Updated trunk to 0.5.3

git-svn-id: https://jcvi-tools.svn.sourceforge.net/svnroot/jcvi-tools/JCVI-Translator/trunk@20 a16988d2-67c9-4db9-a897-0603be59747b

Changes

@@ -1,8 +1,13 @@
 Revision history for JCVI-Translator
 
+0.5.3   Apr 13 2009
+ - Fixed inconsistencies in the way the start codon is being stored in the
+   reverse translation table
+ - Added test for above condition
+
 0.5.2   Apr 13 2009
  - Even more documentation in Utils
- - codons/regex/find method in Utils can take "stop" for stop codons 
+ - codons/regex/find method in Utils can take "stop" for stop codons
 
 0.5.1   Apr 13 2009
  - Better documentation in Utils

lib/JCVI/Translator.pm

@@ -67,7 +67,7 @@ use strict;
 use warnings;
 
 use version;
-our $VERSION = qv('0.5.2');
+our $VERSION = qv('0.5.3');
 
 use base qw(Class::Accessor::Fast);
 __PACKAGE__->mk_accessors(qw(table base));

lib/JCVI/Translator/Table.pm

@@ -391,8 +391,8 @@ sub add_translation {
     $table->store( $residue, $$codon_ref, $$rc_codon_ref );
 
     # Store the reverse lookup
-    $residue = 'start' if ( $p{start} );
-    $self->_reverse->push( 'start', $$codon_ref, $$rc_codon_ref );
+    $residue = '+' if ( $p{start} );
+    $self->_reverse->push( $residue, $$codon_ref, $$rc_codon_ref );
 }
 
 =head2 bootstrap

lib/JCVI/Translator/Utils.pm

@@ -57,7 +57,7 @@ our $BOOLEAN_REGEX       = qr/^[01]$/;
 our $INTEGER_REGEX       = qr/^\d+$/;
 our $STRAND_REGEX        = qr/^[+-]?1$/;
 our $SEARCH_STRAND_REGEX = qr/^[+-]?[01]$/;
-our $RESIDUE_REGEX       = qr/^(?:$aa_match|start|stop|lower|upper)$/;
+our $RESIDUE_REGEX       = qr/^(?:$aa_match|\+|start|stop|lower|upper)$/;
 our $STRICT_REGEX        = qr/^[012]$/;
 
 sub _new {
@@ -79,8 +79,9 @@ Returns a list of codons for a particular residue or start codon. In addition
 to the one-letter codes for amino acids, the following are valid inputs for the
 residue:
 
-    start:  Start codons
-    stop:   Stop codons (you may also use "*" which is the 1 letter code)
+    start:  Start codons (you may also use "+" which is what the translator
+            uses as the 1-letter code for start codons)
+    stop:   Stop codons (you may also use "*" which is the 1-letter code)
     lower:  Start or stop codons, depending up on strand
     upper:  Start or stop codons, depending up on strand
 
@@ -124,17 +125,14 @@ sub codons {
     # Set the reverse comlement variable
     my $rc = $p{strand} == 1 ? 0 : 1;
 
-    # If residue is 'stop' change it to '*'
-    if ( $residue eq 'stop' ) { $residue = '*' }
-
-    # Do nothing if residue is "start" (don't want to capitalize)
-    elsif ( $residue eq 'start' ) { }
-
-    # Lower bound is "*" on the - strand, "start" on the + strand
-    elsif ( $residue eq 'lower' ) { $residue = $rc ? '*' : 'start' }
+    # Format start/stop to be '+' and '*' which is how translator stores them
+    if    ( $residue eq 'stop' )  { $residue = '*' } 
+    elsif ( $residue eq 'start' ) { $residue = '+' }
 
-    # Upper bound is "start" on the - strand, or "*" on the + strand
-    elsif ( $residue eq 'upper' ) { $residue = $rc ? 'start' : '*' }
+    # Lower bound is stop on the - strand, start on the + strand. Upper bound
+    # is the reverse
+    elsif ( $residue eq 'lower' ) { $residue = $rc ? '*' : '+' }
+    elsif ( $residue eq 'upper' ) { $residue = $rc ? '+' : '*' }
 
     # Capitalize all other residues
     else { $residue = uc $residue }
@@ -157,7 +155,8 @@ Returns a regular expression matching codons for a particular amino acid
 residue. In addition to the one-letter codes for amino acids, the following are
 valid inputs for the residue:
 
-    start:  Start codons
+    start:  Start codons (you may also use "+" which is what the translator
+            uses as the 1-letter code for start codons)
     stop:   Stop codons (you may also use "*" which is the 1 letter code)
     lower:  Start or stop codons, depending up on strand
     upper:  Start or stop codons, depending up on strand
@@ -225,7 +224,8 @@ Find the indexes of a given residue in a sequence. In addition to the
 one-letter codes for amino acids, the following are valid inputs for the
 residue:
 
-    start:  Start codons
+    start:  Start codons (you may also use "+" which is what the translator
+            uses as the 1-letter code for start codons)
     stop:   Stop codons (you may also use "*" which is the 1 letter code)
     lower:  Start or stop codons, depending up on strand
     upper:  Start or stop codons, depending up on strand

t/21-codons.t

@@ -23,18 +23,31 @@ ok( !$@, 'codons ran with strand = 1' );
 
 eval { $utils->codons( 'F', { strand => -1 } ) };
 ok( !$@, 'codons ran with strand = -1' );
-
 eval { $utils->codons( 'F', { strand => 2 } ) };
 ok( $@, 'codons died with strand = 2' );
 
-my @expected = ( [qw(TTT TTC TTY)], [qw(AAA GAA RAA)] );
-
-foreach my $strand ( 1, -1 ) {
-    my $codons = $utils->codons( 'F', { strand => $strand } );
-    my $expected = shift @expected;
-    is( scalar(@$codons), scalar(@$expected), 'Expected number of codons' );
-
-    my $lc = List::Compare->new( $expected, $codons );
-
-    is( scalar( $lc->get_symdiff ), 0, '0 differences between lists' );
+my @entries = (
+    [ 'F' => [ [qw(TTT TTC TTY)], [qw(AAA GAA RAA)] ] ],
+    [
+        'start' => [
+            [qw(YTG WTG MTG HTG TTG CTG ATG)],
+            [qw(CAR CAW CAK CAD CAA CAG CAT)]
+        ]
+    ]
+);
+
+foreach my $entry (@entries) {
+    my ( $residue, $expecteds ) = @$entry;
+
+    foreach my $strand ( 1, -1 ) {
+        my $codons = $utils->codons( $residue, { strand => $strand } );
+        my $expected = shift @$expecteds;
+        is( scalar(@$codons), scalar(@$expected),
+            "Got expected number of codons for $residue" );
+
+        my $lc = List::Compare->new( $expected, $codons );
+
+        is( scalar( $lc->get_symdiff ), 0,
+            '0 differences between codon lists' );
+    }
 }