From 02bf2b081aed88622e3279e816f0aeb45cba546e Mon Sep 17 00:00:00 2001 From: "cmzmasek@gmail.com" Date: Wed, 28 Jan 2015 03:15:33 +0000 Subject: [PATCH] in progress --- .../org/forester/application/msa_compactor.java | 4 + forester/java/src/org/forester/msa/MsaMethods.java | 38 ++++++++++ .../org/forester/msa_compactor/MsaCompactor.java | 79 ++++++++++---------- .../org/forester/msa_compactor/MsaProperties.java | 8 ++ 4 files changed, 89 insertions(+), 40 deletions(-) diff --git a/forester/java/src/org/forester/application/msa_compactor.java b/forester/java/src/org/forester/application/msa_compactor.java index 4b68a89..15c2d4c 100644 --- a/forester/java/src/org/forester/application/msa_compactor.java +++ b/forester/java/src/org/forester/application/msa_compactor.java @@ -427,6 +427,10 @@ public class msa_compactor { System.out.println( " Min sequence length : " + ( ( int ) initial_msa_stats.getMin() ) ); System.out.println( " Gap ratio : " + NF_4.format( MsaMethods.calcGapRatio( msa ) ) ); + System.out.println( " Mean gap count per 100 residues : " + + NF_1.format( MsaMethods.calcNumberOfGapsPer100Stats( msa ).arithmeticMean() ) ); + System.out.println( " Normalized Shannon Entropy (entn7) : " + + NF_4.format( MsaMethods.calcNormalizedShannonsEntropy( 7, msa ) ) ); System.out.println( " Normalized Shannon Entropy (entn21): " + NF_4.format( MsaMethods.calcNormalizedShannonsEntropy( 21, msa ) ) ); } diff --git a/forester/java/src/org/forester/msa/MsaMethods.java b/forester/java/src/org/forester/msa/MsaMethods.java index ee5188e..881e622 100644 --- a/forester/java/src/org/forester/msa/MsaMethods.java +++ b/forester/java/src/org/forester/msa/MsaMethods.java @@ -107,6 +107,44 @@ public final class MsaMethods { _ignored_seqs_ids = new ArrayList(); } + public static final DescriptiveStatistics calcNumberOfGapsPer100Stats( final Msa msa ) { + final int[] gaps = calcNumberOfGapsInMsa( msa ); + final DescriptiveStatistics stats = new BasicDescriptiveStatistics(); + final double n = 100.0 / msa.getLength(); + for( final int gap : gaps ) { + stats.addValue( n * gap ); + } + return stats; + } + + public static final int[] calcNumberOfGapsInMsa( final Msa msa ) { + final int seqs = msa.getNumberOfSequences(); + final int[] gaps= new int[ seqs ]; + for( int i = 0; i < seqs; ++i ) { + gaps[ i ] = calcNumberOfGaps( msa.getSequence( i ) ); + } + return gaps; + } + + + + public final static int calcNumberOfGaps( final MolecularSequence seq ) { + int gaps = 0; + boolean was_gap = false; + for( int i = 0; i < seq.getLength(); ++i ) { + if ( seq.isGapAt( i ) ) { + if ( !was_gap ) { + ++gaps; + was_gap = true; + } + } + else { + was_gap = false; + } + } + return gaps; + } + public static DescriptiveStatistics calcBasicGapinessStatistics( final Msa msa ) { final DescriptiveStatistics stats = new BasicDescriptiveStatistics(); for( int i = 0; i < msa.getLength(); ++i ) { diff --git a/forester/java/src/org/forester/msa_compactor/MsaCompactor.java b/forester/java/src/org/forester/msa_compactor/MsaCompactor.java index da74dec..687951c 100644 --- a/forester/java/src/org/forester/msa_compactor/MsaCompactor.java +++ b/forester/java/src/org/forester/msa_compactor/MsaCompactor.java @@ -230,34 +230,32 @@ public class MsaCompactor { it.next().getNodeData().setNodeVisualData( vis ); } } - - for( int i = 0; i < msa_props.size(); ++i ) { - final MsaProperties msa_prop = msa_props.get( i ); - final String id = msa_prop.getRemovedSeq(); - if ( !ForesterUtil.isEmpty( id ) ) { - final PhylogenyNode n = phy.getNode( id ); - n.setName( n.getName() + " [" + i + "]" ); - if ( !chart_only ) { - final NodeVisualData vis = new NodeVisualData(); - vis.setFillType( NodeFill.SOLID ); - vis.setShape( NodeShape.RECTANGLE ); - vis.setNodeColor( ForesterUtil.calcColor( msa_prop.getLength(), min, max, mean_color, max_color ) ); - n.getNodeData().setNodeVisualData( vis ); - } - else { - n.getNodeData() - .getNodeVisualData() - .setNodeColor( ForesterUtil.calcColor( msa_prop.getLength(), - min, - max, - mean, - min_color, - max_color, - mean_color ) ); - } - } - + for( int i = 0; i < msa_props.size(); ++i ) { + final MsaProperties msa_prop = msa_props.get( i ); + final String id = msa_prop.getRemovedSeq(); + if ( !ForesterUtil.isEmpty( id ) ) { + final PhylogenyNode n = phy.getNode( id ); + n.setName( n.getName() + " [" + i + "]" ); + if ( !chart_only ) { + final NodeVisualData vis = new NodeVisualData(); + vis.setFillType( NodeFill.SOLID ); + vis.setShape( NodeShape.RECTANGLE ); + vis.setNodeColor( ForesterUtil.calcColor( msa_prop.getLength(), min, max, mean_color, max_color ) ); + n.getNodeData().setNodeVisualData( vis ); + } + else { + n.getNodeData() + .getNodeVisualData() + .setNodeColor( ForesterUtil.calcColor( msa_prop.getLength(), + min, + max, + mean, + min_color, + max_color, + mean_color ) ); } + } + } } final public void deleteGapColumns( final double max_allowed_gap_ratio ) { @@ -304,7 +302,7 @@ public class MsaCompactor { } public final List removeViaGapAverage( final double mean_gapiness ) throws IOException, - InterruptedException { + InterruptedException { final GapContribution stats[] = calcGapContribtionsStats( _norm ); final List to_remove_ids = new ArrayList(); final List msa_props = new ArrayList(); @@ -349,7 +347,7 @@ public class MsaCompactor { System.out.println( msg ); } if ( _phylogentic_inference ) { - decorateTree( phy, msa_props, false ); + decorateTree( phy, msa_props, false ); displayTree( phy ); } return msa_props; @@ -400,14 +398,14 @@ public class MsaCompactor { System.out.println( msg ); } if ( _phylogentic_inference ) { - decorateTree( phy, msa_props, false ); + decorateTree( phy, msa_props, false ); displayTree( phy ); } return msa_props; } public final List removeWorstOffenders( final int to_remove ) throws IOException, - InterruptedException { + InterruptedException { final GapContribution stats[] = calcGapContribtionsStats( _norm ); final List to_remove_ids = new ArrayList(); final List msa_props = new ArrayList(); @@ -450,16 +448,13 @@ public class MsaCompactor { System.out.println( msg ); } if ( _phylogentic_inference ) { - decorateTree( phy, msa_props, false ); + decorateTree( phy, msa_props, false ); displayTree( phy ); - - - System.out.println( "calculating phylogentic tree..." ); - System.out.println(); - Phylogeny phy2 = calcTree(); - addSeqs2Tree( _msa, phy2 ); - displayTree( phy2 ); - + System.out.println( "calculating phylogentic tree..." ); + System.out.println(); + final Phylogeny phy2 = calcTree(); + addSeqs2Tree( _msa, phy2 ); + displayTree( phy2 ); } return msa_props; } @@ -641,6 +636,8 @@ public class MsaCompactor { sb.append( msa_properties.getLength() ); sb.append( "\t" ); sb.append( NF_4.format( msa_properties.getGapRatio() ) ); + sb.append( "\t" ); + sb.append( NF_1.format( msa_properties.getAvgNumberOfGapsPer100() ) ); if ( _calculate_shannon_entropy ) { sb.append( "\t" ); sb.append( NF_4.format( msa_properties.getEntropy7() ) ); @@ -708,6 +705,8 @@ public class MsaCompactor { System.out.print( "\t" ); System.out.print( "Length" ); System.out.print( "\t" ); + System.out.print( "Gap R" ); + System.out.print( "\t" ); System.out.print( "Gaps" ); System.out.print( "\t" ); if ( _calculate_shannon_entropy ) { diff --git a/forester/java/src/org/forester/msa_compactor/MsaProperties.java b/forester/java/src/org/forester/msa_compactor/MsaProperties.java index ae575b2..5343aec 100644 --- a/forester/java/src/org/forester/msa_compactor/MsaProperties.java +++ b/forester/java/src/org/forester/msa_compactor/MsaProperties.java @@ -34,6 +34,7 @@ public final class MsaProperties { final private double _gap_ratio; final private int _length; final private int _number_of_sequences; + final private double _avg_number_of_gaps_per_100; final private String _removed_seq; public MsaProperties( final int number_of_sequences, @@ -41,12 +42,14 @@ public final class MsaProperties { final double gap_ratio, final double entropy7, final double entropy21, + final double avg_number_of_gaps_per_100, final String removed_seq ) { _number_of_sequences = number_of_sequences; _length = length; _gap_ratio = gap_ratio; _entropy7 = entropy7; _entropy21 = entropy21; + _avg_number_of_gaps_per_100 = avg_number_of_gaps_per_100; _removed_seq = removed_seq; } @@ -55,6 +58,7 @@ public final class MsaProperties { _length = msa.getLength(); _gap_ratio = MsaMethods.calcGapRatio( msa ); _removed_seq = removed_seq; + _avg_number_of_gaps_per_100 = MsaMethods.calcNumberOfGapsPer100Stats( msa ).arithmeticMean(); if ( calculate_normalized_shannon_entropy ) { _entropy7 = MsaMethods.calcNormalizedShannonsEntropy( 7, msa ); _entropy21 = MsaMethods.calcNormalizedShannonsEntropy( 21, msa ); @@ -77,6 +81,10 @@ public final class MsaProperties { return _gap_ratio; } + public final double getAvgNumberOfGapsPer100() { + return _avg_number_of_gaps_per_100; + } + public final int getLength() { return _length; } -- 1.7.10.2