import jalview.analysis.AlignSeq;
import jalview.api.DBRefEntryI;
+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.Collections;
import java.util.Enumeration;
import java.util.List;
import java.util.Vector;
String vamsasId;
- DBRefEntryI sourceDBRef;
-
DBRefEntry[] dbrefs;
RNA rna;
}
/**
- * Creates a new Sequence object with new features, DBRefEntries,
- * AlignmentAnnotations, and PDBIds but inherits any existing dataset sequence
- * reference.
+ * Creates a new Sequence object with new AlignmentAnnotations but inherits
+ * any existing dataset sequence reference. If non exists, everything is
+ * copied.
*
* @param seq
- * DOCUMENT ME!
+ * if seq is a dataset sequence, behaves like a plain old copy
+ * constructor
*/
public Sequence(SequenceI seq)
{
}
+ /**
+ * 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)
{
- initSeqAndName(seq.getName(), seq.getSequence(), seq.getStart(),
- seq.getEnd());
+ {
+ char[] oseq = seq.getSequence();
+ initSeqAndName(seq.getName(), Arrays.copyOf(oseq, oseq.length),
+ seq.getStart(), seq.getEnd());
+ }
description = seq.getDescription();
- sourceDBRef = seq.getSourceDBRef();
- if (seq.getSequenceFeatures() != null)
+ if (seq != datasetSequence)
{
- SequenceFeature[] sf = seq.getSequenceFeatures();
- for (int i = 0; i < sf.length; i++)
- {
- addSequenceFeature(new SequenceFeature(sf[i]));
- }
+ setDatasetSequence(seq.getDatasetSequence());
}
- setDatasetSequence(seq.getDatasetSequence());
if (datasetSequence == null && seq.getDBRefs() != null)
{
- // only copy DBRefs if we really are a dataset sequence
+ // only copy DBRefs and seqfeatures if we really are a dataset sequence
DBRefEntry[] dbr = seq.getDBRefs();
for (int i = 0; i < dbr.length; i++)
{
addDBRef(new DBRefEntry(dbr[i]));
}
+ if (seq.getSequenceFeatures() != null)
+ {
+ SequenceFeature[] sf = seq.getSequenceFeatures();
+ for (int i = 0; i < sf.length; i++)
+ {
+ addSequenceFeature(new SequenceFeature(sf[i]));
+ }
+ }
}
if (seq.getAnnotation() != null)
{
}
if (seq.getAllPDBEntries() != null)
{
- Vector ids = seq.getAllPDBEntries();
- Enumeration e = ids.elements();
- while (e.hasMoreElements())
+ Vector<PDBEntry> 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!
- */
@Override
public void setSequenceFeatures(SequenceFeature[] features)
{
- sequenceFeatures = features;
+ if (datasetSequence == null)
+ {
+ sequenceFeatures = features;
+ }
+ else
+ {
+ if (datasetSequence.getSequenceFeatures() != features
+ && datasetSequence.getSequenceFeatures() != null
+ && datasetSequence.getSequenceFeatures().length > 0)
+ {
+ new Exception(
+ "Warning: JAL-2046 side effect ? Possible implementation error: overwriting dataset sequence features by setting sequence features on alignment")
+ .printStackTrace();
+ }
+ datasetSequence.setSequenceFeatures(features);
+ }
}
@Override
public synchronized void addSequenceFeature(SequenceFeature sf)
{
+ if (sequenceFeatures == null && datasetSequence != null)
+ {
+ datasetSequence.addSequenceFeature(sf);
+ return;
+ }
if (sequenceFeatures == null)
{
sequenceFeatures = new SequenceFeature[0];
{
if (sequenceFeatures == null)
{
+ if (datasetSequence != null)
+ {
+ datasetSequence.deleteFeature(sf);
+ }
return;
}
}
@Override
- public void addPDBId(PDBEntry entry)
+ public boolean addPDBId(PDBEntry entry)
{
if (pdbIds == null)
{
pdbIds = new Vector<PDBEntry>();
+ pdbIds.add(entry);
+ return true;
}
- if (pdbIds.contains(entry))
- {
- updatePDBEntry(pdbIds.get(pdbIds.indexOf(entry)), entry);
