/*
-* Jalview - A Sequence Alignment Editor and Viewer
-* Copyright (C) 2005 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
-* as published by the Free Software Foundation; either version 2
-* of the License, or (at your option) any later version.
-*
-* This program is distributed in the hope that it will be useful,
-* but WITHOUT ANY WARRANTY; without even the implied warranty of
-* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-* GNU General Public License for more details.
-*
-* You should have received a copy of the GNU General Public License
-* along with this program; if not, write to the Free Software
-* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
-*/
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
+ *
+ * This file is part of Jalview.
+ *
+ * Jalview is free software: you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
+ *
+ * Jalview is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ * PURPOSE. See the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
package jalview.datamodel;
-import java.awt.*;
-
-import java.util.*;
-
+import jalview.analysis.AlignSeq;
+import jalview.api.DBRefEntryI;
+import jalview.datamodel.features.SequenceFeatures;
+import jalview.datamodel.features.SequenceFeaturesI;
+import jalview.util.Comparison;
+import jalview.util.DBRefUtils;
+import jalview.util.MapList;
+import jalview.util.StringUtils;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.BitSet;
+import java.util.Collections;
+import java.util.Enumeration;
+import java.util.Iterator;
+import java.util.List;
+import java.util.ListIterator;
+import java.util.Vector;
+
+import fr.orsay.lri.varna.models.rna.RNA;
/**
- * DOCUMENT ME!
- *
+ *
+ * Implements the SequenceI interface for a char[] based sequence object.
+ *
* @author $author$
* @version $Revision$
*/
-public class Sequence implements SequenceI
+public class Sequence extends ASequence implements SequenceI
{
- SequenceI datasetSequence;
- String name;
- private String sequence;
- String description;
- int start;
- int end;
- Color color = Color.white;
- Vector pdbIds;
- String vamsasId;
- DBRefEntry [] dbrefs;
-
- /** This annotation is displayed below the alignment but the
- * positions are tied to the residues of this sequence */
- Vector annotation;
-
- /** DOCUMENT ME!! */
- 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!
- */
- public Sequence(String name, String sequence, int start, int end)
+ SequenceI datasetSequence;
+
+ String name;
+
+ private char[] sequence;
+
+ String description;
+
+ int start;
+
+ int end;
+
+ Vector<PDBEntry> pdbIds;
+
+ String vamsasId;
+
+ DBRefEntry[] dbrefs;
+
+ RNA rna;
+
+ /**
+ * This annotation is displayed below the alignment but the positions are tied
+ * to the residues of this sequence
+ *
+ * TODO: change to List<>
+ */
+ Vector<AlignmentAnnotation> annotation;
+
+ private SequenceFeaturesI sequenceFeatureStore;
+
+ /*
+ * A cursor holding the approximate current view position to the sequence,
+ * as determined by findIndex or findPosition or findPositions.
+ * Using a cursor as a hint allows these methods to be more performant for
+ * large sequences.
+ */
+ private SequenceCursor cursor;
+
+ /*
+ * A number that should be incremented whenever the sequence is edited.
+ * If the value matches the cursor token, then we can trust the cursor,
+ * if not then it should be recomputed.
+ */
+ private int changeCount;
+
+ /**
+ * Creates a new Sequence object.
+ *
+ * @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();
+ initSeqAndName(name, sequence.toCharArray(), start, end);
+ }
+
+ public Sequence(String name, char[] sequence, int start, int end)
+ {
+ this();
+ initSeqAndName(name, sequence, start, end);
+ }
+
+ /**
+ * Stage 1 constructor - assign name, sequence, and set start and end fields.
+ * start and end are updated values from name2 if it ends with /start-end
+ *
+ * @param name2
+ * @param sequence2
+ * @param start2
+ * @param end2
+ */
+ protected void initSeqAndName(String name2, char[] sequence2, int start2,
+ int end2)
+ {
+ this.name = name2;
+ this.sequence = sequence2;
+ this.start = start2;
+ this.end = end2;
+ parseId();
+ checkValidRange();
+ }
+
+ /**
+ * If 'name' ends in /i-j, where i >= j > 0 are integers, extracts i and j as
+ * start and end respectively and removes the suffix from the name
+ */
+ void parseId()
+ {
+ if (name == null)
{
- this.name = name;
- this.sequence = sequence;
- this.start = start;
- this.end = end;
-
- parseId();
-
- checkValidRange();
+ System.err.println(
+ "POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
+ name = "";
}
+ int slashPos = name.lastIndexOf('/');
+ if (slashPos > -1 && slashPos < name.length() - 1)
+ {
+ String suffix = name.substring(slashPos + 1);
+ String[] range = suffix.split("-");
+ if (range.length == 2)
+ {
+ try
+ {
+ int from = Integer.valueOf(range[0]);
+ int to = Integer.valueOf(range[1]);
+ if (from > 0 && to >= from)
+ {
+ name = name.substring(0, slashPos);
+ setStart(from);
+ setEnd(to);
+ checkValidRange();
+ }
+ } catch (NumberFormatException e)
+ {
+ // leave name unchanged if suffix is invalid
+ }
+ }
+ }
+ }
+
+ /**
+ * Ensures that 'end' is not before the end of the sequence, that is,
+ * (end-start+1) is at least as long as the count of ungapped positions. Note
+ * that end is permitted to be beyond the end of the sequence data.
+ */
+ void checkValidRange()
+ {
+ // Note: JAL-774 :
+ // http://issues.jalview.org/browse/JAL-774?focusedCommentId=11239&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-11239
+ {
+ int endRes = 0;
+ for (int j = 0; j < sequence.length; j++)
+ {
+ if (!Comparison.isGap(sequence[j]))
+ {
+ endRes++;
+ }
+ }
+ if (endRes > 0)
+ {
+ endRes += start - 1;
+ }
- com.stevesoft.pat.Regex limitrx = new com.stevesoft.pat.Regex(
- "[/][0-9]{1,}[-][0-9]{1,}$");
- com.stevesoft.pat.Regex endrx = new com.stevesoft.pat.Regex(
- "[0-9]{1,}$");
+ if (end < endRes)
+ {
+ end = endRes;
+ }
+ }
- void parseId()
+ }
+
+ /**
+ * default constructor
+ */
+ private Sequence()
+ {
+ sequenceFeatureStore = new SequenceFeatures();
+ }
+
+ /**
+ * Creates a new Sequence object.
+ *
+ * @param name
+ * DOCUMENT ME!
+ * @param sequence
+ * DOCUMENT ME!
+ */
+ public Sequence(String name, String sequence)
+ {
+ this(name, sequence, 1, -1);
+ }
+
+ /**
+ * Creates a new Sequence object with new AlignmentAnnotations but inherits
+ * any existing dataset sequence reference. If non exists, everything is
+ * copied.
+ *
+ * @param seq
+ * if seq is a dataset sequence, behaves like a plain old copy
+ * constructor
+ */
+ public Sequence(SequenceI seq)
+ {
+ 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();
+ initSeqFrom(seq, alAnnotation);
+ }
+
+ /**
+ * does the heavy lifting when cloning a dataset sequence, or coping data from
+ * dataset to a new derived sequence.
+ *
+ * @param seq
+ * - source of attributes.
