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.List;
+import java.util.ListIterator;
import java.util.Vector;
+import com.stevesoft.pat.Regex;
+
import fr.orsay.lri.varna.models.rna.RNA;
/**
*/
public class Sequence extends ASequence implements SequenceI
{
+ private static final Regex limitrx = new Regex(
+ "[/][0-9]{1,}[-][0-9]{1,}$");
+
+ private static final Regex endrx = new Regex("[0-9]{1,}$");
+
SequenceI datasetSequence;
String name;
*/
int index = -1;
- /** array of sequence features - may not be null for a valid sequence object */
- public SequenceFeature[] sequenceFeatures;
+ private SequenceFeatures sequenceFeatureStore;
/**
* Creates a new Sequence object.
*/
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);
}
checkValidRange();
}
- com.stevesoft.pat.Regex limitrx = new com.stevesoft.pat.Regex(
- "[/][0-9]{1,}[-][0-9]{1,}$");
-
- com.stevesoft.pat.Regex endrx = new com.stevesoft.pat.Regex("[0-9]{1,}$");
-
void parseId()
{
if (name == null)
}
/**
+ * default constructor
+ */
+ private Sequence()
+ {
+ sequenceFeatureStore = new SequenceFeatures();
+ }
+
+ /**
* Creates a new Sequence object.
*
* @param name
*/
public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation)
{
+ this();
initSeqFrom(seq, alAnnotation);
-
}
/**
protected void initSeqFrom(SequenceI seq,
AlignmentAnnotation[] alAnnotation)
{
- {
- char[] oseq = seq.getSequence();
- initSeqAndName(seq.getName(), Arrays.copyOf(oseq, oseq.length),
- seq.getStart(),
- seq.getEnd());
- }
+ char[] oseq = seq.getSequence();
+ initSeqAndName(seq.getName(), Arrays.copyOf(oseq, oseq.length),
+ seq.getStart(), seq.getEnd());
+
description = seq.getDescription();
if (seq != datasetSequence)
{
setDatasetSequence(seq.getDatasetSequence());
}
- if (datasetSequence == null && seq.getDBRefs() != null)
+
+ /*
+ * only copy DBRefs and seqfeatures if we really are a dataset sequence
+ */
+ if (datasetSequence == null)
{
- // 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)
+ if (seq.getDBRefs() != null)
{
- SequenceFeature[] sf = seq.getSequenceFeatures();
- for (int i = 0; i < sf.length; i++)
+ DBRefEntry[] dbr = seq.getDBRefs();
+ for (int i = 0; i < dbr.length; i++)
{
- addSequenceFeature(new SequenceFeature(sf[i]));
+ addDBRef(new DBRefEntry(dbr[i]));
}
}
+
+ /*
+ * make copies of any sequence features
+ */
+ for (SequenceFeature sf : seq.getSequenceFeatures())
+ {
+ addSequenceFeature(new SequenceFeature(sf));
+ }
}
+
if (seq.getAnnotation() != null)
{
AlignmentAnnotation[] sqann = seq.getAnnotation();
}
}
-
@Override
- public void setSequenceFeatures(SequenceFeature[] features)
+ public void setSequenceFeatures(List<SequenceFeature> features)
{
- if (datasetSequence == null)
- {
- sequenceFeatures = features;
- }
- else
+ if (datasetSequence != null)
{
- 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);
+ return;
}
+ sequenceFeatureStore = new SequenceFeatures(features);
}
@Override
- public synchronized void addSequenceFeature(SequenceFeature sf)
+ public synchronized boolean addSequenceFeature(SequenceFeature sf)
{
- if (sequenceFeatures==null && datasetSequence != null)
+ if (sf.getType() == null)
{
- datasetSequence.addSequenceFeature(sf);
- return;
- }
- if (sequenceFeatures == 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)
- {
- if (datasetSequence!=null) {
- datasetSequence.deleteFeature(sf);
- }
- 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);
}
}
/**
- * Returns the sequence features (if any), looking first on the sequence, then
- * on its dataset sequence, and so on until a non-null value is found (or
- * none). This supports retrieval of sequence features stored on the sequence
- * (as in the applet) or on the dataset sequence (as in the Desktop version).
