X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FSequence.java;h=c8b94cecf83e3d4710d2186fc8d0b4a551b03d4d;hb=19c22459dca6c69d6f1fd0eed029c01e21c99854;hp=293528d13c186a1ef6db43720f0e8b615e0d4be1;hpb=1ad6a9b44ff7828a2a001ad1520b0ee652a466e2;p=jalview.git
diff --git a/src/jalview/datamodel/Sequence.java b/src/jalview/datamodel/Sequence.java
index 293528d..32d295a 100755
--- a/src/jalview/datamodel/Sequence.java
+++ b/src/jalview/datamodel/Sequence.java
@@ -1,67 +1,125 @@
/*
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)
- * Copyright (C) 2011 J Procter, AM Waterhouse, J Engelhardt, LM Lui, G Barton, M Clamp, S Searle
+ * 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.
- *
+ * 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 .
+ * You should have received a copy of the GNU General Public License
+ * along with Jalview. If not, see .
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
*/
package jalview.datamodel;
-import jalview.analysis.AlignSeq;
-
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.BitSet;
+import java.util.Collection;
+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;
+import jalview.analysis.AlignSeq;
+import jalview.analysis.AlignmentUtils;
+import jalview.analysis.SeqsetUtils;
+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 jalview.ws.datamodel.alphafold.MappableContactMatrix;
+
/**
*
- * Implements the SequenceI interface for a char[] based sequence object.
- *
- * @author $author$
- * @version $Revision$
+ * Implements the SequenceI interface for a char[] based sequence object
*/
-public class Sequence implements SequenceI
+public class Sequence extends ASequence implements SequenceI
{
+
+ /**
+ * A subclass that gives us access to modCount, which tracks whether there
+ * have been any changes. We use this to update
+ *
+ * @author hansonr
+ *
+ * @param
+ */
+ @SuppressWarnings("serial")
+ public class DBModList extends ArrayList
+ {
+
+ protected int getModCount()
+ {
+ return modCount;
+ }
+
+ }
+
SequenceI datasetSequence;
- String name;
+ private String name;
private char[] sequence;
- String description;
+ private String description;
+
+ private int start;
- int start;
+ private int end;
- int end;
+ private Vector pdbIds;
- Vector pdbIds;
+ private String vamsasId;
- String vamsasId;
+ private DBModList dbrefs; // controlled access
+
+ /**
+ * a flag to let us know that elements have changed in dbrefs
+ *
+ * @author Bob Hanson
+ */
+ private int refModCount = 0;
- DBRefEntry[] dbrefs;
+ private 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 annotation;
-
- /**
- * The index of the sequence in a MSA
+ private Vector 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.
*/
- int index = -1;
+ private SequenceCursor cursor;
- /** array of seuqence features - may not be null for a valid sequence object */
- public SequenceFeature[] sequenceFeatures;
+ /*
+ * 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.
@@ -78,56 +136,88 @@ public class Sequence implements SequenceI
*/
public Sequence(String name, String sequence, int start, int end)
{
- this.name = name;
- this.sequence = sequence.toCharArray();
- this.start = start;
- this.end = end;
- parseId();
- checkValidRange();
+ this();
+ initSeqAndName(name, sequence.toCharArray(), start, end);
}
public Sequence(String name, char[] sequence, int start, int end)
{
- this.name = name;
- this.sequence = sequence;
- this.start = start;
- this.end = 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();
}
- 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 '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)
{
- System.err
- .println("POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
+ System.err.println(
+ "POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
name = "";
}
- // Does sequence have the /start-end signiature?
- if (limitrx.search(name))
+ int slashPos = name.lastIndexOf('/');
+ if (slashPos > -1 && slashPos < name.length() - 1)
{
- name = limitrx.left();
- endrx.search(limitrx.stringMatched());
- setStart(Integer.parseInt(limitrx.stringMatched().substring(1,
- endrx.matchedFrom() - 1)));
- setEnd(Integer.parseInt(endrx.stringMatched()));
+ 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
+ // 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 (!jalview.util.Comparison.isGap(sequence[j]))
+ if (!Comparison.isGap(sequence[j]))
{
endRes++;
}
@@ -137,7 +227,8 @@ public class Sequence implements SequenceI
endRes += start - 1;
}
- if (end dbr = seq.getDBRefs();
+ if (dbr != null)
+ {
+ for (int i = 0, n = dbr.size(); i < n; i++)
+ {
+ addDBRef(new DBRefEntry(dbr.get(i)));
+ }
+ }
+
+ /*
+ * make copies of any sequence features
+ */
+ for (SequenceFeature sf : seq.getSequenceFeatures())
{
- addDBRef(new DBRefEntry(dbr[i]));
+ addSequenceFeature(new SequenceFeature(sf));
}
}
+
if (seq.getAnnotation() != null)
{
AlignmentAnnotation[] sqann = seq.getAnnotation();
@@ -223,115 +352,108 @@ public class Sequence implements SequenceI
{
// only copy the given annotation
AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
+ ContactMatrixI cm = seq.getContactMatrixFor(sqann[i]);
+ if (cm!=null)
+ {
+ addContactListFor(newann, cm);
+ }
addAlignmentAnnotation(newann);
}
}
}
- if (seq.getPDBId() != null)
+ if (seq.getAllPDBEntries() != null)
{
- Vector ids = seq.getPDBId();
- Enumeration e = ids.elements();
- while (e.hasMoreElements())
+ Vector ids = seq.getAllPDBEntries();
+ for (PDBEntry pdb : ids)
{
- this.addPDBId(new PDBEntry((PDBEntry) e.nextElement()));
+ this.addPDBId(new PDBEntry(pdb));
}
}
}
- /**
- * DOCUMENT ME!
