From ad46292d0fd8f1d757ae876d5c9771964bce27e7 Mon Sep 17 00:00:00 2001
From: martinahansen <martinahansen@74ccb610-7750-0410-82ae-013aeee3265d>
Date: Fri, 10 Jul 2009 18:38:43 +0000
Subject: [PATCH] speedup of analyze_seq 3 fold

git-svn-id: http://biopieces.googlecode.com/svn/trunk@564 74ccb610-7750-0410-82ae-013aeee3265d
---
 bp_bin/analyze_seq      | 71 +++++++++++++++++++++++++++++++++++++----
 code_perl/Maasha/Seq.pm | 28 +++++++++-------
 2 files changed, 80 insertions(+), 19 deletions(-)

diff --git a/bp_bin/analyze_seq b/bp_bin/analyze_seq
index b6f7d8c..1418889 100755
--- a/bp_bin/analyze_seq
+++ b/bp_bin/analyze_seq
@@ -22,7 +22,8 @@
 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> DESCRIPTION <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
 
 
-# Analyze BED entries in the stream.
+# Analyze sequences in the stream.
+
 
 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
 
@@ -30,6 +31,7 @@
 use warnings;
 use strict;
 use Maasha::Biopieces;
+use Maasha::Common;
 use Maasha::Seq;
 
 
@@ -45,12 +47,7 @@ $out = Maasha::Biopieces::write_stream( $options->{ "stream_out" } );
 
 while ( $record = Maasha::Biopieces::get_record( $in ) ) 
 {
-    if ( $record->{ "SEQ" } )
-    {
-        $analysis = Maasha::Seq::seq_analyze( $record->{ "SEQ" } );
-
-        map { $record->{ $_ } = $analysis->{ $_ } } keys %{ $analysis };
-    }
+    seq_analyze( $record ) if $record->{ "SEQ" };
 
     Maasha::Biopieces::put_record( $record, $out );
 }
@@ -62,6 +59,66 @@ Maasha::Biopieces::close_stream( $out );
 # >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
 
 
+sub seq_analyze
+{
+    # Martin A. Hansen, July 2009.
+
+    # Analyzes the composition of the string in the record and appends
+    # the analysis to the record.
+
+    my ( $record,   # Biopiece record with a SEQ entry.
+       ) = @_;
+
+    # Returns nothing.
+
+    my ( %char_hash, @indels, @alph, $char, $gc, $at, $lc, $max, $indels );
+
+    %char_hash = Maasha::Common::str_analyze( $record->{ 'SEQ' } );
+
+    $record->{ 'SEQ_TYPE' } = Maasha::Seq::seq_guess_type( $record->{ 'SEQ' } );
+    $record->{ 'SEQ_LEN' }  = length $record->{ 'SEQ' };
+
+    @alph   = Maasha::Seq::seq_alph( $record->{ 'SEQ_TYPE' } . "_AMBI" );
+    @indels = qw( - ~ . _ );
+
+    $max = 0;
+
+    foreach $char ( @alph )
+    {
+        $char_hash{ $char } += $char_hash{ lc $char } || 0;
+
+        $record->{ "RES[$char]" } = $char_hash{ $char };
+
+        $max = $char_hash{ $char } if $char_hash{ $char } > $max;
+
+        $record->{ "RES_SUM" } += $char_hash{ $char };
+    }
+
+    $indels = 0;
+
+    map { $record->{ "RES[$_]" } = $char_hash{ $_ }; $indels += $char_hash{ $_ } } @indels;
+
+    if ( $record->{ "SEQ_TYPE" } =~ /DNA|RNA/i )
+    {
+        $gc = $char_hash{ "G" } + $char_hash{ "C" };
+        $at = $char_hash{ "A" } + $char_hash{ "T" } + $char_hash{ "U" };
+
+        $lc = 0;
+
+        map { $lc += $char_hash{ lc $_ } || 0 } @alph;
+
+        $record->{ "MIX_INDEX" }  = sprintf( "%.2f", $max / ( $record->{ "SEQ_LEN" } - $indels ) );
+        $record->{ "GC%" }        = sprintf( "%.2f", 100 * $gc / ( $record->{ "SEQ_LEN" } - $indels ) );
+        $record->{ "SOFT_MASK%" } = sprintf( "%.2f", 100 * $lc / ( $record->{ "SEQ_LEN" } - $indels ) );
+        $record->{ "HARD_MASK%" } = sprintf( "%.2f", 100 * ( $char_hash{ "n" } + $char_hash{ "N" } ) / ( $record->{ "SEQ_LEN" } - $indels ) );
+        $record->{ "MELT_TEMP" }  = sprintf( "%.2f", 4 * $gc + 2 * $at );
+    }
+}
+
+
+# >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
+
+
 BEGIN
 {
     Maasha::Biopieces::status_set();
diff --git a/code_perl/Maasha/Seq.pm b/code_perl/Maasha/Seq.pm
index cfedd89..2c64f15 100644
--- a/code_perl/Maasha/Seq.pm
+++ b/code_perl/Maasha/Seq.pm
@@ -65,11 +65,11 @@ sub seq_guess_type
     }
 
     if ( $count = $check_seq =~ tr/FLPQIEflpqie// and $count > 0 ) {
-        return "protein";
+        return "PROTEIN";
     } elsif ( $count = $check_seq =~ tr/Uu// and $count > 0 ) {
-        return "rna";
+        return "RNA";
     } else {
-        return "dna";
+        return "DNA";
     }
 }
 
@@ -914,26 +914,30 @@ sub seq_alph
 {
     # Martin A. Hansen, May 2007.
 
-    # returns a requested alphabet
+    # Returns a requested sequence alphabet.
 
     my ( $type,   # alphabet type
        ) = @_;
 
     # returns list
 
-    my ( @alph );
+    my ( %alph_hash, $alph );
 
-    if ( $type =~ /^dna$/i ) {
-        @alph = qw( A T C G );
-    } elsif ( $type =~ /^rna$/i ) {
-        @alph = qw( A U C G );
-    } elsif ( $type =~ /^prot/i ) {
-        @alph = qw( F L S Y C W P H Q R I M T N K V A D E G );
+    %alph_hash = (
+        DNA      => [ qw( A T C G ) ],
+        DNA_AMBI => [ qw( A G C U T R Y W S M K H D V B N ) ],
+        RNA      => [ qw( A U C G ) ],
+        RNA_AMBI => [ qw( A G C U T R Y W S M K H D V B N ) ],
+        PROTEIN  => [ qw( F L S Y C W P H Q R I M T N K V A D E G ) ],
+    );
+
+    if ( exists $alph_hash{ $type } ) {
+        $alph = $alph_hash{ $type };
     } else {
         die qq(ERROR: Unknown alphabet type: "$type"\n);
     }
 
-    return wantarray ? @alph : \@alph;
+    return wantarray ? @{ $alph } : $alph;
 }
 
 
-- 
2.39.5