package jalview.util;
+import java.util.ArrayList;
+import java.util.Collections;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+
import jalview.analysis.AlignmentSorter;
import jalview.api.AlignViewportI;
import jalview.commands.CommandI;
import jalview.datamodel.SequenceGroup;
import jalview.datamodel.SequenceI;
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.Iterator;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
-
/**
* Helper methods for manipulations involving sequence mappings.
*
* @param mapTo
* @return
*/
- public static SequenceGroup mapSequenceGroup(SequenceGroup sg,
- AlignViewportI mapFrom, AlignViewportI mapTo)
+ public static SequenceGroup mapSequenceGroup(final SequenceGroup sg,
+ final AlignViewportI mapFrom, final AlignViewportI mapTo)
{
/*
* Note the SequenceGroup holds aligned sequences, the mappings hold dataset
AlignViewportI protein = targetIsNucleotide ? mapFrom : mapTo;
Set<AlignedCodonFrame> codonFrames = protein.getAlignment()
.getCodonFrames();
-
/*
- * Copy group name, name colours, but not sequences or sequence colour
- * scheme
+ * Copy group name, colours etc, but not sequences or sequence colour scheme
*/
SequenceGroup mappedGroup = new SequenceGroup(sg);
mappedGroup.cs = mapTo.getGlobalColourScheme();
mappedGroup.clear();
- // TODO set width of mapped group
+ int minStartCol = -1;
+ int maxEndCol = -1;
+ final int selectionStartRes = sg.getStartRes();
+ final int selectionEndRes = sg.getEndRes();
for (SequenceI selected : sg.getSequences())
{
+ /*
+ * Find the widest range of non-gapped positions in the selection range
+ */
+ int firstUngappedPos = selectionStartRes;
+ while (firstUngappedPos <= selectionEndRes
+ && Comparison.isGap(selected.getCharAt(firstUngappedPos)))
+ {
+ firstUngappedPos++;
+ }
+
+ /*
+ * If this sequence is only gaps in the selected range, skip it
+ */
+ if (firstUngappedPos > selectionEndRes)
+ {
+ continue;
+ }
+
+ int lastUngappedPos = selectionEndRes;
+ while (lastUngappedPos >= selectionStartRes
+ && Comparison.isGap(selected.getCharAt(lastUngappedPos)))
+ {
+ lastUngappedPos--;
+ }
+
+ /*
+ * Find the selected start/end residue positions in sequence
+ */
+ int startResiduePos = selected.findPosition(firstUngappedPos);
+ int endResiduePos = selected.findPosition(lastUngappedPos);
+
for (AlignedCodonFrame acf : codonFrames)
{
SequenceI mappedSequence = targetIsNucleotide ? acf
{
for (SequenceI seq : mapTo.getAlignment().getSequences())
{
+ int mappedStartResidue = 0;
+ int mappedEndResidue = 0;
if (seq.getDatasetSequence() == mappedSequence)
{
+ /*
+ * Found a sequence mapping. Locate the start/end mapped residues.
+ */
+ SearchResults sr = buildSearchResults(selected,
+ startResiduePos, Collections.singleton(acf));
+ for (Match m : sr.getResults())
+ {
+ mappedStartResidue = m.getStart();
+ mappedEndResidue = m.getEnd();
+ }
+ sr = buildSearchResults(selected, endResiduePos,
+ Collections.singleton(acf));
+ for (Match m : sr.getResults())
+ {
+ mappedStartResidue = Math.min(mappedStartResidue,
+ m.getStart());
+ mappedEndResidue = Math.max(mappedEndResidue, m.getEnd());
+ }
+
+ /*
+ * Find the mapped aligned columns, save the range. Note findIndex
+ * returns a base 1 position, SequenceGroup uses base 0
+ */
+ int mappedStartCol = seq.findIndex(mappedStartResidue) - 1;
+ minStartCol = minStartCol == -1 ? mappedStartCol : Math.min(
+ minStartCol, mappedStartCol);
+ int mappedEndCol = seq.findIndex(mappedEndResidue) - 1;
+ maxEndCol = maxEndCol == -1 ? mappedEndCol : Math.max(
+ maxEndCol, mappedEndCol);
mappedGroup.addSequence(seq, false);
break;
}
}
}
}
+ mappedGroup.setStartRes(minStartCol < 0 ? 0 : minStartCol);
+ mappedGroup.setEndRes(maxEndCol < 0 ? 0 : maxEndCol);
return mappedGroup;
}