- *
- * @param v
- * DOCUMENT ME!
- */
- public void setSequenceFeatures(SequenceFeature[] features)
+ @Override
+ public void setSequenceFeatures(List features)
{
- sequenceFeatures = features;
+ if (datasetSequence != null)
+ {
+ datasetSequence.setSequenceFeatures(features);
+ return;
+ }
+ sequenceFeatureStore = new SequenceFeatures(features);
}
- public synchronized void addSequenceFeature(SequenceFeature sf)
+ @Override
+ public synchronized boolean addSequenceFeature(SequenceFeature sf)
{
- if (sequenceFeatures == null)
+ if (sf.getType() == null)
{
- sequenceFeatures = new SequenceFeature[0];
+ System.err.println(
+ "SequenceFeature type may not be null: " + sf.toString());
+ return false;
}
- for (int i = 0; i < sequenceFeatures.length; i++)
+ if (datasetSequence != null)
{
- if (sequenceFeatures[i].equals(sf))
- {
- return;
- }
+ return datasetSequence.addSequenceFeature(sf);
}
- SequenceFeature[] temp = new SequenceFeature[sequenceFeatures.length + 1];
- System.arraycopy(sequenceFeatures, 0, temp, 0, sequenceFeatures.length);
- temp[sequenceFeatures.length] = sf;
-
- sequenceFeatures = temp;
+ return sequenceFeatureStore.add(sf);
}
+ @Override
public void deleteFeature(SequenceFeature sf)
{
- if (sequenceFeatures == null)
- {
- return;
- }
-
- int index = 0;
- for (index = 0; index < sequenceFeatures.length; index++)
- {
- if (sequenceFeatures[index].equals(sf))
- {
- break;
- }
- }
-
- if (index == sequenceFeatures.length)
- {
- return;
- }
-
- int sfLength = sequenceFeatures.length;
- if (sfLength < 2)
+ if (datasetSequence != null)
{
- sequenceFeatures = null;
+ datasetSequence.deleteFeature(sf);
}
else
{
- SequenceFeature[] temp = new SequenceFeature[sfLength - 1];
- System.arraycopy(sequenceFeatures, 0, temp, 0, index);
-
- if (index < sfLength)
- {
- System.arraycopy(sequenceFeatures, index + 1, temp, index,
- sequenceFeatures.length - index - 1);
- }
-
- sequenceFeatures = temp;
+ sequenceFeatureStore.delete(sf);
}
}
/**
- * DOCUMENT ME!
+ * {@inheritDoc}
*
- * @return DOCUMENT ME!
+ * @return
*/
- public SequenceFeature[] getSequenceFeatures()
+ @Override
+ public List getSequenceFeatures()
+ {
+ if (datasetSequence != null)
+ {
+ return datasetSequence.getSequenceFeatures();
+ }
+ return sequenceFeatureStore.getAllFeatures();
+ }
+
+ @Override
+ public SequenceFeaturesI getFeatures()
{
- return sequenceFeatures;
+ return datasetSequence != null ? datasetSequence.getFeatures()
+ : sequenceFeatureStore;
}
- public void addPDBId(PDBEntry entry)
+ @Override
+ public boolean addPDBId(PDBEntry entry)
{
if (pdbIds == null)
{
- pdbIds = new Vector();
+ pdbIds = new Vector<>();
+ pdbIds.add(entry);
+ return true;
}
- if (!pdbIds.contains(entry))
+
+ for (PDBEntry pdbe : pdbIds)
{
- pdbIds.addElement(entry);
+ if (pdbe.updateFrom(entry))
+ {
+ return false;
+ }
}
+ pdbIds.addElement(entry);
+ return true;
}
/**
@@ -340,7 +462,8 @@ public class Sequence implements SequenceI
* @param id
* DOCUMENT ME!
*/
- public void setPDBId(Vector id)
+ @Override
+ public void setPDBId(Vector id)
{
pdbIds = id;
}
@@ -350,36 +473,40 @@ public class Sequence implements SequenceI
*
* @return DOCUMENT ME!
*/
- public Vector getPDBId()
+ @Override
+ public Vector getAllPDBEntries()
{
- return pdbIds;
+ return pdbIds == null ? new Vector<>() : pdbIds;
}
/**
- * DOCUMENT ME!
+ * Answers the sequence name, with '/start-end' appended if jvsuffix is true
*
- * @return DOCUMENT ME!
+ * @return
*/
+ @Override
public String getDisplayId(boolean jvsuffix)
{
- StringBuffer result = new StringBuffer(name);
- if (jvsuffix)
+ if (!jvsuffix)
{
- result.append("/" + start + "-" + end);
+ return name;
}
+ StringBuilder result = new StringBuilder(name);
+ result.append("/").append(start).append("-").append(end);
return result.toString();
}
/**
- * DOCUMENT ME!
+ * 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 name
- * DOCUMENT ME!
+ * @param theName
*/
- public void setName(String name)
+ @Override
+ public void setName(String theName)
{
- this.name = name;
+ this.name = theName;
this.parseId();
}
@@ -388,6 +515,7 @@ public class Sequence implements SequenceI
*
* @return DOCUMENT ME!