+ * @param alAnnotation
+ * - alignment annotation present on seq that should be copied onto
+ * this sequence
+ */
+ protected void initSeqFrom(SequenceI seq,
+ AlignmentAnnotation[] alAnnotation)
+ {
+ char[] oseq = seq.getSequence(); // returns a copy of the array
+ initSeqAndName(seq.getName(), oseq, seq.getStart(), seq.getEnd());
+
+ description = seq.getDescription();
+ if (seq != datasetSequence)
{
- // Does sequence have the /start-end signiature?
- if(limitrx.search(name))
- {
- name = limitrx.left();
- endrx.search(limitrx.stringMatched());
- setStart( Integer.parseInt( limitrx.stringMatched().substring(1,endrx.matchedFrom()-1 )));
- setEnd( Integer.parseInt( endrx.stringMatched() ));
- }
+ setDatasetSequence(seq.getDatasetSequence());
}
+
+ /*
+ * only copy DBRefs and seqfeatures if we really are a dataset sequence
+ */
+ if (datasetSequence == null)
+ {
+ if (seq.getDBRefs() != null)
+ {
+ DBRefEntry[] dbr = seq.getDBRefs();
+ for (int i = 0; i < dbr.length; i++)
+ {
+ addDBRef(new DBRefEntry(dbr[i]));
+ }
+ }
- void checkValidRange()
+ /*
+ * make copies of any sequence features
+ */
+ for (SequenceFeature sf : seq.getSequenceFeatures())
+ {
+ addSequenceFeature(new SequenceFeature(sf));
+ }
+ }
+
+ if (seq.getAnnotation() != null)
{
- if (end < 1)
+ AlignmentAnnotation[] sqann = seq.getAnnotation();
+ for (int i = 0; i < sqann.length; i++)
{
- int endRes = 0;
- char ch;
- for (int j = 0; j < sequence.length(); j++)
+ if (sqann[i] == null)
+ {
+ continue;
+ }
+ boolean found = (alAnnotation == null);
+ if (!found)
{
- ch = sequence.charAt(j);
- if (!jalview.util.Comparison.isGap( (ch)))
+ for (int apos = 0; !found && apos < alAnnotation.length; apos++)
{
- endRes++;
+ found = (alAnnotation[apos] == sqann[i]);
}
}
- if (endRes > 0)
+ if (found)
{
- endRes += start - 1;
+ // only copy the given annotation
+ AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
+ addAlignmentAnnotation(newann);
}
-
- this.end = endRes;
}
-
}
+ if (seq.getAllPDBEntries() != null)
+ {
+ Vector<PDBEntry> ids = seq.getAllPDBEntries();
+ for (PDBEntry pdb : ids)
+ {
+ this.addPDBId(new PDBEntry(pdb));
+ }
+ }
+ }
- /**
- * Creates a new Sequence object.
- *
- * @param name DOCUMENT ME!
- * @param sequence DOCUMENT ME!
- */
- public Sequence(String name, String sequence)
+ @Override
+ public void setSequenceFeatures(List<SequenceFeature> features)
+ {
+ if (datasetSequence != null)
{
- this(name, sequence, 1, -1);
+ datasetSequence.setSequenceFeatures(features);
+ return;
}
+ sequenceFeatureStore = new SequenceFeatures(features);
+ }
- /**
- * Creates a new Sequence object.
- *
- * @param seq DOCUMENT ME!
- */
- public Sequence(SequenceI seq)
+ @Override
+ public synchronized boolean addSequenceFeature(SequenceFeature sf)
+ {
+ if (sf.getType() == null)
{
- this(seq.getName(), seq.getSequence(), seq.getStart(), seq.getEnd());
+ System.err.println("SequenceFeature type may not be null: "
+ + sf.toString());
+ return false;
}
- /**
- * DOCUMENT ME!
- *
- * @param v DOCUMENT ME!
- */
- public void setSequenceFeatures(SequenceFeature [] features)
+ if (datasetSequence != null)
{
- sequenceFeatures = features;
+ return datasetSequence.addSequenceFeature(sf);
}
- public synchronized void addSequenceFeature(SequenceFeature sf)
+ return sequenceFeatureStore.add(sf);
+ }
+
+ @Override
+ public void deleteFeature(SequenceFeature sf)
+ {
+ if (datasetSequence != null)
{
- if(sequenceFeatures==null)
- {
- sequenceFeatures = new SequenceFeature[0];
- }
+ datasetSequence.deleteFeature(sf);
+ }
+ else
+ {
+ sequenceFeatureStore.delete(sf);
+ }
+ }
+
+ /**
+ * {@inheritDoc}
+ *
+ * @return
+ */
+ @Override
+ public List<SequenceFeature> getSequenceFeatures()
+ {
+ if (datasetSequence != null)
+ {
+ return datasetSequence.getSequenceFeatures();
+ }
+ return sequenceFeatureStore.getAllFeatures();
+ }
+
+ @Override
+ public SequenceFeaturesI getFeatures()
+ {
+ return datasetSequence != null ? datasetSequence.getFeatures()
+ : sequenceFeatureStore;
+ }
+
+ @Override
+ public boolean addPDBId(PDBEntry entry)
+ {
+ if (pdbIds == null)
+ {
+ pdbIds = new Vector<>();
+ pdbIds.add(entry);
+ return true;
+ }
- for(int i=0; i<sequenceFeatures.length; i++)
+ for (PDBEntry pdbe : pdbIds)
+ {
+ if (pdbe.updateFrom(entry))
{
- if(sequenceFeatures[i].equals(sf))
- {
- return;
- }
+ return false;
}
+ }
+ pdbIds.addElement(entry);
+ return true;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param id
+ * DOCUMENT ME!
+ */
+ @Override
+ public void setPDBId(Vector<PDBEntry> id)
+ {
+ pdbIds = id;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ @Override
+ public Vector<PDBEntry> getAllPDBEntries()
+ {
+ return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ @Override
+ public String getDisplayId(boolean jvsuffix)
+ {
+ StringBuffer result = new StringBuffer(name);
+ if (jvsuffix)
+ {
+ result.append("/" + start + "-" + end);
+ }
- SequenceFeature [] temp = new SequenceFeature[sequenceFeatures.length+1];
- System.arraycopy(sequenceFeatures, 0, temp, 0, sequenceFeatures.length);
- temp[sequenceFeatures.length] = sf;
+ return result.toString();
+ }
+
+ /**
+ * Sets the sequence name. If the name ends in /start-end, then the start-end
+ * values are parsed out and set, and the suffix is removed from the name.
+ *
+ * @param theName
+ */
+ @Override
+ public void setName(String theName)
+ {
+ this.name = theName;
+ this.parseId();
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ @Override
+ public String getName()
+ {
+ return this.name;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param start
+ * DOCUMENT ME!
+ */
+ @Override
+ public void setStart(int start)
+ {
+ this.start = start;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ @Override
+ public int getStart()
+ {
+ return this.start;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param end
+ * DOCUMENT ME!
+ */
+ @Override
+ public void setEnd(int end)
+ {
+ this.end = end;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ @Override
+ public int getEnd()
+ {
+ return this.end;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ @Override
+ public int getLength()
+ {
+ return this.sequence.length;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param seq
+ * DOCUMENT ME!
