/*
* Jalview - A Sequence Alignment Editor and Viewer
-* Copyright (C) 2006 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
+ * Copyright (C) 2007 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
*/
package jalview.datamodel;
-import java.awt.*;
import java.util.*;
+import jalview.analysis.*;
/**
- * DOCUMENT ME!
+ *
+ * Implements the SequenceI interface for a char[] based sequence object.
*
* @author $author$
* @version $Revision$
*/
-public class Sequence implements SequenceI
+public class Sequence
+ implements SequenceI
{
SequenceI datasetSequence;
String name;
- private String sequence;
+ private char [] sequence;
String description;
int start;
int end;
- Color color = Color.white;
Vector pdbIds;
String vamsasId;
DBRefEntry[] dbrefs;
* positions are tied to the residues of this sequence */
Vector annotation;
- /** DOCUMENT ME!! */
+ /** array of seuqence features - may not be null for a valid sequence object */
public SequenceFeature[] sequenceFeatures;
- /** This array holds hidden sequences
- * of which this sequence is the representitive member of a group
- */
- SequenceGroup hiddenSequences;
/**
* Creates a new Sequence object.
*
- * @param name DOCUMENT ME!
- * @param sequence DOCUMENT ME!
- * @param start DOCUMENT ME!
- * @param end DOCUMENT ME!
+ * @param name display name string
+ * @param sequence string to form a possibly gapped sequence out of
+ * @param start first position of non-gap residue in the sequence
+ * @param end last position of ungapped residues (nearly always only used for display purposes)
*/
public Sequence(String name, String sequence, int start, int end)
{
this.name = name;
- this.sequence = sequence;
+ this.sequence = sequence.toCharArray();
this.start = start;
this.end = end;
-
parseId();
+ checkValidRange();
+ }
+ public Sequence(String name, char [] sequence, int start, int end)
+ {
+ this.name = name;
+ this.sequence = sequence;
+ this.start = start;
+ this.end = end;
+ parseId();
checkValidRange();
}
void parseId()
{
+ if (name==null)
+ {
+ System.err.println("POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
+ name = "";
+ }
// Does sequence have the /start-end signiature?
if (limitrx.search(name))
{
if (end < 1)
{
int endRes = 0;
- char ch;
- for (int j = 0; j < sequence.length(); j++)
+ for (int j = 0; j < sequence.length; j++)
{
- ch = sequence.charAt(j);
- if (!jalview.util.Comparison.isGap( (ch)))
+ if (!jalview.util.Comparison.isGap( sequence[j] ))
{
endRes++;
}
}
/**
- * Creates a new Sequence object.
- *
+ * Creates a new Sequence object with new features, DBRefEntries, AlignmentAnnotations, and PDBIds
+ * but inherits any existing dataset sequence reference.
* @param seq DOCUMENT ME!
*/
public Sequence(SequenceI seq)
{
- this(seq.getName(), seq.getSequence(), seq.getStart(), seq.getEnd());
+ this(seq, seq.getAnnotation());
+ }
+ /**
+ * Create a new sequence object with new features, DBRefEntries, and PDBIds
+ * but inherits any existing dataset sequence reference, and duplicate of
+ * any annotation that is present in the given annotation array.
+ * @param seq the sequence to be copied
+ * @param alAnnotation an array of annotation including some associated with seq
+ */
+ public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation)
+ {
+ this(seq.getName(),
+ seq.getSequence(),
+ seq.getStart(),
+ seq.getEnd());
+ description = seq.getDescription();
+ if (seq.getSequenceFeatures()!=null) {
+ SequenceFeature[] sf = seq.getSequenceFeatures();
+ for (int i=0; i<sf.length; i++) {
+ addSequenceFeature(new SequenceFeature(sf[i]));
+ }
+ }
+ if (seq.getDBRef()!=null) {
+ DBRefEntry[] dbr = seq.getDBRef();
+ for (int i=0; i<dbr.length; i++) {
+ addDBRef(new DBRefEntry(dbr[i]));
+ }
+ }
+ setDatasetSequence(seq.getDatasetSequence());
+ if (seq.getAnnotation()!=null) {
+ AlignmentAnnotation[] sqann = seq.getAnnotation();
+ for (int i=0;i<sqann.length; i++)
+ {
+ if (sqann[i]==null)
+ {
+ continue;
+ }
+ boolean found = (alAnnotation==null);
+ if (!found)
+ {
+ for (int apos = 0; !found && apos<alAnnotation.length; apos++)
+ {
+ found = (alAnnotation[apos] == sqann[i]);
+ }
+ }
+ if (found)
+ {
+ // only copy the given annotation
+ AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
+ addAlignmentAnnotation(newann);
+ }
+ }
+ }
+ if (seq.getPDBId()!=null) {
+ Vector ids = seq.getPDBId();
+ Enumeration e = ids.elements();
+ while (e.hasMoreElements()) {
+ this.addPDBId(new PDBEntry((PDBEntry) e.nextElement()));
+ }
+ }
}
/**
public void deleteFeature(SequenceFeature sf)
{
if(sequenceFeatures==null)
+ {
return;
+ }
int index=0;
for (index = 0; index < sequenceFeatures.length; index++)
if(index==sequenceFeatures.length)
+ {
return;
+ }
int sfLength = sequenceFeatures.length;
if(sfLength<2)
System.arraycopy(sequenceFeatures, 0, temp, 0, index);
if(index<sfLength)
+ {
System.arraycopy(sequenceFeatures,
index + 1,
temp,
index, sequenceFeatures.length - index -1);
+ }
sequenceFeatures = temp;
}
public void addPDBId(PDBEntry entry)
{
if (pdbIds == null)
+ {
pdbIds = new Vector();
+ }
pdbIds.addElement(entry);
}
*/
public int getLength()
{
- return this.sequence.length();
+ return this.sequence.length;
}
/**
*/
public void setSequence(String seq)
{
- this.sequence = seq;
+ this.sequence = seq.toCharArray();
checkValidRange();
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public String getSequence()
+
+ public String getSequenceAsString()
+ {
+ return new String(sequence);
+ }
+
+ public String getSequenceAsString(int start, int end)
+ {
+ return new String(getSequence(start, end));
+ }
+
+
+ public char [] getSequence()
{
- return this.sequence;
+ return sequence;
}
/**
*
* @return DOCUMENT ME!
*/
- public String getSequence(int start, int end)
+ public char [] getSequence(int start, int end)
{
+ if (start<0)
+ start=0;
// JBPNote - left to user to pad the result here (TODO:Decide on this policy)
- if (start >= sequence.length())
+ if (start >= sequence.length)
{
- return "";
+ return new char[0];
}
- if (end >= sequence.length())
+ if (end >= sequence.length)
{
- end = sequence.length();
+ end = sequence.length;
}
- return this.sequence.substring(start, end);
+ char [] reply = new char[end-start];
+ System.arraycopy(sequence, start, reply, 0, end-start);
+
+ return reply;
}
+
/**
* make a new Sequence object from start to end (including gaps) over this seqeunce
* @param start int
public SequenceI getSubSequence(int start, int end)
{
if (start < 0)
+ {
start = 0;
- String seq = getSequence(start, end);
- if (seq == "")
+ }
+ char [] seq = getSequence(start, end);
+ if (seq.length == 0)
+ {
return null;
+ }
int nstart = findPosition(start);
int nend = findPosition(end) - 1;
// JBPNote - this is an incomplete copy.
SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
- nseq.setDatasetSequence(getDatasetSequence());
+ nseq.setDescription(description);
+ if (datasetSequence!=null)
+ {
+ nseq.setDatasetSequence(datasetSequence);
+ }
+ else
+ {
+ nseq.setDatasetSequence(this);
+ }
return nseq;
}
*/
public char getCharAt(int i)
{
- if (i < sequence.length())
+ if (i < sequence.length)
{
- return sequence.charAt(i);
+ return sequence[i];
}
else
{
}
/**
- * DOCUMENT ME!
+ * Return the alignment position for a sequence position
*
- * @param pos DOCUMENT ME!
+ * @param pos lying from start to end
*
- * @return DOCUMENT ME!
+ * @return aligned position of residue pos
*/
public int findIndex(int pos)
{
int j = start;
int i = 0;
- while ( (i < sequence.length()) && (j <= end) && (j <= pos))
+ while ( (i < sequence.length) && (j <= end) && (j <= pos))
{
- if (!jalview.util.Comparison.isGap(sequence.charAt(i)))
+ if (!jalview.util.Comparison.isGap(sequence[i]))
{
j++;
}
{
int j = 0;
int pos = start;
- int seqlen = sequence.length();
+ int seqlen = sequence.length;
while ( (j < i) && (j < seqlen))
{
- if (!jalview.util.Comparison.isGap( (sequence.charAt(j))))
+ if (!jalview.util.Comparison.isGap( sequence[j] ))
{
pos++;
}
public int[] gapMap()
{
String seq = jalview.analysis.AlignSeq.extractGaps(jalview.util.Comparison.
- GapChars, sequence);
+ GapChars, new String(sequence));
int[] map = new int[seq.length()];
int j = 0;
int p = 0;
- while (j < sequence.length())
+ while (j < sequence.length)
{
- if (!jalview.util.Comparison.isGap(sequence.charAt(j)))
+ if (!jalview.util.Comparison.isGap(sequence[j]))
{
map[p++] = j;
}
return map;
}
- /**
- * DOCUMENT ME!