*/
+ @Override
public String getName()
{
return this.name;
@@ -399,9 +527,11 @@ public class Sequence implements SequenceI
* @param start
* DOCUMENT ME!
*/
+ @Override
public void setStart(int start)
{
this.start = start;
+ sequenceChanged();
}
/**
@@ -409,6 +539,7 @@ public class Sequence implements SequenceI
*
* @return DOCUMENT ME!
*/
+ @Override
public int getStart()
{
return this.start;
@@ -420,6 +551,7 @@ public class Sequence implements SequenceI
* @param end
* DOCUMENT ME!
*/
+ @Override
public void setEnd(int end)
{
this.end = end;
@@ -430,6 +562,7 @@ public class Sequence implements SequenceI
*
* @return DOCUMENT ME!
*/
+ @Override
public int getEnd()
{
return this.end;
@@ -440,6 +573,7 @@ public class Sequence implements SequenceI
*
* @return DOCUMENT ME!
*/
+ @Override
public int getLength()
{
return this.sequence.length;
@@ -451,25 +585,32 @@ public class Sequence implements SequenceI
* @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;
+ return sequence == null ? null
+ : Arrays.copyOf(sequence, sequence.length);
}
/*
@@ -477,10 +618,13 @@ public class Sequence implements SequenceI
*
* @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)
@@ -499,16 +643,7 @@ public class Sequence implements SequenceI
return reply;
}
- /**
- * make a new Sequence object from start to end (including gaps) over this
- * seqeunce
- *
- * @param start
- * int
- * @param end
- * int
- * @return SequenceI
- */
+ @Override
public SequenceI getSubSequence(int start, int end)
{
if (start < 0)
@@ -537,16 +672,15 @@ public class Sequence implements SequenceI
}
/**
- * DOCUMENT ME!
- *
- * @param i
- * DOCUMENT ME!
+ * 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 DOCUMENT ME!
+ * @return
*/
+ @Override
public char getCharAt(int i)
{
- if (i < sequence.length)
+ if (i >= 0 && i < sequence.length)
{
return sequence[i];
}
@@ -557,227 +691,662 @@ public class Sequence implements SequenceI
}
/**
- * DOCUMENT ME!
+ * Sets the sequence description, and also parses out any special formats of
+ * interest
*
* @param desc
- * DOCUMENT ME!
*/
+ @Override
public void setDescription(String desc)
{
this.description = desc;
}
+ @Override
+ public void setGeneLoci(String speciesId, String assemblyId,
+ String chromosomeId, MapList map)
+ {
+ addDBRef(new GeneLocus(speciesId, assemblyId, chromosomeId,
+ new Mapping(map)));
+ }
+
/**
- * DOCUMENT ME!
+ * Returns the gene loci mapping for the sequence (may be null)
*
- * @return DOCUMENT ME!
+ * @return
+ */
+ @Override
+ public GeneLociI getGeneLoci()
+ {
+ List refs = getDBRefs();
+ if (refs != null)
+ {
+ for (final DBRefEntry ref : refs)
+ {
+ if (ref instanceof GeneLociI)
+ {
+ return (GeneLociI) ref;
+ }
+ }
+ }
+ return null;
+ }
+
+ /**
+ * Answers the description
+ *
+ * @return
*/
+ @Override
public String getDescription()
{
return this.description;
}
/**
- * Return the alignment position for a sequence position
- *
- * @param pos
- * lying from start to end
- *
- * @return aligned position of residue pos
+ * {@inheritDoc}
*/
+ @Override
public int findIndex(int pos)
{
- // returns the alignment position for a residue
+ /*
+ * use a valid, hopefully nearby, cursor if available
+ */
+ if (isValidCursor(cursor))
+ {
+ return findIndex(pos, cursor);
+ }
+
int j = start;
int i = 0;
- // Rely on end being at least as long as the length of the sequence.
+ 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
+ */
while ((i < sequence.length) && (j <= end) && (j <= pos))
{
- if (!jalview.util.Comparison.isGap(sequence[i]))
+ if (!Comparison.isGap(sequence[i]))
{
+ if (j == start)
+ {
+ startColumn = i;
+ }
j++;
}
-
i++;
}
- if ((j == end) && (j < pos))
+ if (j == end && j < pos)
{
return end + 1;
}
- else
- {
- return i;
- }
+
+ updateCursor(pos, i, startColumn);
+ return i;
}
/**
- * Returns the sequence position for an alignment position
+ * Updates the cursor to the latest found residue and column position
*
- * @param i
- * column index in alignment (from 1)
- *
- * @return residue number for residue (left of and) nearest ith column
+ * @param residuePos
+ * (start..)
+ * @param column
+ * (1..)
+ * @param startColumn
+ * column position of the first sequence residue
*/
- public int findPosition(int i)
+ protected void updateCursor(int residuePos, int column, int startColumn)
{
- int j = 0;
- int pos = start;
- int seqlen = sequence.length;
- while ((j < i) && (j < seqlen))
- {
- if (!jalview.util.Comparison.isGap(sequence[j]))
- {
- pos++;
- }
+ /*
+ * preserve end residue column provided cursor was valid
+ */
+ int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0;
- j++;
+ if (residuePos == this.end)
+ {
+ endColumn = column;
}
- return pos;
+ cursor = new SequenceCursor(this, residuePos, column, startColumn,
+ endColumn, this.changeCount);
}
/**
- * Returns an int array where indices correspond to each residue in the
- * sequence and the element value gives its position in the alignment
+ * 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.
