Merge branch 'release/v3.0.0'

.gitignore

@@ -8,3 +8,4 @@
 /tabix.tar
 /tabix/
 /perltidy.LOG
+/MYMETA.*

CHANGES.md

@@ -0,0 +1,4 @@
+### 3.0.0
+* Removed use of legacy Tabix codebase and switches to samtools 1.2+
+* Switches to gihub version of vcftools and makes several patches unnecessary
+* Several bugfixes

MANIFEST

@@ -4,6 +4,7 @@ bin/Admin_EnsemblTranscriptFastaFilter.pl
 bin/Admin_GeneRegionBedDumper.pl
 bin/AnnotateVcf.pl
 bin/cpanm
+CHANGES.md
 docs.tar.gz
 lib/Sanger/CGP/Vagrent.pm
 lib/Sanger/CGP/Vagrent/Annotators/AbstractAnnotator.pm
@@ -41,8 +42,6 @@ LICENSE.TXT
 Makefile.PL
 MANIFEST			This list of files
 MANIFEST.SKIP
-patches/tabixPerlLinker.diff
-patches/vcfToolsInstLocs.diff
 patches/vcfToolsProcessLog.diff
 prerelease.sh
 README.md

Makefile.PL

@@ -38,7 +38,7 @@ WriteMakefile(
                         bin/AnnotateVcf.pl)],
   PREREQ_PM     => {
                     'Attribute::Abstract' => '1.02',
-                    'Bio::DB::Sam' => '1.39',
+                    'Bio::DB::HTS' => '1.11',
                     'Bio::Root::Version' => '1.006924',# this is the only way to get Bio::Perl to behave
                                                        # and return the version correctly
                     'Capture::Tiny' => 0.24,

README.md

@@ -1,36 +1,3 @@
-LICENCE
-=======
-
-Copyright (c) 2014 Genome Research Ltd.
-
-Author: Cancer Genome Project <[email protected]>
-
-This file is part of VAGrENT.
-
-VAGrENT is free software: you can redistribute it and/or modify it under
-the terms of the GNU Affero General Public License as published by the Free
-Software Foundation; either version 3 of the License, or (at your option) any
-later version.
-
-This program is distributed in the hope that it will be useful, but WITHOUT
-ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
-FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
-details.
-
-You should have received a copy of the GNU Affero General Public License
-along with this program. If not, see <http://www.gnu.org/licenses/>.
-
-1. The usage of a range of years within a copyright statement contained within
-this distribution should be interpreted as being equivalent to a list of years
-including the first and last year specified and all consecutive years between
-them. For example, a copyright statement that reads ‘Copyright (c) 2005, 2007-
-2009, 2011-2012’ should be interpreted as being identical to a statement that
-reads ‘Copyright (c) 2005, 2007, 2008, 2009, 2011, 2012’ and a copyright
-statement that reads ‘Copyright (c) 2005-2012’ should be interpreted as being
-identical to a statement that reads ‘Copyright (c) 2005, 2006, 2007, 2008,
-2009, 2010, 2011, 2012’."
-
-
 VAGrENT
 ======
 
@@ -45,16 +12,17 @@ as well as Sequence Ontology terms to classify its consequences.
 ---
 
 ###Dependencies/Install
-Some of the code included in this package has dependencies on several C packages:
+Some of the code included in this package has dependencies on several packages:
 
- * [Samtools](https://github.com/samtools/samtools) - max 0.1.20 until perl bindings are updated
- * [tabix](https://github.com/samtools/samtools/tabix)
- * [vcftools](http://vcftools.sourceforge.net/)
+ * [Samtools v1.3+](https://github.com/samtools/samtools)
+ * [vcftools](https://vcftools.github.io/)
+ * [Bio::DB::HTS](http://search.cpan.org/~rishidev/Bio-DB-HTS/)
 
 And various perl modules.
 