+ */
+ @Override
+ public void setSequence(String seq)
+ {
+ this.sequence = seq.toCharArray();
+ checkValidRange();
+ sequenceChanged();
+ }
+
+ @Override
+ public String getSequenceAsString()
+ {
+ return new String(sequence);
+ }
+
+ @Override
+ public String getSequenceAsString(int start, int end)
+ {
+ return new String(getSequence(start, end));
+ }
+
+ @Override
+ public char[] getSequence()
+ {
+ // return sequence;
+ return sequence == null ? null : Arrays.copyOf(sequence,
+ sequence.length);
+ }
+
+ /*
+ * (non-Javadoc)
+ *
+ * @see jalview.datamodel.SequenceI#getSequence(int, int)
+ */
+ @Override
+ 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)
+ {
+ return new char[0];
+ }
- sequenceFeatures = temp;
+ if (end >= sequence.length)
+ {
+ end = sequence.length;
}
+ char[] reply = new char[end - start];
+ System.arraycopy(sequence, start, reply, 0, end - start);
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public SequenceFeature [] getSequenceFeatures()
+ return reply;
+ }
+
+ @Override
+ public SequenceI getSubSequence(int start, int end)
+ {
+ if (start < 0)
{
- return sequenceFeatures;
+ start = 0;
}
-
- public void addPDBId(PDBEntry entry)
+ char[] seq = getSequence(start, end);
+ if (seq.length == 0)
{
- if(pdbIds == null)
- pdbIds = new Vector();
-
- pdbIds.addElement(entry);
+ return null;
}
-
- /**
- * DOCUMENT ME!
- *
- * @param id DOCUMENT ME!
- */
- public void setPDBId(Vector id)
+ 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.setDescription(description);
+ if (datasetSequence != null)
{
- pdbIds = id;
+ nseq.setDatasetSequence(datasetSequence);
}
+ else
+ {
+ nseq.setDatasetSequence(this);
+ }
+ return nseq;
+ }
+
+ /**
+ * Returns the character of the aligned sequence at the given position (base
+ * zero), or space if the position is not within the sequence's bounds
+ *
+ * @return
+ */
+ @Override
+ public char getCharAt(int i)
+ {
+ if (i >= 0 && i < sequence.length)
+ {
+ return sequence[i];
+ }
+ else
+ {
+ return ' ';
+ }
+ }
+
+ /**
+ * Sets the sequence description, and also parses out any special formats of
+ * interest
+ *
+ * @param desc
+ */
+ @Override
+ public void setDescription(String desc)
+ {
+ this.description = desc;
+ }
+
+ @Override
+ public void setGeneLoci(String speciesId, String assemblyId,
+ String chromosomeId, MapList map)
+ {
+ addDBRef(new DBRefEntry(speciesId, assemblyId, DBRefEntry.CHROMOSOME
+ + ":" + chromosomeId, new Mapping(map)));
+ }
+
+ /**
+ * Returns the gene loci mapping for the sequence (may be null)
+ *
+ * @return
+ */
+ @Override
+ public GeneLociI getGeneLoci()
+ {
+ DBRefEntry[] refs = getDBRefs();
+ if (refs != null)
+ {
+ for (final DBRefEntry ref : refs)
+ {
+ if (ref.isChromosome())
+ {
+ return new GeneLociI()
+ {
+ @Override
+ public String getSpeciesId()
+ {
+ return ref.getSource();
+ }
+
+ @Override
+ public String getAssemblyId()
+ {
+ return ref.getVersion();
+ }
+
+ @Override
+ public String getChromosomeId()
+ {
+ // strip off "chromosome:" prefix to chrId
+ return ref.getAccessionId().substring(
+ DBRefEntry.CHROMOSOME.length() + 1);
+ }
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+ @Override
+ public MapList getMap()
+ {
+ return ref.getMap().getMap();
+ }
+ };
+ }
+ }
+ }
+ return null;
+ }
+
+ /**
+ * Answers the description
+ *
+ * @return
+ */
+ @Override
+ public String getDescription()
+ {
+ return this.description;
+ }
+
+ /**
+ * {@inheritDoc}
+ */
+ @Override
+ public int findIndex(int pos)
+ {
+ /*
+ * use a valid, hopefully nearby, cursor if available
*/
- public Vector getPDBId()
+ if (isValidCursor(cursor))
{
- return pdbIds;
+ return findIndex(pos, cursor);
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+ int j = start;
+ int i = 0;
+ int startColumn = 0;
+
+ /*
+ * traverse sequence from the start counting gaps; make a note of
+ * the column of the first residue to save in the cursor
*/
- public String getDisplayId(boolean jvsuffix)
+ while ((i < sequence.length) && (j <= end) && (j <= pos))
{
- StringBuffer result = new StringBuffer(name);
- if (jvsuffix)
+ if (!Comparison.isGap(sequence[i]))
{
- result.append("/" + start + "-" + end);
+ if (j == start)
+ {
+ startColumn = i;
+ }
+ j++;
}
+ i++;
+ }
- return result.toString();
+ if (j == end && j < pos)
+ {
+ return end + 1;
}
- /**
- * DOCUMENT ME!
- *
- * @param name DOCUMENT ME!
+ updateCursor(pos, i, startColumn);
+ return i;
+ }
+
+ /**
+ * Updates the cursor to the latest found residue and column position
+ *
+ * @param residuePos
+ * (start..)
+ * @param column
+ * (1..)
+ * @param startColumn
+ * column position of the first sequence residue
+ */
+ protected void updateCursor(int residuePos, int column, int startColumn)
+ {
+ /*
+ * preserve end residue column provided cursor was valid
*/
- public void setName(String name)
+ int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0;
+ if (residuePos == this.end)
{
- this.name = name;
- this.parseId();
+ endColumn = column;
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public String getName()
+ cursor = new SequenceCursor(this, residuePos, column, startColumn,
+ endColumn, this.changeCount);
+ }
+
+ /**
+ * Answers the aligned column position (1..) for the given residue position
+ * (start..) given a 'hint' of a residue/column location in the neighbourhood.
+ * The hint may be left of, at, or to the right of the required position.
+ *
+ * @param pos
+ * @param curs
+ * @return
+ */
+ protected int findIndex(int pos, SequenceCursor curs)
+ {
+ if (!isValidCursor(curs))
{
- return this.name;
+ /*
+ * wrong or invalidated cursor, compute de novo
+ */
+ return findIndex(pos);
}
- /**
- * DOCUMENT ME!
- *
- * @param start DOCUMENT ME!
- */
- public void setStart(int start)
+ if (curs.residuePosition == pos)
{
- this.start = start;
+ return curs.columnPosition;
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+ /*
+ * move left or right to find pos from hint.position
*/
- public int getStart()
+ int col = curs.columnPosition - 1; // convert from base 1 to 0-based array
+ // index
+ int newPos = curs.residuePosition;
+ int delta = newPos > pos ? -1 : 1;
+
+ while (newPos != pos)
{
- return this.start;
+ col += delta; // shift one column left or right
+ if (col < 0 || col == sequence.length)
+ {
+ break;
+ }
+ if (!Comparison.isGap(sequence[col]))
+ {
+ newPos += delta;
+ }
}
- /**
- * DOCUMENT ME!
- *
- * @param end DOCUMENT ME!