*
- * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
- * residues in SequenceI object
+ * @param pos
+ * @param curs
+ * @return
*/
- public int[] gapMap()
+ protected int findIndex(final int pos, SequenceCursor curs)
{
- 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 (!isValidCursor(curs))
{
- if (!jalview.util.Comparison.isGap(sequence[j]))
- {
- map[p++] = j;
- }
+ /*
+ * wrong or invalidated cursor, compute de novo
+ */
+ return findIndex(pos);
+ }
- j++;
+ if (curs.residuePosition == pos)
+ {
+ return curs.columnPosition;
}
- return map;
- }
+ /*
+ * move left or right to find pos from hint.position
+ */
+ int col = curs.columnPosition - 1; // convert from base 1 to base 0
+ int newPos = curs.residuePosition;
+ int delta = newPos > pos ? -1 : 1;
- /*
- * (non-Javadoc)
- *
- * @see jalview.datamodel.SequenceI#findPositionMap()
- */
- public int[] findPositionMap()
- {
- int map[] = new int[sequence.length];
- int j = 0;
- int pos = start;
- int seqlen = sequence.length;
- while ((j < seqlen))
+ while (newPos != pos)
{
- map[j] = pos;
- if (!jalview.util.Comparison.isGap(sequence[j]))
+ col += delta; // shift one column left or right
+ if (col < 0)
{
- pos++;
+ break;
+ }
+ if (col == sequence.length)
+ {
+ col--; // return last column if we failed to reach pos
+ break;
}
+ if (!Comparison.isGap(sequence[col]))
+ {
+ newPos += delta;
+ }
+ }
- j++;
+ col++; // convert back to base 1
+
+ /*
+ * only update cursor if we found the target position
+ */
+ if (newPos == pos)
+ {
+ updateCursor(pos, col, curs.firstColumnPosition);
}
- return map;
+
+ return col;
}
- /*
- * (non-Javadoc)
- *
- * @see jalview.datamodel.SequenceI#deleteChars(int, int)
+ /**
+ * {@inheritDoc}
*/
- public void deleteChars(int i, int j)
+ @Override
+ public int findPosition(final int column)
{
- int newstart = start, newend = end;
- if (i >= sequence.length)
+ /*
+ * use a valid, hopefully nearby, cursor if available
+ */
+ if (isValidCursor(cursor))
{
- return;
+ return findPosition(column + 1, cursor);
}
- char[] tmp;
+ // TODO recode this more naturally i.e. count residues only
+ // as they are found, not 'in anticipation'
+
+ /*
+ * traverse the sequence counting gaps; note the column position
+ * of the first residue, to save in the cursor
+ */
+ int firstResidueColumn = 0;
+ int lastPosFound = 0;
+ int lastPosFoundColumn = 0;
+ int seqlen = sequence.length;
- if (j >= sequence.length)
+ if (seqlen > 0 && !Comparison.isGap(sequence[0]))
{
- tmp = new char[i];
- System.arraycopy(sequence, 0, tmp, 0, i);
+ lastPosFound = start;
+ lastPosFoundColumn = 0;
}
- else
+
+ int j = 0;
+ int pos = start;
+
+ while (j < column && j < seqlen)
{
- tmp = new char[sequence.length - j + i];
- System.arraycopy(sequence, 0, tmp, 0, i);
- System.arraycopy(sequence, j, tmp, i, sequence.length - j);
+ if (!Comparison.isGap(sequence[j]))
+ {
+ lastPosFound = pos;
+ lastPosFoundColumn = j;
+ if (pos == this.start)
+ {
+ firstResidueColumn = j;
+ }
+ pos++;
+ }
+ j++;
+ }
+ if (j < seqlen && !Comparison.isGap(sequence[j]))
+ {
+ lastPosFound = pos;
+ lastPosFoundColumn = j;
+ if (pos == this.start)
+ {
+ firstResidueColumn = j;
+ }
+ }
+
+ /*
+ * update the cursor to the last residue position found (if any)
+ * (converting column position to base 1)
+ */
+ if (lastPosFound != 0)
+ {
+ updateCursor(lastPosFound, lastPosFoundColumn + 1,
+ firstResidueColumn + 1);
+ }
+
+ 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
+ */
+ if (curs.columnPosition < 0 || curs.columnPosition > sequence.length)
+ {
+ 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
+ }
+
+ if (curs.columnPosition == col)
+ {
+ cursor = curs; // in case this method becomes public
+ return curs.residuePosition; // easy case :-)
+ }
+
+ if (curs.lastColumnPosition > 0 && curs.lastColumnPosition < col)
+ {
+ /*
+ * sequence lies entirely to the left of col
+ * - return last residue + 1
+ */
+ return end + 1;
+ }
+
+ if (curs.firstColumnPosition > 0 && curs.firstColumnPosition > col)
+ {
+ /*
+ * sequence lies entirely to the right of col
+ * - return first residue
+ */
+ return start;
+ }
+
+ // 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
+ */
+ 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)
+ {
+ 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)
+ {
+ firstResidueColumn = column + 1;
+ }
+ }
+ }
+
+ if (cursor == null || lastFoundPosition != cursor.residuePosition)
+ {
+ updateCursor(lastFoundPosition, lastFoundPositionColumn,
+ firstResidueColumn);
+ }
+
+ /*
+ * hack to give position to the right if on a gap