+ * {@inheritDoc}
*
* @return
*/
@Override
- public SequenceFeature[] getSequenceFeatures()
+ public List<SequenceFeature> getSequenceFeatures()
{
- SequenceFeature[] features = sequenceFeatures;
-
- SequenceI seq = this;
- int count = 0; // failsafe against loop in sequence.datasetsequence...
- while (features == null && seq.getDatasetSequence() != null
- && count++ < 10)
+ if (datasetSequence != null)
{
- seq = seq.getDatasetSequence();
- features = ((Sequence) seq).sequenceFeatures;
+ return datasetSequence.getSequenceFeatures();
}
- return features;
+ return sequenceFeatureStore.getAllFeatures();
+ }
+
+ @Override
+ public SequenceFeaturesI getFeatures()
+ {
+ return datasetSequence != null ? datasetSequence.getFeatures()
+ : sequenceFeatureStore;
}
@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;
}
/**
// Rely on end being at least as long as the length of the sequence.
while ((i < sequence.length) && (j <= end) && (j <= pos))
{
- if (!jalview.util.Comparison.isGap(sequence[i]))
+ if (!Comparison.isGap(sequence[i]))
{
j++;
}
int seqlen = sequence.length;
while ((j < i) && (j < seqlen))
{
- if (!jalview.util.Comparison.isGap(sequence[j]))
+ if (!Comparison.isGap(sequence[j]))
{
pos++;
}
}
@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(int i, int j)
{
int newstart = start, newend = end;
@Override
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)
{
- // TODO add to dataset sequence instead if there is one?
+ if (datasetSequence != null)
+ {
+ datasetSequence.addDBRef(entry);
+ return;
+ }
+
if (dbrefs == null)
{
dbrefs = new DBRefEntry[0];
temp[temp.length - 1] = entry;
dbrefs = temp;
+
+ DBRefUtils.ensurePrimaries(this);
}
@Override
public void setDatasetSequence(SequenceI seq)
{
- // TODO check for circular reference before setting?
+ 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()
{
- Sequence seq=null;
+ Sequence seq = null;
if (datasetSequence == null)
{
if (isValidDatasetSequence())
else
{
// Create a new, valid dataset sequence
- createDatasetSequence();
+ createDatasetSequence();
}
}
return new Sequence(this);
private long _seqhash = 0;
+ /**
+ * Answers false if the sequence is more than 85% nucleotide (ACGTU), else
+ * true
+ */
@Override
public boolean isProtein()
{
if (_seqhash != sequence.hashCode())
{
_seqhash = sequence.hashCode();
- _isNa=jalview.util.Comparison.isNucleotide(new SequenceI[] { this });
+ _isNa = Comparison.isNucleotide(this);
}
return !_isNa;
};
dsseq.setDescription(description);
// move features and database references onto dataset sequence
- dsseq.sequenceFeatures = sequenceFeatures;
- sequenceFeatures=null;
+ dsseq.sequenceFeatureStore = sequenceFeatureStore;
+ sequenceFeatureStore = null;
dsseq.dbrefs = dbrefs;
- dbrefs=null;
+ dbrefs = null;
// TODO: search and replace any references to this sequence with
// references to the dataset sequence in Mappings on dbref
dsseq.pdbIds = pdbIds;
return null;
}
- Vector subset = new Vector();
- Enumeration e = annotation.elements();
+ Vector<AlignmentAnnotation> subset = new Vector<AlignmentAnnotation>();
+ Enumeration<AlignmentAnnotation> 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);
e = subset.elements();
while (e.hasMoreElements())
{
- anns[i++] = (AlignmentAnnotation) e.nextElement();