- }
- else
- {
- pdbIds.addElement(entry);
- }
- }
- private static void updatePDBEntry(PDBEntry oldEntry, PDBEntry newEntry)
- {
- if (newEntry.getFile() != null)
+ for (PDBEntry pdbe : pdbIds)
{
- oldEntry.setFile(newEntry.getFile());
+ if (pdbe.updateFrom(entry))
+ {
+ return false;
+ }
}
+ pdbIds.addElement(entry);
+ return true;
}
/**
@Override
public Vector<PDBEntry> getAllPDBEntries()
{
- return pdbIds;
+ return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
}
/**
}
/**
- * 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];
}
}
@Override
- public void setDBRef(DBRefEntry[] dbref)
+ public void setDBRefs(DBRefEntry[] dbref)
{
+ if (dbrefs == null && datasetSequence != null
+ && this != datasetSequence)
+ {
+ datasetSequence.setDBRefs(dbref);
+ return;
+ }
dbrefs = dbref;
+ if (dbrefs != null)
+ {
+ DBRefUtils.ensurePrimaries(this);
+ }
}
@Override
@Override
public void addDBRef(DBRefEntry entry)
{
+ if (datasetSequence != null)
+ {
+ datasetSequence.addDBRef(entry);
+ return;
+ }
+
if (dbrefs == null)
{
dbrefs = new DBRefEntry[0];
}
- int i, iSize = dbrefs.length;
-
- for (i = 0; i < iSize; i++)
+ for (DBRefEntryI dbr : dbrefs)
{
- if (dbrefs[i].equalRef(entry))
+ if (dbr.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);
+ /*
+ * 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;
+
+ DBRefUtils.ensurePrimaries(this);
}
@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 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 long _seqhash = 0;
+
+ /**
+ * 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;
+ };
+
/*
* (non-Javadoc)
*
{
if (datasetSequence == null)
{
- datasetSequence = new Sequence(getName(), AlignSeq.extractGaps(
+ 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(getDBRefs());
- setDBRef(null);
- datasetSequence.setPDBId(getAllPDBEntries());
- setPDBId(null);
+
+ datasetSequence = dsseq;
+
+ dsseq.setDescription(description);
+ // move features and database references onto dataset sequence
+ dsseq.sequenceFeatures = sequenceFeatures;
+ sequenceFeatures = 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);
{
return false;
}
- Vector newpdb = new Vector();
- for (int i = 0; i < dbrefs.length; i++)
+ boolean added = false;
+ for (DBRefEntry dbr : dbrefs)
{
- if (DBRefSource.PDB.equals(dbrefs[i].getSource()))
+ 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;
}
@Override
@Override
public PDBEntry getPDBEntry(String pdbIdStr)
{
- if (getDatasetSequence() == null
- || getDatasetSequence().getAllPDBEntries() == null)
+ if (getDatasetSequence() != null)
+ {
+ return getDatasetSequence().getPDBEntry(pdbIdStr);
+ }
+ if (pdbIds == null)
{
return null;
}
- List<PDBEntry> entries = getDatasetSequence().getAllPDBEntries();
+ List<PDBEntry> entries = getAllPDBEntries();
for (PDBEntry entry : entries)
{
if (entry.getId().equalsIgnoreCase(pdbIdStr))
}
@Override
- public void setSourceDBRef(DBRefEntryI dbRef)
- {
- this.sourceDBRef = dbRef;
- }
-
- @Override
- public DBRefEntryI getSourceDBRef()
+ public List<DBRefEntry> getPrimaryDBRefs()
{
- return this.sourceDBRef;
+ if (datasetSequence != null)
+ {
+ return datasetSequence.getPrimaryDBRefs();
+ }
+ if (dbrefs == null || dbrefs.length == 0)
+ {
+ return Collections.emptyList();
+ }
+ synchronized (dbrefs)
+ {
+ List<DBRefEntry> primaries = new ArrayList<DBRefEntry>();
+ 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;
+ }
}
}
git://source.jalview.org / jalview.git / blobdiff