+ * or beyond the length of the sequence (see JAL-2562)
+ */
+ if (delta > 0 && (gapped || column >= sequence.length))
+ {
+ newPos++;
+ }
+
+ return newPos;
+ }
+
+ /**
+ * {@inheritDoc}
+ */
+ @Override
+ public ContiguousI findPositions(int fromColumn, int toColumn)
+ {
+ if (toColumn < fromColumn || fromColumn < 1)
+ {
+ return null;
+ }
+
+ /*
+ * find the first non-gapped position, if any
+ */
+ int firstPosition = 0;
+ int col = fromColumn - 1;
+ int length = sequence.length;
+ while (col < length && col < toColumn)
+ {
+ if (!Comparison.isGap(sequence[col]))
+ {
+ firstPosition = findPosition(col++);
+ break;
+ }
+ col++;
+ }
+
+ if (firstPosition == 0)
+ {
+ return null;
+ }
+
+ /*
+ * find the last non-gapped position
+ */
+ int lastPosition = firstPosition;
+ while (col < length && col < toColumn)
+ {
+ if (!Comparison.isGap(sequence[col++]))
+ {
+ lastPosition++;
+ }
+ }
+
+ 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;
+ }
+
+ j++;
+ }
+
+ 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 (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++;
+ }
+
+ j++;
+ }
+ return map;
+ }
+
+ @Override
+ public List getInsertions()
+ {
+ ArrayList map = new ArrayList<>();
+ 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)
+ {
+ map.add(new int[] { lastj, j - 1 });
+ lastj = -1;
+ }
+ }
+ 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)
+ {
+ 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;
- for (int s = i; s < j; s++)
+ // 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;
+
+ for (int s = i; s < j && s < sequence.length; s++)
{
- if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
+ if (Comparison.isGap(sequence[s]))
+ {
+ continue;
+ }
+ deleteCount++;
+ if (startDeleteColumn == -1)
{
- if (createNewDs)
+ startDeleteColumn = findPosition(s) - start;
+ }
+ if (createNewDs)
+ {
+ newend--;
+ }
+ else
+ {
+ if (startIndex == s)
{
- newend--;
+ /*
+ * 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
{
- int sindex = findIndex(start) - 1;
- if (sindex == s)
+ if (endIndex < j)
{
- // delete characters including start of sequence
- newstart = findPosition(j);
- break; // don't need to search for any more residue characters.
+ /*
+ * 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
{
- // 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
- }
+ createNewDs = true;
+ newend--;
}
}
}
}
- // deletion occured in the middle of the sequence
+
if (createNewDs && this.datasetSequence != null)
{
- // construct a new sequence
+ /*
+ * 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 ?)
- ds.deleteChars(i, j);
- datasetSequence = ds;
}
start = newstart;
end = newend;
sequence = tmp;
+ sequenceChanged();
}
- /**
- * DOCUMENT ME!
- *
- * @param i
- * DOCUMENT ME!
- * @param c
- * DOCUMENT ME!
- * @param chop
- * DOCUMENT ME!
- */
+ @Override
public void insertCharAt(int i, int length, char c)
{
char[] tmp = new char[sequence.length + length];
@@ -805,101 +1374,137 @@ public class Sequence implements SequenceI
}
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;
}
- public void setDBRef(DBRefEntry[] dbref)
+ @Deprecated
+ @Override
+ public void setDBRefs(DBModList newDBrefs)
{
- dbrefs = dbref;
+ if (dbrefs == null && datasetSequence != null
+ && this != datasetSequence)
+ {
+ datasetSequence.setDBRefs(newDBrefs);
+ return;
+ }
+ dbrefs = newDBrefs;
+ refModCount = 0;
}
- public DBRefEntry[] getDBRef()
+ @Override
+ public DBModList getDBRefs()
{
if (dbrefs == null && datasetSequence != null
&& this != datasetSequence)
{
- return datasetSequence.getDBRef();
+ return datasetSequence.getDBRefs();
}
return dbrefs;
}
+ @Override
public void addDBRef(DBRefEntry entry)
{
- if (dbrefs == null)
+ // TODO JAL-3980 maintain as sorted list
+ if (datasetSequence != null)
{
- dbrefs = new DBRefEntry[0];
+ datasetSequence.addDBRef(entry);
+ return;
}
- int i, iSize = dbrefs.length;
+ if (dbrefs == null)
+ {
+ dbrefs = new DBModList<>();
+ }
+ // TODO JAL-3979 LOOK UP RATHER THAN SWEEP FOR EFFICIENCY
- for (i = 0; i < iSize; i++)
+ for (int ib = 0, nb = dbrefs.size(); ib < nb; ib++)
{
- if (dbrefs[i].equalRef(entry))
+ if (dbrefs.get(ib).updateFrom(entry))
{
- if (entry.getMap() != null)
- {
- if (dbrefs[i].getMap() == null)
- {
- // overwrite with 'superior' entry that contains a mapping.
- dbrefs[i] = entry;
- }
- }
+ /*
+ * found a dbref that either matched, or could be
+ * updated from, the new entry - no need to add it
+ */
return;
}
}
- DBRefEntry[] temp = new DBRefEntry[iSize + 1];
- System.arraycopy(dbrefs, 0, temp, 0, iSize);
- temp[temp.length - 1] = entry;
+ // /// BH OUCH!