+ anns[i++] = e.nextElement();
}
subset.removeAllElements();
return anns;
{
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
if (entry.getSequenceFeatures() != null)
{
- SequenceFeature[] sfs = entry.getSequenceFeatures();
- for (int si = 0; si < sfs.length; si++)
+ List<SequenceFeature> sfs = entry.getSequenceFeatures();
+ for (SequenceFeature feature : sfs)
{
- SequenceFeature sf[] = (mp != null) ? mp.locateFeature(sfs[si])
- : new SequenceFeature[] { new SequenceFeature(sfs[si]) };
- if (sf != null && sf.length > 0)
+ SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
+ : new SequenceFeature[] { new SequenceFeature(feature) };
+ if (sf != null)
{
for (int sfi = 0; sfi < sf.length; sfi++)
{
// transfer PDB entries
if (entry.getAllPDBEntries() != null)
{
- Enumeration e = entry.getAllPDBEntries().elements();
+ Enumeration<PDBEntry> e = entry.getAllPDBEntries().elements();
while (e.hasMoreElements())
{
- PDBEntry pdb = (PDBEntry) e.nextElement();
+ PDBEntry pdb = e.nextElement();
addPDBId(pdb);
}
}
return null;
}
-
@Override
public List<DBRefEntry> getPrimaryDBRefs()
{
- if (datasetSequence!=null)
+ if (datasetSequence != null)
{
return datasetSequence.getPrimaryDBRefs();
}
- if (dbrefs==null || dbrefs.length==0)
+ if (dbrefs == null || dbrefs.length == 0)
{
- return Arrays.asList(new DBRefEntry[0]);
+ return Collections.emptyList();
}
synchronized (dbrefs)
{
List<DBRefEntry> primaries = new ArrayList<DBRefEntry>();
- DBRefEntry tmp[] = new DBRefEntry[1], res[] = null;
+ DBRefEntry[] tmp = new DBRefEntry[1];
for (DBRefEntry ref : dbrefs)
{
- if (!ref.isPrimary())
+ if (!ref.isPrimaryCandidate())
{
continue;
}
}
// check standard protein or dna sources
tmp[0] = ref;
- res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
+ DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
if (res != null && res[0] == tmp[0])
{
primaries.add(ref);
}
}
+ /**
+ * {@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 = findPosition(toColumn - 1);
+
+ List<SequenceFeature> result = new ArrayList<>();
+ if (datasetSequence != null)
+ {
+ result = datasetSequence.getFeatures().findFeatures(startPos, endPos,
+ types);
+ }
+ else
+ {
+ result = sequenceFeatureStore.findFeatures(startPos, endPos, types);
+ }
+
+ /*
+ * if the start or end column is gapped, startPos or endPos may be to the
+ * left or right, and we may have included adjacent or enclosing features;
+ * remove any that are not enclosing, non-contact features
+ */
+ if (endPos > this.end || Comparison.isGap(sequence[fromColumn - 1])
+ || Comparison.isGap(sequence[toColumn - 1]))
+ {
+ ListIterator<SequenceFeature> it = result.listIterator();
+ while (it.hasNext())
+ {
+ SequenceFeature sf = it.next();
+ int featureStartColumn = findIndex(sf.getBegin());
+ int featureEndColumn = findIndex(sf.getEnd());
+ boolean noOverlap = featureStartColumn > toColumn
+ || featureEndColumn < fromColumn;
+
+ /*
+ * reject an 'enclosing' feature if it is actually a contact feature
+ */
+ if (sf.isContactFeature() && featureStartColumn < fromColumn
+ && featureEndColumn > toColumn)
+ {
+ noOverlap = true;
+ }
+ if (noOverlap)
+ {
+ it.remove();
+ }
+ }
+ }
+
+ return result;
+ }
}
git://source.jalview.org / jalview.git / blobdiff