+ col++; // convert back to base 1
+ updateCursor(pos, col, curs.firstColumnPosition);
+
+ return col;
+ }
+
+ /**
+ * {@inheritDoc}
+ */
+ @Override
+ public int findPosition(final int column)
+ {
+ /*
+ * use a valid, hopefully nearby, cursor if available
*/
- public void setEnd(int end)
+ if (isValidCursor(cursor))
{
- this.end = end;
+ return findPosition(column + 1, cursor);
}
+
+ // TODO recode this more naturally i.e. count residues only
+ // as they are found, not 'in anticipation'
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+ /*
+ * traverse the sequence counting gaps; note the column position
+ * of the first residue, to save in the cursor
*/
- public int getEnd()
+ int firstResidueColumn = 0;
+ int lastPosFound = 0;
+ int lastPosFoundColumn = 0;
+ int seqlen = sequence.length;
+
+ if (seqlen > 0 && !Comparison.isGap(sequence[0]))
{
- return this.end;
+ lastPosFound = start;
+ lastPosFoundColumn = 0;
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public int getLength()
+ int j = 0;
+ int pos = start;
+
+ while (j < column && j < seqlen)
+ {
+ if (!Comparison.isGap(sequence[j]))
+ {
+ lastPosFound = pos;
+ lastPosFoundColumn = j;
+ if (pos == this.start)
+ {
+ firstResidueColumn = j;
+ }
+ pos++;
+ }
+ j++;
+ }
+ if (j < seqlen && !Comparison.isGap(sequence[j]))
{
- return this.sequence.length();
+ lastPosFound = pos;
+ lastPosFoundColumn = j;
+ if (pos == this.start)
+ {
+ firstResidueColumn = j;
+ }
}
- /**
- * DOCUMENT ME!
- *
- * @param seq DOCUMENT ME!
+ /*
+ * update the cursor to the last residue position found (if any)
+ * (converting column position to base 1)
*/
- public void setSequence(String seq)
+ if (lastPosFound != 0)
{
- this.sequence = seq;
- checkValidRange();
+ updateCursor(lastPosFound, lastPosFoundColumn + 1,
+ firstResidueColumn + 1);
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+ return pos;
+ }
+
+ /**
+ * Answers true if the given cursor is not null, is for this sequence object,
+ * and has a token value that matches this object's changeCount, else false.
+ * This allows us to ignore a cursor as 'stale' if the sequence has been
+ * modified since the cursor was created.
+ *
+ * @param curs
+ * @return
+ */
+ protected boolean isValidCursor(SequenceCursor curs)
+ {
+ if (curs == null || curs.sequence != this || curs.token != changeCount)
+ {
+ return false;
+ }
+ /*
+ * sanity check against range
*/
- public String getSequence()
+ if (curs.columnPosition < 0 || curs.columnPosition > sequence.length)
{
- return this.sequence;
+ return false;
+ }
+ if (curs.residuePosition < start || curs.residuePosition > end)
+ {
+ return false;
+ }
+ return true;
+ }
+
+ /**
+ * Answers the sequence position (start..) for the given aligned column
+ * position (1..), given a hint of a cursor in the neighbourhood. The cursor
+ * may lie left of, at, or to the right of the column position.
+ *
+ * @param col
+ * @param curs
+ * @return
+ */
+ protected int findPosition(final int col, SequenceCursor curs)
+ {
+ if (!isValidCursor(curs))
+ {
+ /*
+ * wrong or invalidated cursor, compute de novo
+ */
+ return findPosition(col - 1);// ugh back to base 0
}
- /**
- * DOCUMENT ME!
- *
- * @param start DOCUMENT ME!
- * @param end DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public String getSequence(int start, int end)
+ if (curs.columnPosition == col)
{
- // JBPNote - left to user to pad the result here (TODO:Decide on this policy)
- if (start >= sequence.length())
- {
- return "";
- }
+ cursor = curs; // in case this method becomes public
+ return curs.residuePosition; // easy case :-)
+ }
- if (end >= sequence.length())
- {
- end = sequence.length();
- }
+ if (curs.lastColumnPosition > 0 && curs.lastColumnPosition < col)
+ {
+ /*
+ * sequence lies entirely to the left of col
+ * - return last residue + 1
+ */
+ return end + 1;
+ }
- return this.sequence.substring(start, end);
+ if (curs.firstColumnPosition > 0 && curs.firstColumnPosition > col)
+ {
+ /*
+ * sequence lies entirely to the right of col
+ * - return first residue
+ */
+ return start;
}
- /**
- * make a new Sequence object from start to end (including gaps) over this seqeunce
- * @param start int
- * @param end int
- * @return SequenceI
- */
- public SequenceI getSubSequence(int start, int end) {
- if (start<0)
- start = 0;
- String seq = getSequence(start, end);
- if (seq=="")
- return null;
- int nstart = findPosition(start);
- int nend=findPosition(end-1)-1;
- // JBPNote - this is an incomplete copy.
- SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
- nseq.setDatasetSequence(getDatasetSequence());
- return nseq;
- }
- /**
- * DOCUMENT ME!
- *
- * @param i DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+
+ // todo could choose closest to col out of column,
+ // firstColumnPosition, lastColumnPosition as a start point
+
+ /*
+ * move left or right to find pos from cursor position
*/
- public char getCharAt(int i)
+ int firstResidueColumn = curs.firstColumnPosition;
+ int column = curs.columnPosition - 1; // to base 0
+ int newPos = curs.residuePosition;
+ int delta = curs.columnPosition > col ? -1 : 1;
+ boolean gapped = false;
+ int lastFoundPosition = curs.residuePosition;
+ int lastFoundPositionColumn = curs.columnPosition;
+
+ while (column != col - 1)
{
- if (i < sequence.length())
- {
- return sequence.charAt(i);
- }
- else
+ column += delta; // shift one column left or right
+ if (column < 0 || column == sequence.length)
+ {
+ break;
+ }
+ gapped = Comparison.isGap(sequence[column]);
+ if (!gapped)
+ {
+ newPos += delta;
+ lastFoundPosition = newPos;
+ lastFoundPositionColumn = column + 1;
+ if (lastFoundPosition == this.start)
{
- return ' ';
+ firstResidueColumn = column + 1;
}
+ }
}
- /**
- * DOCUMENT ME!
- *
- * @param desc DOCUMENT ME!
- */
- public void setDescription(String desc)
+ if (cursor == null || lastFoundPosition != cursor.residuePosition)
{
- this.description = desc;
+ updateCursor(lastFoundPosition, lastFoundPositionColumn,
+ firstResidueColumn);
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+ /*
+ * hack to give position to the right if on a gap
+ * or beyond the length of the sequence (see JAL-2562)
*/
- public String getDescription()
+ if (delta > 0 && (gapped || column >= sequence.length))
{
- return this.description;
+ newPos++;
}
- /**
- * DOCUMENT ME!