+ // /*
+ // * extend the array to make room for one more
+ // */
+ // // TODO use an ArrayList instead
+ // int j = dbrefs.length;
+ // List temp = new DBRefEntry[j + 1];
+ // System.arraycopy(dbrefs, 0, temp, 0, j);
+ // temp[temp.length - 1] = entry;
+ //
+ // dbrefs = temp;
- dbrefs = temp;
+ dbrefs.add(entry);
}
+ @Override
public void setDatasetSequence(SequenceI seq)
{
+ if (seq == this)
+ {
+ throw new IllegalArgumentException(
+ "Implementation Error: self reference passed to SequenceI.setDatasetSequence");
+ }
+ if (seq != null && seq.getDatasetSequence() != null)
+ {
+ throw new IllegalArgumentException(
+ "Implementation error: cascading dataset sequences are not allowed.");
+ }
datasetSequence = seq;
}
+ @Override
public SequenceI getDatasetSequence()
{
return datasetSequence;
}
+ @Override
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 annotation == null ? null
+ : annotation
+ .toArray(new AlignmentAnnotation[annotation.size()]);
+ }
- return ret;
+ @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();
+ this.annotation = new Vector<>();
}
if (!this.annotation.contains(annotation))
{
@@ -908,13 +1513,16 @@ public class Sequence implements SequenceI
annotation.setSequenceRef(this);
}
+ @Override
public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
{
if (this.annotation != null)
{
this.annotation.removeElement(annotation);
if (this.annotation.size() == 0)
+ {
this.annotation = null;
+ }
}
}
@@ -938,38 +1546,52 @@ public class Sequence implements SequenceI
return true;
}
- /*
- * (non-Javadoc)
- *
- * @see jalview.datamodel.SequenceI#deriveSequence()
- */
+ @Override
public SequenceI deriveSequence()
{
- SequenceI seq = new Sequence(this);
- if (datasetSequence != null)
- {
- // duplicate current sequence with same dataset
- seq.setDatasetSequence(datasetSequence);
- }
- else
+ 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
- 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.
+ createDatasetSequence();
}
}
- return seq;
+ return new Sequence(this);
+ }
+
+ private boolean _isNa;
+
+ private int _seqhash = 0;
+
+ private List primaryRefs;
+
+ /**
+ * Answers false if the sequence is more than 85% nucleotide (ACGTU), else
+ * true
+ */
+ @Override
+ public boolean isProtein()
+ {
+ if (datasetSequence != null)
+ {
+ return datasetSequence.isProtein();
+ }
+ if (_seqhash != sequence.hashCode())
+ {
+ _seqhash = sequence.hashCode();
+ _isNa = Comparison.isNucleotide(this);
+ }
+ return !_isNa;
}
/*
@@ -977,22 +1599,54 @@ public class Sequence implements SequenceI
*
* @see jalview.datamodel.SequenceI#createDatasetSequence()
*/
+ @Override
public SequenceI createDatasetSequence()
{
if (datasetSequence == null)
{
- datasetSequence = new Sequence(getName(), AlignSeq.extractGaps(
- jalview.util.Comparison.GapChars, getSequenceAsString()),
+ Sequence dsseq = 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);
- datasetSequence.setPDBId(getPDBId());
- setPDBId(null);
+
+ 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);
+
+ if (_cmholder != null)
+ { // transfer contact matrices
+ ContactMatrixI cm = _cmholder.getContactMatrixFor(aa);
+ if (cm != null)
+ {
+ datasetSequence.addContactListFor(_aa, cm);
+ datasetSequence.addContactListFor(aa, cm);
+ }
+ }
+ }
+ }
+ // all matrices should have been transferred. so we clear the local holder
+ _cmholder=null;
}
return datasetSequence;
}
@@ -1004,6 +1658,7 @@ public class Sequence implements SequenceI
* jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
* annotations)
*/
+ @Override
public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
{
if (annotation != null)
@@ -1015,16 +1670,14 @@ public class Sequence implements SequenceI
for (int i = 0; i < annotations.length; i++)
{
if (annotations[i] != null)
+ {
addAlignmentAnnotation(annotations[i]);
+ }
}
}
}
- /*
- * (non-Javadoc)
- *
- * @see jalview.datamodel.SequenceI#getAnnotation(java.lang.String)
- */
+ @Override
public AlignmentAnnotation[] getAnnotation(String label)
{
if (annotation == null || annotation.size() == 0)
@@ -1032,11 +1685,11 @@ public class Sequence implements SequenceI
return null;
}
- Vector subset = new Vector();
- Enumeration e = annotation.elements();
+ Vector subset = new Vector<>();
+ Enumeration e = annotation.elements();
while (e.hasMoreElements())
{
- AlignmentAnnotation ann = (AlignmentAnnotation) e.nextElement();
+ AlignmentAnnotation ann = e.nextElement();
if (ann.label != null && ann.label.equals(label))
{
subset.addElement(ann);
@@ -1051,12 +1704,13 @@ public class Sequence implements SequenceI
e = subset.elements();
while (e.hasMoreElements())
{
- anns[i++] = (AlignmentAnnotation) e.nextElement();
+ anns[i++] = e.nextElement();
}
subset.removeAllElements();
return anns;
}
+ @Override
public boolean updatePDBIds()
{
if (datasetSequence != null)
@@ -1064,59 +1718,30 @@ public class Sequence implements SequenceI
// TODO: could merge DBRefs
return datasetSequence.updatePDBIds();
}
- if (dbrefs == null || dbrefs.length == 0)
+ if (dbrefs == null || dbrefs.size() == 0)
{
return false;
}
- Vector newpdb = new Vector();
- for (int i = 0; i < dbrefs.length; i++)
+ boolean added = false;
+ for (int ib = 0, nb = dbrefs.size(); ib < nb; ib++)
{
- if (DBRefSource.PDB.equals(dbrefs[i].getSource()))
