X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fdatamodel%2FSequence.java;h=441d8d04df759b273b8bf56b519e1fc02b3c20a4;hb=537da9c65518c52985308b67c499fe6a60a607ce;hp=2f1da7f9ad9c125d334178b5e34a4bbc40422b24;hpb=be762d8d9c71a7aa3121e845c45911c7192b7827;p=jalview.git diff --git a/src/jalview/datamodel/Sequence.java b/src/jalview/datamodel/Sequence.java index 2f1da7f..441d8d0 100755 --- a/src/jalview/datamodel/Sequence.java +++ b/src/jalview/datamodel/Sequence.java @@ -38,8 +38,6 @@ 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; /** @@ -51,11 +49,6 @@ 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; @@ -84,12 +77,7 @@ public class Sequence extends ASequence implements SequenceI */ Vector 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, @@ -151,6 +139,10 @@ public class Sequence extends ASequence implements SequenceI 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) @@ -159,17 +151,37 @@ public class Sequence extends ASequence implements SequenceI "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 : @@ -178,7 +190,7 @@ public class Sequence extends ASequence implements SequenceI int endRes = 0; for (int j = 0; j < sequence.length; j++) { - if (!jalview.util.Comparison.isGap(sequence[j])) + if (!Comparison.isGap(sequence[j])) { endRes++; } @@ -453,15 +465,15 @@ public class Sequence extends ASequence implements SequenceI } /** - * 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(); } @@ -645,10 +657,10 @@ public class Sequence extends ASequence 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) @@ -656,10 +668,67 @@ public class Sequence extends ASequence implements SequenceI this.description = desc; } + @Override + public void setGeneLoci(String speciesId, String assemblyId, + String chromosomeId, MapList map) + { + addDBRef(new DBRefEntry(speciesId, assemblyId, DBRefEntry.CHROMOSOME + + ":" + chromosomeId, new Mapping(map))); + } + /** - * DOCUMENT ME! + * Returns the gene loci mapping for the sequence (may be null) * - * @return DOCUMENT ME! + * @return + */ + @Override + public GeneLociI getGeneLoci() + { + DBRefEntry[] refs = getDBRefs(); + if (refs != null) + { + for (final DBRefEntry ref : refs) + { + if (ref.isChromosome()) + { + return new GeneLociI() + { + @Override + public String getSpeciesId() + { + return ref.getSource(); + } + + @Override + public String getAssemblyId() + { + return ref.getVersion(); + } + + @Override + public String getChromosomeId() + { + // strip off "chromosome:" prefix to chrId + return ref.getAccessionId().substring( + DBRefEntry.CHROMOSOME.length() + 1); + } + + @Override + public MapList getMap() + { + return ref.getMap().getMap(); + } + }; + } + } + } + return null; + } + + /** + * Answers the description + * + * @return */ @Override public String getDescription() @@ -1148,7 +1217,7 @@ public class Sequence extends ASequence implements SequenceI } @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) @@ -1160,62 +1229,75 @@ public class Sequence extends ASequence implements SequenceI 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) { - 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 { - if (!scalc) - { - sindex = findIndex(start) - 1; - scalc = true; - } - 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. - 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; @@ -1665,30 +1747,6 @@ public class Sequence extends ASequence implements SequenceI } } - /** - * @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) { @@ -1827,7 +1885,9 @@ public class Sequence extends ASequence implements SequenceI * 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 it = result.listIterator(); while (it.hasNext())