-Please use `setup.sh` to install the dependencies.  Please be aware that this expects basic C
-compilation libraries and tools to be available, most are listed in `INSTALL`.
+Please use `setup.sh` to install the dependencies.  Setting the environment variable `CGP_PERLLIBS` allows you to to append to `PERL5LIB` during install.  Without this all dependancies are installed into the target area.
+
+Please be aware that this expects basic C compilation libraries and tools to be available.
 
 ---
 
@@ -65,9 +33,41 @@ compilation libraries and tools to be available, most are listed in `INSTALL`.
 
 ####Cutting the release
 1. Update `lib/Sanger/CGP/Vagrent.pm` to the correct version.
-2. Update `Changes` to show major items.
+2. Update `CHANGES.md` to show major items.
 3. Run `./prerelease.sh`
 4. Check all tests and coverage reports are acceptable.
 5. Commit the updated docs tree and updated module/version.
 6. Push commits.
 7. Use the GitHub tools to draft a release.
+
+LICENCE
+=======
+
+Copyright (c) 2014-2016 Genome Research Ltd.
+
+Author: Cancer Genome Project <[email protected]>
+
+This file is part of VAGrENT.
+
+VAGrENT is free software: you can redistribute it and/or modify it under
+the terms of the GNU Affero General Public License as published by the Free
+Software Foundation; either version 3 of the License, or (at your option) any
+later version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT
+ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
+FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
+details.
+
+You should have received a copy of the GNU Affero General Public License
+along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+1. The usage of a range of years within a copyright statement contained within
+this distribution should be interpreted as being equivalent to a list of years
+including the first and last year specified and all consecutive years between
+them. For example, a copyright statement that reads ‘Copyright (c) 2005, 2007-
+2009, 2011-2012’ should be interpreted as being identical to a statement that
+reads ‘Copyright (c) 2005, 2007, 2008, 2009, 2011, 2012’ and a copyright
+statement that reads ‘Copyright (c) 2005-2012’ should be interpreted as being
+identical to a statement that reads ‘Copyright (c) 2005, 2006, 2007, 2008,
+2009, 2010, 2011, 2012’."

bin/Admin_EnsemblGtf2CacheConverter.pl

@@ -38,7 +38,7 @@
 use File::Type;
 use Const::Fast qw(const);
 
-use Bio::DB::Sam;
+use Bio::DB::HTS;
 
 use Sanger::CGP::Vagrent::Data::Transcript;
 use Sanger::CGP::Vagrent::Data::Exon;
@@ -68,7 +68,7 @@
 sub convertGtf {
   my ($opts,$lookup,$ccds) = @_;
   my $type = getInputFileType($opts->{'g'});
-	my $fasta = Bio::DB::Sam::Fai->load($opts->{'f'});
+	my $fasta = Bio::DB::HTS::Fai->load($opts->{'f'});
 	my ($inFh,$outFh);
 	if($type eq 'text') {
 		open $inFh, '<'.$opts->{'g'} || croak("unable to open gtf input file:".$opts->{'g'});

lib/Sanger/CGP/Vagrent.pm

@@ -1,7 +1,7 @@
 package Sanger::CGP::Vagrent;
 
 ##########LICENCE##########
-# Copyright (c) 2014 Genome Research Ltd.
+# Copyright (c) 2014-2016 Genome Research Ltd.
 #
 # Author: Cancer Genome Project [email protected]
 #
@@ -26,7 +26,7 @@ use strict;
 use Const::Fast qw(const);
 
 use base 'Exporter';
-our $VERSION = '2.1.3';
+our $VERSION = '3.0.0';
 our @EXPORT = qw($VERSION);
 
 1;

lib/Sanger/CGP/Vagrent/TranscriptSource/FileBasedTranscriptSource.pm

@@ -2,21 +2,21 @@ package Sanger::CGP::Vagrent::TranscriptSource::FileBasedTranscriptSource;
 