- *
- * @param i DOCUMENT ME!
+ /* (non-Javadoc)
+ * @see jalview.datamodel.SequenceI#deleteChars(int, int)
*/
- public void deleteCharAt(int i)
+ public void deleteChars(int i, int j)
{
- if (i >= sequence.length())
+ int newstart=start,newend=end;
+ if (i >= sequence.length)
{
return;
}
- sequence = sequence.substring(0, i) + sequence.substring(i + 1);
- }
+ char [] tmp;
- /**
- * DOCUMENT ME!
- *
- * @param i DOCUMENT ME!
- * @param j DOCUMENT ME!
- */
- public void deleteChars(int i, int j)
- {
- if (i >= sequence.length())
+ if (j >= sequence.length)
{
- return;
+ tmp = new char[i];
+ System.arraycopy(sequence,0,tmp,0,i);
}
-
- if (j >= sequence.length())
+ else
{
- sequence = sequence.substring(0, i);
+ tmp = new char[sequence.length-j+i];
+ System.arraycopy(sequence,0,tmp,0,i);
+ System.arraycopy(sequence,j,tmp,i,sequence.length-j);
}
- else
+ boolean createNewDs=false;
+ for (int s = i; s < j; s++)
{
- sequence = sequence.substring(0, i) + sequence.substring(j);
+ if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
+ {
+ if (createNewDs)
+ {
+ newend--;
+ } else {
+ int sindex = findIndex(start)-1;
+ if (sindex==s)
+ {
+ // delete characters including start of sequence
+ newstart = findPosition(j);
+ break; // don't need to search for any more residue characters.
+ } else {
+ // delete characters after start.
+ int eindex = findIndex(end)-1;
+ if (eindex<j)
+ {
+ // delete characters at end of sequence
+ newend = findPosition(i-1);
+ break; // don't need to search for any more residue characters.
+ } else {
+ createNewDs=true;
+ newend--; // decrease end position by one for the deleted residue and search further
+ }
+ }
+ }
+ }
}
+ // deletion occured in the middle of the sequence
+ if (createNewDs && this.datasetSequence != null)
+ {
+ // construct a new sequence
+ Sequence ds = new Sequence(datasetSequence);
+ // TODO: remove any non-inheritable properties ?
+ // TODO: create a sequence mapping (since there is a relation here ?)
+ ds.deleteChars(i, j);
+ datasetSequence = ds;
+ }
+ start = newstart;
+ end = newend;
+ sequence = tmp;
}
+
/**
* DOCUMENT ME!
*
*/
public void insertCharAt(int i, int length, char c)
{
- StringBuffer tmp;
+ char [] tmp = new char[sequence.length+length];
- if (i >= sequence.length())
+ if (i >= sequence.length)
{
- tmp = new StringBuffer(sequence);
+ System.arraycopy(sequence, 0, tmp, 0, sequence.length);
+ i = sequence.length;
}
else
- tmp = new StringBuffer(sequence.substring(0, i));
+ {
+ System.arraycopy(sequence, 0, tmp, 0, i);
+ }
+
+ int index = i;
while (length > 0)
{
- tmp.append(c);
+ tmp[ index++ ] = c;
length--;
}
- if (i < sequence.length())
+ if (i < sequence.length)
{
- tmp.append(sequence.substring(i));
+ System.arraycopy(sequence, i, tmp, index, sequence.length-i );
}
- sequence = tmp.toString();
+ sequence = tmp;
}
public void insertCharAt(int i, char c)
insertCharAt(i, 1, c);
}
- /**
- * DOCUMENT ME!
- *
- * @param c DOCUMENT ME!
- */
- public void setColor(Color c)
- {
- this.color = c;
- }
-
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public Color getColor()
- {
- return color;
- }
-
public String getVamsasId()
{
return vamsasId;
public void addDBRef(DBRefEntry entry)
{
if (dbrefs == null)
+ {
dbrefs = new DBRefEntry[0];
+ }
- DBRefEntry[] temp = new DBRefEntry[dbrefs.length + 1];
- System.arraycopy(dbrefs, 0, temp, 0, dbrefs.length);
+ int i, iSize = dbrefs.length;
+
+ for(i=0; i<iSize; i++)
+ {
+ if(dbrefs[i].equalRef(entry))
+ {
+ if (entry.getMap()!=null)
+ {
+ if (dbrefs[i].getMap()==null)
+ {
+ // overwrite with 'superior' entry that contains a mapping.