+ DBRefEntry dbr = dbrefs.get(ib);
+ if (DBRefSource.PDB.equals(dbr.getSource()))
{
- PDBEntry pdbe = new PDBEntry();
- pdbe.setId(dbrefs[i].getAccessionId());
- if (pdbIds == null || pdbIds.size() == 0)
- {
- newpdb.addElement(pdbe);
- }
- else
- {
- Enumeration en = pdbIds.elements();
- boolean matched = false;
- while (!matched && en.hasMoreElements())
- {
- PDBEntry anentry = (PDBEntry) en.nextElement();
- if (anentry.getId().equals(pdbe.getId()))
- {
- matched = true;
- }
- }
- if (!matched)
- {
- newpdb.addElement(pdbe);
- }
- }
- }
- }
- if (newpdb.size() > 0)
- {
- Enumeration en = newpdb.elements();
- while (en.hasMoreElements())
- {
- addPDBId((PDBEntry) en.nextElement());
+ /*
+ * '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 true;
}
- return false;
+ return added;
}
- /*
- * (non-Javadoc)
- *
- * @see
- * jalview.datamodel.SequenceI#transferAnnotation(jalview.datamodel.SequenceI,
- * jalview.datamodel.Mapping)
- */
+ @Override
public void transferAnnotation(SequenceI entry, Mapping mp)
{
if (datasetSequence != null)
@@ -1133,13 +1758,13 @@ public class Sequence implements SequenceI
if (entry.getSequenceFeatures() != null)
{
- SequenceFeature[] sfs = entry.getSequenceFeatures();
- for (int si = 0; si < sfs.length; si++)
+ List sfs = entry.getSequenceFeatures();
+ for (SequenceFeature feature : sfs)
{
- SequenceFeature sf[] = (mp != null) ? mp.locateFeature(sfs[si])
+ SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
: new SequenceFeature[]
- { new SequenceFeature(sfs[si]) };
- if (sf != null && sf.length > 0)
+ { new SequenceFeature(feature) };
+ if (sf != null)
{
for (int sfi = 0; sfi < sf.length; sfi++)
{
@@ -1150,22 +1775,22 @@ public class Sequence implements SequenceI
}
// transfer PDB entries
- if (entry.getPDBId() != null)
+ if (entry.getAllPDBEntries() != null)
{
- Enumeration e = entry.getPDBId().elements();
+ Enumeration e = entry.getAllPDBEntries().elements();
while (e.hasMoreElements())
{
- PDBEntry pdb = (PDBEntry) e.nextElement();
+ PDBEntry pdb = e.nextElement();
addPDBId(pdb);
}
}
// transfer database references
- DBRefEntry[] entryRefs = entry.getDBRef();
+ List entryRefs = entry.getDBRefs();
if (entryRefs != null)
{
- for (int r = 0; r < entryRefs.length; r++)
+ for (int r = 0, n = entryRefs.size(); r < n; r++)
{
- DBRefEntry newref = new DBRefEntry(entryRefs[r]);
+ DBRefEntry newref = new DBRefEntry(entryRefs.get(r));
if (newref.getMap() != null && mp != null)
{
// remap ref using our local mapping
@@ -1180,17 +1805,405 @@ public class Sequence implements SequenceI
}
}
+ @Override
+ public void setRNA(RNA r)
+ {
+ rna = r;
+ }
+
+ @Override
+ public RNA getRNA()
+ {
+ return rna;
+ }
+
+ @Override
+ public List getAlignmentAnnotations(String calcId,
+ String label)
+ {
+ return getAlignmentAnnotations(calcId, label, null, true);
+ }
+
+ @Override
+ public List getAlignmentAnnotations(String calcId,
+ String label, String description)
+ {
+ return getAlignmentAnnotations(calcId, label, description, false);
+ }
+
+ private List getAlignmentAnnotations(String calcId,
+ String label, String description, boolean ignoreDescription)
+ {
+ List 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))
+ && ((ignoreDescription && description == null)
+ || (ann.description != null
+ && ann.description.equals(description))))
+
+ {
+ 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 entries = getAllPDBEntries();
+ for (PDBEntry entry : entries)
+ {
+ if (entry.getId().equalsIgnoreCase(pdbIdStr))
+ {
+ return entry;
+ }
+ }
+ return null;
+ }
+
+ private List tmpList;
+
+ @Override
+ public List getPrimaryDBRefs()
+ {
+ if (datasetSequence != null)
+ {
+ return datasetSequence.getPrimaryDBRefs();
+ }
+ if (dbrefs == null || dbrefs.size() == 0)
+ {
+ return Collections.emptyList();
+ }
+ synchronized (dbrefs)
+ {
+ if (refModCount == dbrefs.getModCount() && primaryRefs != null)
+ {
+ return primaryRefs; // no changes
+ }
+ refModCount = dbrefs.getModCount();
+ List primaries = (primaryRefs == null
+ ? (primaryRefs = new ArrayList<>())
+ : primaryRefs);
+ primaries.clear();
+ if (tmpList == null)
+ {
+ tmpList = new ArrayList<>();
+ tmpList.add(null); // for replacement
+ }
+ for (int i = 0, n = dbrefs.size(); i < n; i++)
+ {
+ DBRefEntry ref = dbrefs.get(i);
+ 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 (DBRefSource.PDB_CANONICAL_NAME
+ .equals(ref.getCanonicalSourceName()))
+ {
+ // 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)
+ {
+ continue;
+ }
+ }
+ else
+ {
+ // check standard protein or dna sources
+ tmpList.set(0, ref);
+ List res = DBRefUtils.selectDbRefs(!isProtein(),
+ tmpList);
+ if (res == null || res.get(0) != tmpList.get(0))
+ {
+ continue;
+ }
+ }
+ primaries.add(ref);
+ }
+
+ // version must be not null, as otherwise it will not be a candidate,
+ // above
+ DBRefUtils.ensurePrimaries(this, primaries);
+ return primaries;
+ }
+ }
+
+ /**
+ * {@inheritDoc}
+ */
+ @Override
+ public List 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 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 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();
+ }
+ }
+ }
+ }
+
+ return result;
+ }
+
/**
- * @return The index (zero-based) on this sequence in the MSA.