- *
- * @param pos DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public int findIndex(int pos)
- {
- // returns the alignment position for a residue
- int j = start;
- int i = 0;
+ return newPos;
+ }
- while ((i < sequence.length()) && (j <= end) && (j <= pos))
- {
- if (!jalview.util.Comparison.isGap(sequence.charAt(i)))
- {
- j++;
- }
-
- i++;
- }
-
- if ((j == end) && (j < pos))
- {
- return end + 1;
- }
- else
- {
- return i;
- }
+ /**
+ * {@inheritDoc}
+ */
+ @Override
+ public Range findPositions(int fromColumn, int toColumn)
+ {
+ if (toColumn < fromColumn || fromColumn < 1)
+ {
+ return null;
}
- /**
- * Returns the sequence position for an alignment position
- *
- * @param i column index in alignment (from 1)
- *
- * @return residue number for residue (left of and) nearest ith column
+ /*
+ * find the first non-gapped position, if any
*/
- public int findPosition(int i)
+ int firstPosition = 0;
+ int col = fromColumn - 1;
+ int length = sequence.length;
+ while (col < length && col < toColumn)
{
- int j = 0;
- int pos = start;
- int seqlen=sequence.length();
- while ((j < i) && (j < seqlen))
- {
- if (!jalview.util.Comparison.isGap((sequence.charAt(j))))
- {
- pos++;
- }
-
- j++;
- }
+ if (!Comparison.isGap(sequence[col]))
+ {
+ firstPosition = findPosition(col++);
+ break;
+ }
+ col++;
+ }
- return pos;
+ if (firstPosition == 0)
+ {
+ return null;
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
+ /*
+ * find the last non-gapped position
*/
- public int[] gapMap()
+ int lastPosition = firstPosition;
+ while (col < length && col < toColumn)
{
- // Returns an int array giving the position of each residue in the sequence in the alignment
- String seq = jalview.analysis.AlignSeq.extractGaps(jalview.util.Comparison.GapChars, sequence);
- int[] map = new int[seq.length()];
- int j = 0;
- int p = 0;
-
- while (j < sequence.length())
- {
- if (!jalview.util.Comparison.isGap(sequence.charAt(j)))
- {
- map[p++] = j;
- }
+ if (!Comparison.isGap(sequence[col++]))
+ {
+ lastPosition++;
+ }
+ }
- j++;
- }
+ return new Range(firstPosition, lastPosition);
+ }
+
+ /**
+ * Returns an int array where indices correspond to each residue in the
+ * sequence and the element value gives its position in the alignment
+ *
+ * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
+ * residues in SequenceI object
+ */
+ @Override
+ public int[] gapMap()
+ {
+ String seq = jalview.analysis.AlignSeq.extractGaps(
+ jalview.util.Comparison.GapChars, new String(sequence));
+ int[] map = new int[seq.length()];
+ int j = 0;
+ int p = 0;
+
+ while (j < sequence.length)
+ {
+ if (!jalview.util.Comparison.isGap(sequence[j]))
+ {
+ map[p++] = j;
+ }
- return map;
+ j++;
}
- /**
- * DOCUMENT ME!
- *
- * @param i DOCUMENT ME!
- */
- public void deleteCharAt(int i)
+ return map;
+ }
+
+ /**
+ * Build a bitset corresponding to sequence gaps
+ *
+ * @return a BitSet where set values correspond to gaps in the sequence
+ */
+ @Override
+ public BitSet gapBitset()
+ {
+ BitSet gaps = new BitSet(sequence.length);
+ int j = 0;
+ while (j < sequence.length)
{
- if (i >= sequence.length())
- {
- return;
- }
-
- sequence = sequence.substring(0, i) + sequence.substring(i + 1);
+ if (jalview.util.Comparison.isGap(sequence[j]))
+ {
+ gaps.set(j);
+ }
+ j++;
}
+ return gaps;
+ }
+
+ @Override
+ public int[] findPositionMap()
+ {
+ int map[] = new int[sequence.length];
+ int j = 0;
+ int pos = start;
+ int seqlen = sequence.length;
+ while ((j < seqlen))
+ {
+ map[j] = pos;
+ if (!jalview.util.Comparison.isGap(sequence[j]))
+ {
+ pos++;
+ }
- /**
- * DOCUMENT ME!
- *
- * @param i DOCUMENT ME!
- * @param j DOCUMENT ME!
- */
- public void deleteChars(int i, int j)
+ j++;
+ }
+ return map;
+ }
+
+ @Override
+ public List<int[]> getInsertions()
+ {
+ ArrayList<int[]> map = new ArrayList<>();
+ int lastj = -1, j = 0;
+ int pos = start;
+ int seqlen = sequence.length;
+ while ((j < seqlen))
{
- if (i >= sequence.length())
+ if (jalview.util.Comparison.isGap(sequence[j]))
+ {
+ if (lastj == -1)
{
- return;
+ lastj = j;
}
-
- if (j >= sequence.length())
+ }
+ else
+ {
+ if (lastj != -1)
{
- sequence = sequence.substring(0, i);
+ map.add(new int[] { lastj, j - 1 });
+ lastj = -1;
}
- else
+ }
+ j++;
+ }
+ if (lastj != -1)
+ {
+ map.add(new int[] { lastj, j - 1 });
+ lastj = -1;
+ }
+ return map;
+ }
+
+ @Override
+ public BitSet getInsertionsAsBits()
+ {
+ BitSet map = new BitSet();
+ int lastj = -1, j = 0;
+ int pos = start;
+ int seqlen = sequence.length;
+ while ((j < seqlen))
+ {
+ if (jalview.util.Comparison.isGap(sequence[j]))
+ {
+ if (lastj == -1)
+ {
+ lastj = j;
+ }
+ }
+ else
+ {
+ if (lastj != -1)
{
- sequence = sequence.substring(0, i) + sequence.substring(j);
+ map.set(lastj, j);
+ lastj = -1;
}
+ }
+ j++;
+ }
+ if (lastj != -1)
+ {
+ map.set(lastj, j);
+ lastj = -1;
+ }
+ return map;
+ }
+
+ @Override
+ public void deleteChars(final int i, final int j)
+ {
+ int newstart = start, newend = end;
+ if (i >= sequence.length || i < 0)
+ {
+ return;
}
+ char[] tmp = StringUtils.deleteChars(sequence, i, j);
+ boolean createNewDs = false;
+ // TODO: take a (second look) at the dataset creation validation method for
+ // the very large sequence case
+ int startIndex = findIndex(start) - 1;
+ int endIndex = findIndex(end) - 1;
+ int startDeleteColumn = -1; // for dataset sequence deletions
+ int deleteCount = 0;
- /**
- * DOCUMENT ME!
- *
- * @param i DOCUMENT ME!
- * @param c DOCUMENT ME!
- * @param chop DOCUMENT ME!
- */
- public void insertCharAt(int i, char c)
+ for (int s = i; s < j; s++)
{
- String tmp = new String(sequence);
-
- if (i < sequence.length())
+ if (Comparison.isGap(sequence[s]))
+ {
+ continue;
+ }
+ deleteCount++;
+ if (startDeleteColumn == -1)
+ {
+ startDeleteColumn = findPosition(s) - start;
+ }
+ if (createNewDs)
+ {
+ newend--;
+ }
+ else
+ {
+ if (startIndex == s)
{
- sequence = tmp.substring(0, i) + String.valueOf(c) +
- tmp.substring(i);
+ /*
+ * deleting characters from start of sequence; new start is the
+ * sequence position of the next column (position to the right
+ * if the column position is gapped)
+ */
+ newstart = findPosition(j);
+ break;
}
else
{
- // JBPNote : padding char at end of sequence. We'll not get away with this when we insert residues, I bet!
- char[] ch = new char[(1 + i) - sequence.length()];
-
- for (int j = 0, k = ch.length; j < k; j++)
- ch[j] = c;
-
- sequence = tmp + String.valueOf(ch);
+ if (endIndex < j)
+ {
+ /*
+ * deleting characters at end of sequence; new end is the sequence
+ * position of the column before the deletion; subtract 1 if this is
+ * gapped since findPosition returns the next sequence position
+ */
+ newend = findPosition(i - 1);
+ if (Comparison.isGap(sequence[i - 1]))
+ {
+ newend--;
+ }
+ break;
+ }
+ else
+ {
+ createNewDs = true;
+ newend--;
+ }
}
+ }
}
- /**
- * DOCUMENT ME!
