit the current stable ID to persist trohough
+ # different invocations
+ my $init_stable_id = $self->{stable_id_list}{$type};
+
+ if ( !defined($init_stable_id) ) {
+ # use stable ID from configuration if set
+ if ( $init_stable_id =
+ $self->conf->param("starting_${type}_stable_id") )
+ {
+ $self->logger->debug( "Using pre-configured $init_stable_id "
+ . "as base for new $type stable IDs.\n" );
+ return $init_stable_id;
+ }
- # also look in gene_archive to make sure there are no larger Ids there
- unless ($type eq 'exon') {
- $sql = qq(SELECT MAX(${type}_stable_id) FROM gene_archive);
- my $archived_stable_id = $self->fetch_value_from_db($s_dbh, $sql);
- if ($archived_stable_id and $self->is_valid($archived_stable_id) and
- ($archived_stable_id gt $init_stable_id)) {
- $init_stable_id = $archived_stable_id;
+ my $s_dba = $self->cache->get_DBAdaptor('source');
+ my $s_dbh = $s_dba->dbc->db_handle;
+
+ # look in the ${type}_stable_id table first
+ my $sql = qq(SELECT MAX(stable_id) FROM ${type}_stable_id);
+ $init_stable_id = $self->fetch_value_from_db( $s_dbh, $sql );
+
+ # also look in gene_archive to make sure there are no larger Ids
+ # there
+ unless ( $type eq 'exon' ) {
+ $sql = qq(SELECT MAX(${type}_stable_id) FROM gene_archive);
+ my $archived_stable_id =
+ $self->fetch_value_from_db( $s_dbh, $sql );
+ if ( $archived_stable_id
+ and $self->is_valid($archived_stable_id)
+ and ( $archived_stable_id gt $init_stable_id ) )
+ {
+ $init_stable_id = $archived_stable_id;
+ }
}
- }
+ } ## end if ( !defined($init_stable_id...))
if ($init_stable_id) {
# since $init_stable_id now is the highest existing stable Id for this
diff --git a/modules/Bio/EnsEMBL/IdMapping/StableIdMapper.pm b/modules/Bio/EnsEMBL/IdMapping/StableIdMapper.pm
index 234e6bf7521220ac74a7d9bc55202397f64035c1..5d9fda6926da490e24da7b3c1aa16d0b99d06975 100644
--- a/modules/Bio/EnsEMBL/IdMapping/StableIdMapper.pm
+++ b/modules/Bio/EnsEMBL/IdMapping/StableIdMapper.pm
@@ -187,9 +187,10 @@ sub map_stable_ids {
# check if there are any objects of this type at all
my %all_sources = %{ $self->cache->get_by_name("${type}s_by_id", 'source') };
my %all_targets = %{ $self->cache->get_by_name("${type}s_by_id", 'target') };
- unless (scalar(keys %all_sources)) {
- $self->logger->info("No cached ${type}s found.\n\n");
- return;
+ if ( scalar( keys(%all_sources) ) == 0 ) {
+ # EG may be possible to have no sources for new species
+ $self->logger->warning("No cached ${type}s found.\n\n");
+ %all_sources = ();
}
my %stats = map { $_ => 0 }
@@ -637,24 +638,31 @@ sub generate_mapping_stats {
my $novel_total = $stats->{'mapped_novel'} + $stats->{'lost_novel'};
# no split into known and novel for exons
- unless ( $type eq 'exon' ) {
- $result .= sprintf( $fmt2,
- 'known',
- $stats->{'mapped_known'},
- $stats->{'lost_known'},
- ($known_total ? $stats->{'mapped_known'}/$known_total*100 : 0)
- );
+ if ( $type ne 'exon' ) {
+ $result .= sprintf( $fmt2, 'known',
+ $stats->{'mapped_known'},
+ $stats->{'lost_known'}, (
+ $known_total
+ ? $stats->{'mapped_known'}/$known_total*100
+ : 0 ) );
+
+ $result .= sprintf( $fmt2, 'novel',
+ $stats->{'mapped_novel'},
+ $stats->{'lost_novel'}, (
+ $novel_total
+ ? $stats->{'mapped_novel'}/$novel_total*100
+ : 0 ) );
+ }
+ if ( $mapped_total == 0 ) {
+ # EG different calculation needed when no mappings found for new
+ # species
+ $result .= sprintf( $fmt2, 'total', $mapped_total, $lost_total, 0 );
+ } else {
$result .= sprintf( $fmt2,
- 'novel',
- $stats->{'mapped_novel'},
- $stats->{'lost_novel'},
- ($novel_total ? $stats->{'mapped_novel'}/$novel_total*100 : 0)
- );
- } ## end unless ( $type eq 'exon' )
-
- $result .= sprintf($fmt2, 'total', $mapped_total, $lost_total,
- $mapped_total/($known_total + $novel_total)*100);
+ 'total', $mapped_total, $lost_total,
+ $mapped_total/( $known_total + $novel_total )*100 );
+ }
# log result
$self->logger->info($result."\n");
diff --git a/modules/Bio/EnsEMBL/IdMapping/SyntenyFramework.pm b/modules/Bio/EnsEMBL/IdMapping/SyntenyFramework.pm
index dff799eea755adf3017836e6c67779acfe6f92f6..a3d7fcf12617167a075f3c1383c2cab34432e422 100644
--- a/modules/Bio/EnsEMBL/IdMapping/SyntenyFramework.pm
+++ b/modules/Bio/EnsEMBL/IdMapping/SyntenyFramework.pm
@@ -376,12 +376,15 @@ sub rescore_gene_matrix_lsf {
is_component => 1,
);
- my $cmd = qq{$Bin/synteny_rescore.pl $options --index \$LSB_JOBINDEX};
-
- my $pipe = qq{|bsub -J$lsf_name\[1-$num_jobs\] } .
- qq{-o $logpath/synteny_rescore.\%I.out } .
- qq{-e $logpath/synteny_rescore.\%I.err } .
- $self->conf->param('lsf_opt_synteny_rescore');
+ my $cmd = qq{perl -I./modules $Bin/synteny_rescore.pl }
+ . qq{$options --index \$LSB_JOBINDEX};
+
+ my $pipe =
+ qq{|bsub -J$lsf_name\[1-$num_jobs\] }
+ . qq{-o $logpath/synteny_rescore.\%I.out }
+ . qq{-e $logpath/synteny_rescore.\%I.err }
+ . $self->conf()->param('lsf_opt_run')
+ . $self->conf()->param('lsf_opt_synteny_rescore');
# run lsf job array
$self->logger->info("Submitting $num_jobs jobs to lsf.\n");