- * It returns {@code -1} if this information is not available.
+ * Invalidates any stale cursors (forcing recalculation) by incrementing the
+ * token that has to match the one presented by the cursor
*/
- public int getIndex() { return index; }
-
+ @Override
+ public void sequenceChanged()
+ {
+ changeCount++;
+ }
+
/**
- * Defines the position of this sequence in the MSA.
- * Use the value {@code -1} if this information is undefined.
- *
- * @param The position for this sequence. This value is zero-based (zero for this first sequence)
+ * {@inheritDoc}
*/
- public void setIndex(int value) { index = value; }
+ @Override
+ public int replace(char c1, char c2)
+ {
+ if (c1 == c2)
+ {
+ return 0;
+ }
+ int count = 0;
+ synchronized (sequence)
+ {
+ for (int c = 0; c < sequence.length; c++)
+ {
+ if (sequence[c] == c1)
+ {
+ sequence[c] = c2;
+ count++;
+ }
+ }
+ }
+ if (count > 0)
+ {
+ sequenceChanged();
+ }
+
+ return count;
+ }
+
+ @Override
+ public String getSequenceStringFromIterator(Iterator 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]));
+ }
+ }
+
+ return newSequence.toString();
+ }
+
+ @Override
+ public int firstResidueOutsideIterator(Iterator regions)
+ {
+ int start = 0;
+
+ if (!regions.hasNext())
+ {
+ return findIndex(getStart()) - 1;
+ }
+
+ // Simply walk along the sequence whilst watching for region
+ // boundaries
+ int hideStart = getLength();
+ int hideEnd = -1;
+ boolean foundStart = false;
+
+ // 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;
+
+ // 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;
+ }
+
+ ////
+ //// Contact Matrix Holder Boilerplate
+ ////
+ ContactMapHolderI _cmholder = null;
+
+ private ContactMapHolderI getContactMapHolder()
+ {
+ if (datasetSequence!=null) {
+ return ((Sequence)datasetSequence).getContactMapHolder();
+ }
+ if (_cmholder==null)
+ {
+ _cmholder=new ContactMapHolder();
+ }
+ return _cmholder;
+ }
+ @Override
+ public Collection getContactMaps()
+ {
+ return getContactMapHolder().getContactMaps();
+ }
+
+ @Override
+ public ContactMatrixI getContactMatrixFor(AlignmentAnnotation ann)
+ {
+ return getContactMapHolder().getContactMatrixFor(ann);
+ }
+
+ @Override
+ public ContactListI getContactListFor(AlignmentAnnotation _aa, int column)
+ {
+ return getContactMapHolder().getContactListFor(_aa, column);
+ }
+
+ @Override
+ public AlignmentAnnotation addContactList(ContactMatrixI cm)
+ {
+ AlignmentAnnotation aa;
+
+ if (datasetSequence != null)
+ {
+ aa = datasetSequence.addContactList(cm);
+ // clone the annotation for the local sequence
+ aa = new AlignmentAnnotation(aa);
+ aa.restrict(start, end);
+ aa.adjustForAlignment();
+ getContactMapHolder().addContactListFor(aa,cm);
+ addAlignmentAnnotation(aa);
+ return aa;
+ }
+
+ // construct new annotation for matrix on dataset sequence
+ aa = getContactMapHolder().addContactList(cm);
+
+ Annotation _aa[] = new Annotation[getLength()];
+
+ for (int i = 0; i < _aa.length; _aa[i++] = new Annotation(0.0f))
+ {
+ ;
+ }
+ aa.annotations = _aa;
+ aa.setSequenceRef(this);
+ if (cm instanceof MappableContactMatrix
+ && !((MappableContactMatrix) cm).hasReferenceSeq())
+ {
+ ((MappableContactMatrix) cm).setRefSeq(this);
+ }
+ aa.createSequenceMapping(this, getStart(), false);
+ addAlignmentAnnotation(aa);
+ return aa;
+ }
+
+ @Override
+ public void addContactListFor(AlignmentAnnotation annotation,
+ ContactMatrixI cm)
+ {
+ getContactMapHolder().addContactListFor(annotation, cm);
+ }
}