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 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;
*/
Vector<AlignmentAnnotation> annotation;
- /**
- * The index of the sequence in a MSA
- */
- int index = -1;
-
- private SequenceFeatures sequenceFeatureStore;
+ private SequenceFeaturesI sequenceFeatureStore;
/*
* A cursor holding the approximate current view position to the sequence,
*/
private int changeCount;
- private GeneLoci geneLoci;
-
/**
* Creates a new Sequence object.
*
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)
"POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
name = "";
}
- // Does sequence have the /start-end signature?
- 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 :
int endRes = 0;
for (int j = 0; j < sequence.length; j++)
{
- if (!jalview.util.Comparison.isGap(sequence[j]))
+ if (!Comparison.isGap(sequence[j]))
{
endRes++;
}
{
if (pdbIds == null)
{
- pdbIds = new Vector<PDBEntry>();
+ pdbIds = new Vector<>();
pdbIds.add(entry);
return true;
}
@Override
public Vector<PDBEntry> getAllPDBEntries()
{
- return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
+ return pdbIds == null ? new Vector<>() : pdbIds;
}
/**
}
/**
- * 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
*/
@Override
- public void setName(String name)
+ public void setName(String theName)
{
- this.name = name;
+ this.name = theName;
this.parseId();
}
public void setDescription(String desc)
{
this.description = desc;
- parseDescription();
- }
-
- /**
- * Parses and saves fields of an Ensembl-style description e.g.
- * chromosome:GRCh38:17:45051610:45109016:1
- */
- protected void parseDescription()
- {
- if (description == null)
- {
- return;
- }
- String[] tokens = description.split(":");
- if (tokens.length == 6 && "chromosome".equals(tokens[0])) {
- String ref = tokens[1];
- String chrom = tokens[2];
- try {
- int chStart = Integer.parseInt(tokens[3]);
- int chEnd = Integer.parseInt(tokens[4]);
- boolean forwardStrand = "1".equals(tokens[5]);
- String species = ""; // dunno yet!
- int[] from = new int[] { start, end };
- int[] to = new int[] { chStart, chEnd };
- MapList map = new MapList(from, to, 1, 1);
- GeneLoci gl = new GeneLoci(species, ref, chrom, map, forwardStrand);
- setGeneLoci(gl);
- } catch (NumberFormatException e)
- {
- System.err.println("Bad integers in description " + description);
- }
- }
}
- public void setGeneLoci(GeneLoci gl)
+ @Override
+ public void setGeneLoci(String speciesId, String assemblyId,
+ String chromosomeId, MapList map)
{
- geneLoci = gl;
+ addDBRef(new DBRefEntry(speciesId, assemblyId, DBRefEntry.CHROMOSOME
+ + ":" + chromosomeId, new Mapping(map)));
}
/**
*
* @return
*/
- public GeneLoci getGeneLoci()
+ @Override
+ public GeneLociI getGeneLoci()
{
- return geneLoci;
+ 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);
+ }
+
+ @Override
+ public MapList getMap()
+ {
+ return ref.getMap().getMap();
+ }
+ };
+ }
+ }
+ }
+ return null;
}
/**
* @param curs
* @return
*/
- protected int findIndex(int pos, SequenceCursor curs)
+ protected int findIndex(final int pos, SequenceCursor curs)
{
if (!isValidCursor(curs))
{
while (newPos != pos)
{
col += delta; // shift one column left or right
- if (col < 0 || col == sequence.length)
+ if (col < 0)
+ {
+ break;
+ }
+ if (col == sequence.length)
{
+ col--; // return last column if we failed to reach pos
break;
}
if (!Comparison.isGap(sequence[col]))
}
col++; // convert back to base 1
- updateCursor(pos, col, curs.firstColumnPosition);
+
+ /*
+ * only update cursor if we found the target position
+ */
+ if (newPos == pos)
+ {
+ updateCursor(pos, col, curs.firstColumnPosition);
+ }
return col;
}
* {@inheritDoc}
*/
@Override
- public Range findPositions(int fromColumn, int toColumn)
+ public ContiguousI findPositions(int fromColumn, int toColumn)
{
if (toColumn < fromColumn || fromColumn < 1)
{
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()
{
@Override
public List<int[]> getInsertions()
{
- ArrayList<int[]> map = new ArrayList<int[]>();
+ ArrayList<int[]> map = new ArrayList<>();
int lastj = -1, j = 0;
int pos = start;
int seqlen = sequence.length;
}
@Override
- public void deleteChars(int i, int j)
+ public void deleteChars(final int i, final int j)
{
int newstart = start, newend = end;
if (i >= sequence.length || i < 0)
boolean createNewDs = false;
// TODO: take a (second look) at the dataset creation validation method for
// the very large sequence case
- int eindex = -1, sindex = -1;
- boolean ecalc = false, scalc = false;
+ 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++)
{
- if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
+ if (Comparison.isGap(sequence[s]))
+ {
+ continue;
+ }
+ deleteCount++;
+ if (startDeleteColumn == -1)
+ {
+ startDeleteColumn = findPosition(s) - start;
+ }
+ if (createNewDs)
+ {
+ newend--;
+ }
+ else
{
- if (createNewDs)
+ 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
{
- if (!scalc)
+ if (endIndex < j)
{
- sindex = findIndex(start) - 1;
- scalc = true;
- }
- if (sindex == s)
- {
- // 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.
- if (!ecalc)
- {
- eindex = findIndex(end) - 1;
- ecalc = true;
- }
- 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;
{
if (this.annotation == null)
{
- this.annotation = new Vector<AlignmentAnnotation>();
+ this.annotation = new Vector<>();
}
if (!this.annotation.contains(annotation))
{
return null;
}
- Vector<AlignmentAnnotation> subset = new Vector<AlignmentAnnotation>();
+ Vector<AlignmentAnnotation> subset = new Vector<>();
Enumeration<AlignmentAnnotation> e = annotation.elements();
while (e.hasMoreElements())
{
}
}
- /**
- * @return The index (zero-based) on this sequence in the MSA. It returns
- * {@code -1} if this information is not available.
- */
- @Override
- public int getIndex()
- {
- return index;
- }
-
- /**
- * 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)
- */
- @Override
- public void setIndex(int value)
- {
- index = value;
- }
-
@Override
public void setRNA(RNA r)
{
public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
String label)
{
- List<AlignmentAnnotation> result = new ArrayList<AlignmentAnnotation>();
+ List<AlignmentAnnotation> result = new ArrayList<>();
if (this.annotation != null)
{
for (AlignmentAnnotation ann : annotation)
}
synchronized (dbrefs)
{
- List<DBRefEntry> primaries = new ArrayList<DBRefEntry>();
+ List<DBRefEntry> primaries = new ArrayList<>();
DBRefEntry[] tmp = new DBRefEntry[1];
for (DBRefEntry ref : dbrefs)
{
* 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[toColumn - 1]))
+ 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())
return count;
}
+
+ @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]));
+ }
+ }
+
+ return newSequence.toString();
+ }
+
+ @Override
+ public int firstResidueOutsideIterator(Iterator<int[]> 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;
+ }
}