+ dbrefs[i] = entry;
+ }
+ }
+ return;
+ }
+ }
+ DBRefEntry[] temp = new DBRefEntry[iSize + 1];
+ System.arraycopy(dbrefs, 0, temp, 0, iSize);
temp[temp.length - 1] = entry;
dbrefs = temp;
public AlignmentAnnotation[] getAnnotation()
{
if (annotation == null)
+ {
return null;
+ }
AlignmentAnnotation[] ret = new AlignmentAnnotation[annotation.size()];
for (int r = 0; r < ret.length; r++)
+ {
ret[r] = (AlignmentAnnotation) annotation.elementAt(r);
+ }
return ret;
}
public void addAlignmentAnnotation(AlignmentAnnotation annotation)
{
if (this.annotation == null)
+ {
this.annotation = new Vector();
+ }
this.annotation.addElement(annotation);
+ annotation.setSequenceRef(this);
+ }
+
+ public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
+ {
+ if(this.annotation!=null)
+ {
+ this.annotation.removeElement(annotation);
+ if(this.annotation.size()==0)
+ this.annotation = null;
+ }
}
- public SequenceGroup getHiddenSequences()
+
+ /**
+ * test if this is a valid candidate for another
+ * sequence's dataset sequence.
+ *
+ */
+ private boolean isValidDatasetSequence()
+ {
+ if (datasetSequence!=null)
+ {
+ return false;
+ }
+ for (int i=0;i<sequence.length; i++)
+ {
+ if (jalview.util.Comparison.isGap(sequence[i]))
+ {
+ return false;
+ }
+ }
+ return true;
+ }
+ /* (non-Javadoc)
+ * @see jalview.datamodel.SequenceI#deriveSequence()
+ */
+ public SequenceI deriveSequence()
{
- return hiddenSequences;
+ SequenceI seq=new Sequence(this);
+ if (datasetSequence != null)
+ {
+ // duplicate current sequence with same dataset
+ seq.setDatasetSequence(datasetSequence);
+ }
+ else
+ {
+ if (isValidDatasetSequence())
+ {
+ // Use this as dataset sequence
+ seq.setDatasetSequence(this);
+ } else {
+ // Create a new, valid dataset sequence
+ SequenceI ds = seq;
+ ds.setSequence(AlignSeq.extractGaps(jalview.util.Comparison.GapChars, new String(sequence)));
+ setDatasetSequence(ds);
+ ds.setSequenceFeatures(getSequenceFeatures());
+ seq = this; // and return this sequence as the derived sequence.
+ }
+ }
+ return seq;
}
- public void addHiddenSequence(SequenceI seq)
+ /* (non-Javadoc)
+ * @see jalview.datamodel.SequenceI#createDatasetSequence()
+ */
+ public SequenceI createDatasetSequence()
{
- if (hiddenSequences == null)
+ if (datasetSequence==null)
{
- hiddenSequences = new SequenceGroup();
+ datasetSequence = new Sequence(getName(),
+ AlignSeq.extractGaps(
+ jalview.util.Comparison.GapChars,
+ getSequenceAsString()),
+ getStart(),
+ getEnd());
+ datasetSequence.setSequenceFeatures(getSequenceFeatures());
+ datasetSequence.setDescription(getDescription());
+ setSequenceFeatures(null);
+ // move database references onto dataset sequence
+ datasetSequence.setDBRef(getDBRef());
+ setDBRef(null);
+ }
+ return datasetSequence;
+ }
+ /* (non-Javadoc)
+ * @see jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[] annotations)
+ */
+ public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
+ {
+ if (annotation!=null) {
+ annotation.removeAllElements();
+ }
+ if (annotations!=null) {
+ for (int i=0; i<annotations.length; i++)
+ {
+ if (annotations[i]!=null)
+ addAlignmentAnnotation(annotations[i]);
+ }
}
- hiddenSequences.addSequence(seq, false);
}
- public void showHiddenSequence(SequenceI seq)
+ /* (non-Javadoc)
+ * @see jalview.datamodel.SequenceI#getAnnotation(java.lang.String)
+ */
+ public AlignmentAnnotation[] getAnnotation(String label)
{
- hiddenSequences.deleteSequence(seq, false);
- if (hiddenSequences.getSize(false) < 1)
+ if (annotation==null || annotation.size()==0)
{
- hiddenSequences = null;
+ return null;
+ }
+
+ Vector subset = new Vector();
+ Enumeration e = annotation.elements();
+ while (e.hasMoreElements())
+ {
+ AlignmentAnnotation ann = (AlignmentAnnotation) e.nextElement();
+ if (ann.label!=null && ann.label.equals(label))
+ {
+ subset.addElement(ann);
+ }
}
+ if (subset.size()==0)
+ {
+ return null;
+ }
+ AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
+ int i=0;
+ e = subset.elements();
+ while (e.hasMoreElements())
+ {
+ anns[i++] = (AlignmentAnnotation) e.nextElement();
+ }
+ subset.removeAllElements();
+ return anns;
}
+
}