package org.forester.msa;
import java.io.IOException;
+import java.io.StringWriter;
import java.io.Writer;
import java.util.ArrayList;
import java.util.HashSet;
import org.forester.io.writers.SequenceWriter;
import org.forester.io.writers.SequenceWriter.SEQ_FORMAT;
import org.forester.sequence.BasicSequence;
-import org.forester.sequence.Sequence;
-import org.forester.sequence.Sequence.TYPE;
+import org.forester.sequence.MolecularSequence;
+import org.forester.sequence.MolecularSequence.TYPE;
import org.forester.util.ForesterUtil;
public class BasicMsa implements Msa {
- private final char[][] _data;
- private final String[] _identifiers;
- private final TYPE _type;
+ private final char[][] _data;
+ private final String[] _identifiers;
+ private final Set<String> _identifiers_set;
+ private final TYPE _type;
public BasicMsa( final int rows, final int columns, final TYPE type ) {
if ( ( rows < 1 ) || ( columns < 1 ) ) {
}
_data = new char[ rows ][ columns ];
_identifiers = new String[ rows ];
+ _identifiers_set = new HashSet<String>();
_type = type;
}
_data = msa._data;
_identifiers = msa._identifiers;
_type = msa._type;
+ _identifiers_set = msa._identifiers_set;
}
@Override
- public List<Sequence> asSequenceList() {
- final List<Sequence> seqs = new ArrayList<Sequence>();
+ public List<MolecularSequence> asSequenceList() {
+ final List<MolecularSequence> seqs = new ArrayList<MolecularSequence>();
for( int i = 0; i < getNumberOfSequences(); ++i ) {
seqs.add( getSequence( i ) );
}
return seqs;
}
- private int determineMaxIdLength() {
- int max = 0;
+ @Override
+ public List<Character> getColumnAt( final int col ) {
+ final List<Character> column = new ArrayList<Character>();
for( int row = 0; row < getNumberOfSequences(); ++row ) {
- final int l = getIdentifier(row).length();
- if ( l > max ) {
- max = l;
- }
+ column.add( getResidueAt( row, col ) );
}
- return max;
+ return column;
}
@Override
}
@Override
- public Sequence getSequence( final String id ) {
+ public MolecularSequence getSequence( final int row ) {
+ return new BasicSequence( getIdentifier( row ), _data[ row ], getType() );
+ }
+
+ @Override
+ public MolecularSequence getSequence( final String id ) {
for( int i = 0; i < getNumberOfSequences(); ++i ) {
if ( getIdentifier( i ).equals( id ) ) {
return getSequence( i );
}
@Override
- public Sequence getSequence( final int row ) {
- return new BasicSequence( getIdentifier( row ), getSequenceAsArray( row ), getType() );
- }
-
- @Override
public StringBuffer getSequenceAsString( final int row ) {
- final StringBuffer sb = new StringBuffer(getLength() );
+ final StringBuffer sb = new StringBuffer( getLength() );
for( int col = 0; col < getLength(); ++col ) {
sb.append( getResidueAt( row, col ) );
}
return sb;
}
-
- @Override
- public char[] getSequenceAsArray( final int row ) {
- return _data[ row ];
- }
@Override
public TYPE getType() {
}
@Override
+ public boolean isGapAt( final int row, final int col ) {
+ return getResidueAt( row, col ) == MolecularSequence.GAP;
+ }
+
+ @Override
public void setIdentifier( final int row, final String id ) {
+ if ( ForesterUtil.isEmpty( id ) ) {
+ throw new IllegalArgumentException( "illegal attempt to create msa with empty identifier" );
+ }
+ if ( _identifiers_set.contains( id ) ) {
+ throw new IllegalArgumentException( "illegal attempt to create msa with non-unique identifiers [" + id
+ + "]" );
+ }
+ _identifiers_set.add( id );
_identifiers[ row ] = id;
}
@Override
public String toString() {
- final int max = determineMaxIdLength() + 1;
- final StringBuffer sb = new StringBuffer();
- for( int row = 0; row < getNumberOfSequences(); ++row ) {
- sb.append( ForesterUtil.pad( getIdentifier( row ).toString(), max, ' ', false ) );
- for( int col = 0; col < getLength(); ++col ) {
- sb.append( getResidueAt( row, col ) );
- }
- sb.append( ForesterUtil.LINE_SEPARATOR );
+ final Writer w = new StringWriter();
+ try {
+ write( w, MSA_FORMAT.PHYLIP );
+ }
+ catch ( final IOException e ) {
+ e.printStackTrace();
}
- return sb.toString();
+ return w.toString();
}
@Override
case FASTA:
writeToFasta( w );
break;
+ case NEXUS:
+ writeToNexus( w );