- *
- * @param c DOCUMENT ME!
- */
- public void setColor(Color c)
+ if (createNewDs && this.datasetSequence != null)
{
- this.color = c;
+ /*
+ * if deletion occured in the middle of the sequence,
+ * construct a new dataset sequence and delete the residues
+ * that were deleted from the aligned sequence
+ */
+ Sequence ds = new Sequence(datasetSequence);
+ ds.deleteChars(startDeleteColumn, startDeleteColumn + deleteCount);
+ datasetSequence = ds;
+ // TODO: remove any non-inheritable properties ?
+ // TODO: create a sequence mapping (since there is a relation here ?)
+ }
+ start = newstart;
+ end = newend;
+ sequence = tmp;
+ sequenceChanged();
+ }
+
+ @Override
+ public void insertCharAt(int i, int length, char c)
+ {
+ char[] tmp = new char[sequence.length + length];
+
+ if (i >= sequence.length)
+ {
+ System.arraycopy(sequence, 0, tmp, 0, sequence.length);
+ i = sequence.length;
+ }
+ else
+ {
+ System.arraycopy(sequence, 0, tmp, 0, i);
}
- /**
- * DOCUMENT ME!
- *
- * @return DOCUMENT ME!
- */
- public Color getColor()
+ int index = i;
+ while (length > 0)
{
- return color;
+ tmp[index++] = c;
+ length--;
}
- public String getVamsasId()
+ if (i < sequence.length)
{
- return vamsasId;
+ System.arraycopy(sequence, i, tmp, index, sequence.length - i);
}
- public void setVamsasId(String id)
+ sequence = tmp;
+ sequenceChanged();
+ }
+
+ @Override
+ public void insertCharAt(int i, char c)
+ {
+ insertCharAt(i, 1, c);
+ }
+
+ @Override
+ public String getVamsasId()
+ {
+ return vamsasId;
+ }
+
+ @Override
+ public void setVamsasId(String id)
+ {
+ vamsasId = id;
+ }
+
+ @Override
+ public void setDBRefs(DBRefEntry[] dbref)
+ {
+ if (dbrefs == null && datasetSequence != null
+ && this != datasetSequence)
+ {
+ datasetSequence.setDBRefs(dbref);
+ return;
+ }
+ dbrefs = dbref;
+ if (dbrefs != null)
{
- vamsasId = id;
+ DBRefUtils.ensurePrimaries(this);
}
+ }
- public void setDBRef(DBRefEntry [] dbref)
+ @Override
+ public DBRefEntry[] getDBRefs()
+ {
+ if (dbrefs == null && datasetSequence != null
+ && this != datasetSequence)
{
- dbrefs = dbref;
+ return datasetSequence.getDBRefs();
}
+ return dbrefs;
+ }
- public DBRefEntry [] getDBRef()
+ @Override
+ public void addDBRef(DBRefEntry entry)
+ {
+ if (datasetSequence != null)
{
- return dbrefs;
+ datasetSequence.addDBRef(entry);
+ return;
}
- public void addDBRef(DBRefEntry entry)
+ if (dbrefs == null)
{
- if(dbrefs == null)
- dbrefs = new DBRefEntry[0];
+ dbrefs = new DBRefEntry[0];
+ }
- DBRefEntry [] temp = new DBRefEntry[dbrefs.length+1];
- System.arraycopy(dbrefs, 0, temp, 0, dbrefs.length);
+ for (DBRefEntryI dbr : dbrefs)
+ {
+ if (dbr.updateFrom(entry))
+ {
+ /*
+ * found a dbref that either matched, or could be
+ * updated from, the new entry - no need to add it
+ */
+ return;
+ }
+ }
- temp[temp.length-1] = entry;
+ /*
+ * extend the array to make room for one more
+ */
+ // TODO use an ArrayList instead
+ int j = dbrefs.length;
+ DBRefEntry[] temp = new DBRefEntry[j + 1];
+ System.arraycopy(dbrefs, 0, temp, 0, j);
+ temp[temp.length - 1] = entry;
- dbrefs = temp;
- }
+ dbrefs = temp;
- public void setDatasetSequence(SequenceI seq)
+ DBRefUtils.ensurePrimaries(this);
+ }
+
+ @Override
+ public void setDatasetSequence(SequenceI seq)
+ {
+ if (seq == this)
{
- datasetSequence = seq;
+ throw new IllegalArgumentException(
+ "Implementation Error: self reference passed to SequenceI.setDatasetSequence");
}
-
- public SequenceI getDatasetSequence()
+ if (seq != null && seq.getDatasetSequence() != null)
{
- return datasetSequence;
+ throw new IllegalArgumentException(
+ "Implementation error: cascading dataset sequences are not allowed.");
}
+ datasetSequence = seq;
+ }
+
+ @Override
+ public SequenceI getDatasetSequence()
+ {
+ return datasetSequence;
+ }
+
+ @Override
+ public AlignmentAnnotation[] getAnnotation()
+ {
+ return annotation == null ? null
+ : annotation
+ .toArray(new AlignmentAnnotation[annotation.size()]);
+ }
+
+ @Override
+ public boolean hasAnnotation(AlignmentAnnotation ann)
+ {
+ return annotation == null ? false : annotation.contains(ann);
+ }
+
+ @Override
+ public void addAlignmentAnnotation(AlignmentAnnotation annotation)
+ {
+ if (this.annotation == null)
+ {
+ this.annotation = new Vector<>();
+ }
+ if (!this.annotation.contains(annotation))
+ {
+ this.annotation.addElement(annotation);
+ }
+ annotation.setSequenceRef(this);
+ }
- public AlignmentAnnotation [] getAnnotation()
+ @Override
+ public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
+ {
+ if (this.annotation != null)
+ {
+ this.annotation.removeElement(annotation);
+ if (this.annotation.size() == 0)
+ {
+ this.annotation = null;
+ }
+ }
+ }
+
+ /**
+ * test if this is a valid candidate for another sequence's dataset sequence.