 ##########LICENCE##########
 # Copyright (c) 2014 Genome Research Ltd.
-# 
+#
 # Author: Cancer Genome Project [email protected]
-# 
+#
 # This file is part of VAGrENT.
-# 
+#
 # VAGrENT is free software: you can redistribute it and/or modify it under
 # the terms of the GNU Affero General Public License as published by the Free
 # Software Foundation; either version 3 of the License, or (at your option) any
 # later version.
-# 
+#
 # This program is distributed in the hope that it will be useful, but WITHOUT
 # ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
 # FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more
 # details.
-# 
+#
 # You should have received a copy of the GNU Affero General Public License
 # along with this program. If not, see <http://www.gnu.org/licenses/>.
 ##########LICENCE##########
@@ -29,14 +29,8 @@ use Log::Log4perl;
 use Data::Dumper;
 use Const::Fast qw(const);
 
-use Bio::DB::Sam;
-
-# prevent the init warnings from Tabix
-BEGIN {
-  $SIG{__WARN__} = sub {warn $_[0] unless( $_[0] =~ m/^Subroutine Tabix.* redefined/)};
-};
-
-use Tabix;
+use Bio::DB::HTS;
+use Bio::DB::HTS::Tabix;
 
 use Sanger::CGP::Vagrent::Data::Transcript;
 use Sanger::CGP::Vagrent::Data::Exon;
@@ -126,25 +120,23 @@ sub _generateLocationString {
 
 sub _getTranscriptsFromCache {
   my ($self,$gp) = @_;
-  $self->{_cache_tbx} = Tabix->new('-data' => $self->{_cache}) unless defined $self->{_cache_tbx};
+  $self->{_cache_tbx} = Bio::DB::HTS::Tabix->new(filename => $self->{_cache}) unless defined $self->{_cache_tbx};
   my $min;
   my $max = $gp->getMaxPos + $SEARCH_BUFFER;
   if($gp->getMinPos() < $SEARCH_BUFFER){
     $min = 0;
   } else {
     $min = ($gp->getMinPos - $SEARCH_BUFFER) - 1;
   }
-  my $res = $self->{_cache_tbx}->query($gp->getChr(),$min,$max);
-  return undef unless defined $res;
+  my $iter = $self->{_cache_tbx}->query(sprintf '%s:%d-%d', $gp->getChr(),$min,$max);
+  return undef unless defined $iter;
   my $out = undef;
-  if(defined $res->get){
-    while(my $ret = $self->{_cache_tbx}->read($res)){
-      my $raw = (split("\t",$ret))[6];
-      my $VAR1;
-      eval $raw;
-      $VAR1->{_cdnaseq} = $self->_getTranscriptSeq($VAR1);
-      push @$out, $VAR1;
-    }
+  while(my $ret = $iter->next){
+    my $raw = (split("\t",$ret))[6];
+    my $VAR1;
+    eval $raw;
+    $VAR1->{_cdnaseq} = $self->_getTranscriptSeq($VAR1);
+    push @$out, $VAR1;
   }
   return $out;
 }
@@ -185,7 +177,7 @@ sub _setCacheFile {
 sub _getTranscriptSeq {
   my ($self,$trans) = @_;
   unless(defined $self->{_fai_obj}){
-    $self->{_fai_obj} = Bio::DB::Sam::Fai->load($self->{_cache_fa});
+    $self->{_fai_obj} = Bio::DB::HTS::Fai->load($self->{_cache_fa});
   }
   return $self->{_fai_obj}->fetch($trans->getAccession);
 }

prerelease.sh

@@ -43,7 +43,7 @@ cd $MY_PATH
 
 echo '########## Running perl tests ##########'
 
-export HARNESS_PERL_SWITCHES=-MDevel::Cover=-db,reports,-ignore,'t/.*\.t,/Bio',
+export HARNESS_PERL_SWITCHES=-MDevel::Cover=-db,reports,-select='^lib/*\.pm$',-ignore,'^t/'
 rm -rf docs
 mkdir -p docs/reports_text
 prove -w -I lib t