+ break;
default:
throw new RuntimeException( "unknown format " + format );
}
}
+ private short determineMaxIdLength() {
+ short max = 0;
+ for( int row = 0; row < getNumberOfSequences(); ++row ) {
+ final short l = ( short ) getIdentifier( row ).length();
+ if ( l > max ) {
+ max = l;
+ }
+ }
+ return max;
+ }
+
private void writeToFasta( final Writer w ) throws IOException {
SequenceWriter.writeSeqs( asSequenceList(), w, SEQ_FORMAT.FASTA, 100 );
}
+ private void writeToNexus( final Writer w ) throws IOException {
+ final int max = determineMaxIdLength() + 1;
+ w.write( "Begin Data;" );
+ w.write( ForesterUtil.LINE_SEPARATOR );
+ w.write( " Dimensions NTax=" + getNumberOfSequences() );
+ w.write( " NChar=" + getLength() );
+ w.write( ";" );
+ w.write( ForesterUtil.LINE_SEPARATOR );
+ w.write( " Format DataType=Protein Interleave=No gap=-;" );
+ w.write( ForesterUtil.LINE_SEPARATOR );
+ w.write( " Matrix" );
+ w.write( ForesterUtil.LINE_SEPARATOR );
+ for( int row = 0; row < getNumberOfSequences(); ++row ) {
+ final MolecularSequence seq = getSequence( row );
+ final String s = seq.getMolecularSequenceAsString();
+ w.write( " " );
+ w.write( ForesterUtil.pad( getIdentifier( row ).replace( ' ', '_' ), max, ' ', false ).toString() );
+ w.write( " " );
+ w.write( s );
+ w.write( ForesterUtil.LINE_SEPARATOR );
+ }
+ w.write( " ;" );
+ w.write( ForesterUtil.LINE_SEPARATOR );
+ w.write( "End;" );
+ w.write( ForesterUtil.LINE_SEPARATOR );
+ }
+
private void writeToPhylip( final Writer w ) throws IOException {
final int max = determineMaxIdLength() + 1;
+ w.write( getNumberOfSequences() + " " + getLength() );
+ w.write( ForesterUtil.LINE_SEPARATOR );
for( int row = 0; row < getNumberOfSequences(); ++row ) {
- w.write( ForesterUtil.pad( getIdentifier( row ), max, ' ', false ).toString() );
+ w.write( ForesterUtil.pad( getIdentifier( row ).replace( ' ', '_' ), max, ' ', false ).toString() );
for( int col = 0; col < getLength(); ++col ) {
w.write( getResidueAt( row, col ) );
}
}
}
- public static Msa createInstance( final List<Sequence> seqs ) {
+ public static Msa createInstance( final List<MolecularSequence> seqs ) {
if ( seqs.size() < 1 ) {
- throw new IllegalArgumentException( "cannot create basic msa from less than one sequence" );
+ throw new IllegalArgumentException( "cannot create msa from less than one sequence" );
}
- final Set<String> ids = new HashSet<String>();
final int length = seqs.get( 0 ).getLength();
final BasicMsa msa = new BasicMsa( seqs.size(), length, seqs.get( 0 ).getType() );
for( int row = 0; row < seqs.size(); ++row ) {
- final Sequence seq = seqs.get( row );
- //
- // int x = length - seq.getLength();
- // if ( x > 0 ) {
- // String a = "";
- // for( int i = 0; i < x; i++ ) {
- // a += "-";
- // }
- // seq = BasicSequence.createAaSequence( seq.getIdentifier(), seq.getMolecularSequenceAsString() + a );
- // }
- // else {
- // seq = BasicSequence.createAaSequence( seq.getIdentifier(), seq.getMolecularSequenceAsString()
- // .substring( 0, length ) );
- // }
- //
+ final MolecularSequence seq = seqs.get( row );
if ( seq.getLength() != length ) {
throw new IllegalArgumentException( "illegal attempt to build msa from sequences of unequal length ["
+ seq.getIdentifier() + "]" );
throw new IllegalArgumentException( "illegal attempt to build msa from sequences of different type ["
+ seq.getIdentifier() + "]" );
}
- if ( ids.contains( seq.getIdentifier() ) ) {
- throw new IllegalArgumentException( "illegal attempt to create msa with non-unique identifiers ["
- + seq.getIdentifier() + "]" );
- }
- ids.add( seq.getIdentifier() );
msa.setIdentifier( row, seq.getIdentifier() );
for( int col = 0; col < length; ++col ) {
msa._data[ row ][ col ] = seq.getResidueAt( col );
}
return msa;
}
-
- @Override
- public List<Character> getColumnAt( final int col ) {
- final List<Character> column = new ArrayList<Character>();
- for( int row = 0; row < getNumberOfSequences(); ++row ) {
- column.add( getResidueAt( row, col ) );
- }
- return column;
- }
-
- @Override
- public boolean isGapAt( final int row, final int col ) {
- return getResidueAt( row, col ) == Sequence.GAP;
- }
}