+ *
+ */
+ private boolean isValidDatasetSequence()
+ {
+ if (datasetSequence != null)
{
- if(annotation==null)
- return null;
+ return false;
+ }
+ for (int i = 0; i < sequence.length; i++)
+ {
+ if (jalview.util.Comparison.isGap(sequence[i]))
+ {
+ return false;
+ }
+ }
+ return true;
+ }
+
+ @Override
+ public SequenceI deriveSequence()
+ {
+ Sequence seq = null;
+ if (datasetSequence == null)
+ {
+ if (isValidDatasetSequence())
+ {
+ // Use this as dataset sequence
+ seq = new Sequence(getName(), "", 1, -1);
+ seq.setDatasetSequence(this);
+ seq.initSeqFrom(this, getAnnotation());
+ return seq;
+ }
+ else
+ {
+ // Create a new, valid dataset sequence
+ createDatasetSequence();
+ }
+ }
+ return new Sequence(this);
+ }
- AlignmentAnnotation [] ret = new AlignmentAnnotation[annotation.size()];
- for(int r = 0; r<ret.length; r++)
- ret[r] = (AlignmentAnnotation)annotation.elementAt(r);
+ private boolean _isNa;
- return ret;
- }
+ private int _seqhash = 0;
- public void addAlignmentAnnotation(AlignmentAnnotation annotation)
+ /**
+ * Answers false if the sequence is more than 85% nucleotide (ACGTU), else
+ * true
+ */
+ @Override
+ public boolean isProtein()
+ {
+ if (datasetSequence != null)
{
- if(this.annotation==null)
- this.annotation = new Vector();
-
- this.annotation.addElement( annotation );
+ return datasetSequence.isProtein();
+ }
+ if (_seqhash != sequence.hashCode())
+ {
+ _seqhash = sequence.hashCode();
+ _isNa = Comparison.isNucleotide(this);
+ }
+ return !_isNa;
+ };
+
+ /*
+ * (non-Javadoc)
+ *
+ * @see jalview.datamodel.SequenceI#createDatasetSequence()
+ */
+ @Override
+ public SequenceI createDatasetSequence()
+ {
+ if (datasetSequence == null)
+ {
+ Sequence dsseq = new Sequence(getName(),
+ AlignSeq.extractGaps(jalview.util.Comparison.GapChars,
+ getSequenceAsString()),
+ getStart(), getEnd());
+
+ datasetSequence = dsseq;
+
+ dsseq.setDescription(description);
+ // move features and database references onto dataset sequence
+ dsseq.sequenceFeatureStore = sequenceFeatureStore;
+ sequenceFeatureStore = null;
+ dsseq.dbrefs = dbrefs;
+ dbrefs = null;
+ // TODO: search and replace any references to this sequence with
+ // references to the dataset sequence in Mappings on dbref
+ dsseq.pdbIds = pdbIds;
+ pdbIds = null;
+ datasetSequence.updatePDBIds();
+ if (annotation != null)
+ {
+ // annotation is cloned rather than moved, to preserve what's currently
+ // on the alignment
+ for (AlignmentAnnotation aa : annotation)
+ {
+ AlignmentAnnotation _aa = new AlignmentAnnotation(aa);
+ _aa.sequenceRef = datasetSequence;
+ _aa.adjustForAlignment(); // uses annotation's own record of
+ // sequence-column mapping
+ datasetSequence.addAlignmentAnnotation(_aa);
+ }
+ }
+ }
+ return datasetSequence;
+ }
+
+ /*
+ * (non-Javadoc)
+ *
+ * @see
+ * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
+ * annotations)
+ */
+ @Override
+ 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]);
+ }
+ }
}
+ }
- public SequenceGroup getHiddenSequences()
+ @Override
+ public AlignmentAnnotation[] getAnnotation(String label)
+ {
+ if (annotation == null || annotation.size() == 0)
{
- return hiddenSequences;
+ return null;
}
- public void addHiddenSequence(SequenceI seq)
+ Vector<AlignmentAnnotation> subset = new Vector<>();
+ Enumeration<AlignmentAnnotation> e = annotation.elements();
+ while (e.hasMoreElements())
{
- if(hiddenSequences==null)
+ AlignmentAnnotation ann = e.nextElement();
+ if (ann.label != null && ann.label.equals(label))
{
- hiddenSequences = new SequenceGroup();
+ subset.addElement(ann);
}
- hiddenSequences.addSequence(seq, false);
}
-
- public void showHiddenSequence(SequenceI seq)
+ if (subset.size() == 0)
+ {
+ return null;
+ }
+ AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
+ int i = 0;
+ e = subset.elements();
+ while (e.hasMoreElements())
+ {
+ anns[i++] = e.nextElement();
+ }
+ subset.removeAllElements();
+ return anns;
+ }
+
+ @Override
+ public boolean updatePDBIds()
+ {
+ if (datasetSequence != null)
+ {
+ // TODO: could merge DBRefs
+ return datasetSequence.updatePDBIds();
+ }
+ if (dbrefs == null || dbrefs.length == 0)
{
- hiddenSequences.deleteSequence(seq, false);
- if (hiddenSequences.getSize(false) < 1)
+ return false;
+ }
+ boolean added = false;
+ for (DBRefEntry dbr : dbrefs)
+ {
+ if (DBRefSource.PDB.equals(dbr.getSource()))
{
- hiddenSequences = null;
+ /*
+ * 'Add' any PDB dbrefs as a PDBEntry - add is only performed if the
+ * PDB id is not already present in a 'matching' PDBEntry
+ * Constructor parses out a chain code if appended to the accession id
+ * (a fudge used to 'store' the chain code in the DBRef)
+ */
+ PDBEntry pdbe = new PDBEntry(dbr);
+ added |= addPDBId(pdbe);
}
}
+ return added;
+ }
- public void changeCase(boolean toUpper, int start, int end)
+ @Override
+ public void transferAnnotation(SequenceI entry, Mapping mp)
+ {
+ if (datasetSequence != null)
+ {
+ datasetSequence.transferAnnotation(entry, mp);
+ return;
+ }
+ if (entry.getDatasetSequence() != null)
+ {
+ transferAnnotation(entry.getDatasetSequence(), mp);
+ return;
+ }
+ // transfer any new features from entry onto sequence
+ if (entry.getSequenceFeatures() != null)
{
- StringBuffer newSeq = new StringBuffer();
-
- if(end>sequence.length())
- end = sequence.length();
- if (start > 0)
+ List<SequenceFeature> sfs = entry.getSequenceFeatures();
+ for (SequenceFeature feature : sfs)
{
- newSeq.append(sequence.substring(0, start));
+ SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
+ : new SequenceFeature[] { new SequenceFeature(feature) };
+ if (sf != null)
+ {
+ for (int sfi = 0; sfi < sf.length; sfi++)
+ {
+ addSequenceFeature(sf[sfi]);
+ }
+ }
}
+ }
- if (toUpper)
- newSeq.append(sequence.substring(start, end).toUpperCase());
- else
- newSeq.append(sequence.substring(start, end).toLowerCase());
-
- if (end < sequence.length())
- newSeq.append(sequence.substring(end));
+ // transfer PDB entries
+ if (entry.getAllPDBEntries() != null)
+ {
+ Enumeration<PDBEntry> e = entry.getAllPDBEntries().elements();
+ while (e.hasMoreElements())
+ {
+ PDBEntry pdb = e.nextElement();
+ addPDBId(pdb);
+ }
+ }
+ // transfer database references
+ DBRefEntry[] entryRefs = entry.getDBRefs();
+ if (entryRefs != null)
+ {
+ for (int r = 0; r < entryRefs.length; r++)
+ {
+ DBRefEntry newref = new DBRefEntry(entryRefs[r]);
+ if (newref.getMap() != null && mp != null)
+ {
+ // remap ref using our local mapping
+ }
+ // we also assume all version string setting is done by dbSourceProxy
+ /*
+ * if (!newref.getSource().equalsIgnoreCase(dbSource)) {
+ * newref.setSource(dbSource); }
+ */
+ addDBRef(newref);
+ }
+ }
+ }
+
+ @Override
+ public void setRNA(RNA r)
+ {
+ rna = r;
+ }
+
+ @Override
+ public RNA getRNA()
+ {
+ return rna;
+ }
+
+ @Override
+ public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
+ String label)
+ {
+ List<AlignmentAnnotation> result = new ArrayList<>();
+ if (this.annotation != null)
+ {
+ for (AlignmentAnnotation ann : annotation)
+ {
+ if (ann.calcId != null && ann.calcId.equals(calcId)
+ && ann.label != null && ann.label.equals(label))
+ {
+ result.add(ann);
+ }
+ }
+ }
+ return result;
+ }
+
+ @Override
+ public String toString()
+ {
+ return getDisplayId(false);
+ }
+
+ @Override
+ public PDBEntry getPDBEntry(String pdbIdStr)
+ {
+ if (getDatasetSequence() != null)
+ {
+ return getDatasetSequence().getPDBEntry(pdbIdStr);
+ }
+ if (pdbIds == null)
+ {
+ return null;
+ }
+ List<PDBEntry> entries = getAllPDBEntries();
+ for (PDBEntry entry : entries)
+ {
+ if (entry.getId().equalsIgnoreCase(pdbIdStr))
+ {
+ return entry;
+ }
+ }
+ return null;
+ }
- sequence = newSeq.toString();
+ @Override
+ public List<DBRefEntry> getPrimaryDBRefs()
+ {
+ if (datasetSequence != null)
+ {
+ return datasetSequence.getPrimaryDBRefs();
+ }
+ if (dbrefs == null || dbrefs.length == 0)
+ {
+ return Collections.emptyList();
+ }
+ synchronized (dbrefs)
+ {
+ List<DBRefEntry> primaries = new ArrayList<>();
+ DBRefEntry[] tmp = new DBRefEntry[1];
+ for (DBRefEntry ref : dbrefs)
+ {
+ if (!ref.isPrimaryCandidate())
+ {
+ continue;
+ }
+ if (ref.hasMap())
+ {
+ MapList mp = ref.getMap().getMap();
+ if (mp.getFromLowest() > start || mp.getFromHighest() < end)
+ {
+ // map only involves a subsequence, so cannot be primary
+ continue;
+ }
+ }
+ // whilst it looks like it is a primary ref, we also sanity check type
+ if (DBRefUtils.getCanonicalName(DBRefSource.PDB)
+ .equals(DBRefUtils.getCanonicalName(ref.getSource())))
+ {
+ // PDB dbrefs imply there should be a PDBEntry associated
+ // TODO: tighten PDB dbrefs
+ // formally imply Jalview has actually downloaded and
+ // parsed the pdb file. That means there should be a cached file
+ // handle on the PDBEntry, and a real mapping between sequence and
+ // extracted sequence from PDB file
+ PDBEntry pdbentry = getPDBEntry(ref.getAccessionId());
+ if (pdbentry != null && pdbentry.getFile() != null)
+ {
+ primaries.add(ref);
+ }
+ continue;
+ }
+ // check standard protein or dna sources
+ tmp[0] = ref;
+ DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
+ if (res != null && res[0] == tmp[0])
+ {
+ primaries.add(ref);
+ continue;
+ }
+ }
+ return primaries;
+ }
+ }
+
+ /**
+ * {@inheritDoc}
+ */
+ @Override
+ public List<SequenceFeature> findFeatures(int fromColumn, int toColumn,
+ String... types)
+ {
+ int startPos = findPosition(fromColumn - 1); // convert base 1 to base 0
+ int endPos = fromColumn == toColumn ? startPos
+ : findPosition(toColumn - 1);
+
+ List<SequenceFeature> result = getFeatures().findFeatures(startPos,
+ endPos, types);
+
+ /*
+ * if end column is gapped, endPos may be to the right,
+ * and we may have included adjacent or enclosing features;
+ * remove any that are not enclosing, non-contact features
+ */
+ boolean endColumnIsGapped = toColumn > 0 && toColumn <= sequence.length
+ && Comparison.isGap(sequence[toColumn - 1]);
+ if (endPos > this.end || endColumnIsGapped)
+ {
+ ListIterator<SequenceFeature> it = result.listIterator();
+ while (it.hasNext())
+ {
+ SequenceFeature sf = it.next();
+ int sfBegin = sf.getBegin();
+ int sfEnd = sf.getEnd();
+ int featureStartColumn = findIndex(sfBegin);
+ if (featureStartColumn > toColumn)
+ {
+ it.remove();
+ }
+ else if (featureStartColumn < fromColumn)
+ {
+ int featureEndColumn = sfEnd == sfBegin ? featureStartColumn
+ : findIndex(sfEnd);
+ if (featureEndColumn < fromColumn)
+ {
+ it.remove();
+ }
+ else if (featureEndColumn > toColumn && sf.isContactFeature())
+ {
+ /*
+ * remove an enclosing feature if it is a contact feature
+ */
+ it.remove();
+ }
+ }
+ }
}
- public void toggleCase(int start, int end)
+ return result;
+ }
+
+ /**
+ * Invalidates any stale cursors (forcing recalculation) by incrementing the
+ * token that has to match the one presented by the cursor
+ */
+ @Override
+ public void sequenceChanged()
+ {
+ changeCount++;
+ }
+
+ /**
+ * {@inheritDoc}
+ */
+ @Override
+ public int replace(char c1, char c2)
+ {
+ if (c1 == c2)
+ {
+ return 0;
+ }
+ int count = 0;
+ synchronized (sequence)
{
- StringBuffer newSeq = new StringBuffer();
+ for (int c = 0; c < sequence.length; c++)
+ {
+ if (sequence[c] == c1)
+ {
+ sequence[c] = c2;
+ count++;
+ }
+ }
+ }
+ if (count > 0)
+ {
+ sequenceChanged();
+ }
+
+ return count;
+ }
- if(end>sequence.length())
- end = sequence.length();
+ @Override
+ public String getSequenceStringFromIterator(Iterator<int[]> it)
+ {
+ StringBuilder newSequence = new StringBuilder();
+ while (it.hasNext())
+ {
+ int[] block = it.next();
+ if (it.hasNext())
+ {
+ newSequence.append(getSequence(block[0], block[1] + 1));
+ }
+ else
+ {
+ newSequence.append(getSequence(block[0], block[1]));
+ }
+ }
- if (start > 0)
- {
- newSeq.append(sequence.substring(0, start));
- }
+ return newSequence.toString();
+ }
- char nextChar;
- for(int c=start; c<end; c++)
- {
- nextChar = sequence.charAt(c);
- if(Character.isLetter(nextChar))
- {
- if(Character.isUpperCase(nextChar))
- nextChar = Character.toLowerCase(nextChar);
- else
- nextChar = Character.toUpperCase(nextChar);
- }
+ @Override
+ public int firstResidueOutsideIterator(Iterator<int[]> regions)
+ {
+ int start = 0;
+ if (!regions.hasNext())
+ {
+ return findIndex(getStart()) - 1;
+ }
- newSeq.append(nextChar);
- }
+ // Simply walk along the sequence whilst watching for region
+ // boundaries
+ int hideStart = getLength();
+ int hideEnd = -1;
+ boolean foundStart = false;
- if (end < sequence.length())
- newSeq.append(sequence.substring(end));
+ // step through the non-gapped positions of the sequence
+ for (int i = getStart(); i <= getEnd() && (!foundStart); i++)
+ {
+ // get alignment position of this residue in the sequence
+ int p = findIndex(i) - 1;
- sequence = newSeq.toString();
+ // update region start/end
+ while (hideEnd < p && regions.hasNext())
+ {
+ int[] region = regions.next();
+ hideStart = region[0];
+ hideEnd = region[1];
+ }
+ if (hideEnd < p)
+ {
+ hideStart = getLength();
+ }
+ // update boundary for sequence
+ if (p < hideStart)
+ {
+ start = p;
+ foundStart = true;
+ }
}
+ if (foundStart)
+ {
+ return start;
+ }
+ // otherwise, sequence was completely hidden
+ return 0;
+ }
}