*/
package jalview.util;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.HashMap;
+import java.util.Iterator;
+import java.util.List;
+import java.util.Map;
+
import jalview.analysis.AlignmentSorter;
import jalview.api.AlignViewportI;
+import jalview.bin.Cache;
import jalview.commands.CommandI;
import jalview.commands.EditCommand;
import jalview.commands.EditCommand.Action;
import jalview.commands.EditCommand.Edit;
import jalview.commands.OrderCommand;
import jalview.datamodel.AlignedCodonFrame;
+import jalview.datamodel.AlignedCodonFrame.SequenceToSequenceMapping;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.AlignmentOrder;
import jalview.datamodel.ColumnSelection;
import jalview.datamodel.HiddenColumns;
+import jalview.datamodel.Mapping;
import jalview.datamodel.SearchResultMatchI;
import jalview.datamodel.SearchResults;
import jalview.datamodel.SearchResultsI;
import jalview.datamodel.SequenceGroup;
import jalview.datamodel.SequenceI;
-import java.util.ArrayList;
-import java.util.Arrays;
-import java.util.HashMap;
-import java.util.Iterator;
-import java.util.List;
-import java.util.Map;
-
/**
* Helper methods for manipulations involving sequence mappings.
*
action = action.getUndoAction();
}
// TODO write this
- System.err.println("MappingUtils.mapCutOrPaste not yet implemented");
+ Cache.log.error("MappingUtils.mapCutOrPaste not yet implemented");
}
/**
* Shift Delete start position left, as it acts on positions to its
* right.
*/
- int mappedEditPos = action == Action.DELETE_GAP ? match[0]
- - mappedCount : match[0];
+ int mappedEditPos = action == Action.DELETE_GAP
+ ? match[0] - mappedCount
+ : match[0];
Edit e = result.new Edit(action, new SequenceI[] { targetSeq },
mappedEditPos, mappedCount, gapChar);
result.addEdit(e);
*/
if (action == Action.INSERT_GAP)
{
- copyTarget.setSequence(new String(StringUtils.insertCharAt(
- copyTarget.getSequence(), mappedEditPos, mappedCount,
- gapChar)));
+ copyTarget.setSequence(new String(
+ StringUtils.insertCharAt(copyTarget.getSequence(),
+ mappedEditPos, mappedCount, gapChar)));
}
else if (action == Action.DELETE_GAP)
{
- copyTarget.setSequence(new String(StringUtils.deleteChars(
- copyTarget.getSequence(), mappedEditPos,
- mappedEditPos + mappedCount)));
+ copyTarget.setSequence(new String(
+ StringUtils.deleteChars(copyTarget.getSequence(),
+ mappedEditPos, mappedEditPos + mappedCount)));
}
}
}
*/
int startResiduePos = selected.findPosition(firstUngappedPos);
int endResiduePos = selected.findPosition(lastUngappedPos);
-
- for (AlignedCodonFrame acf : codonFrames)
+ for (SequenceI seq : mapTo.getAlignment().getSequences())
{
- SequenceI mappedSequence = targetIsNucleotide ? acf
- .getDnaForAaSeq(selected) : acf.getAaForDnaSeq(selected);
- if (mappedSequence != null)
+ int mappedStartResidue = 0;
+ int mappedEndResidue = 0;
+ for (AlignedCodonFrame acf : codonFrames)
{
- for (SequenceI seq : mapTo.getAlignment().getSequences())
+ for (SequenceToSequenceMapping map: acf.getMappings())
{
- int mappedStartResidue = 0;
- int mappedEndResidue = 0;
- if (seq.getDatasetSequence() == mappedSequence)
+ if (map.covers(selected) && map.covers(seq))
+ {
+ /*
+ * Found a sequence mapping. Locate the start/end mapped residues.
+ */
+ List<AlignedCodonFrame> mapping = Arrays
+ .asList(new AlignedCodonFrame[]
+ { acf });
+ // locate start
+ SearchResultsI sr = buildSearchResults(selected,
+ startResiduePos, mapping);
+ for (SearchResultMatchI m : sr.getResults())
{
- /*
- * Found a sequence mapping. Locate the start/end mapped residues.
- */
- List<AlignedCodonFrame> mapping = Arrays
- .asList(new AlignedCodonFrame[] { acf });
- SearchResultsI sr = buildSearchResults(selected,
- startResiduePos, mapping);
- for (SearchResultMatchI m : sr.getResults())
- {
- mappedStartResidue = m.getStart();
- mappedEndResidue = m.getEnd();
- }
- sr = buildSearchResults(selected, endResiduePos, mapping);
- for (SearchResultMatchI 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;
+ mappedStartResidue = m.getStart();
+ mappedEndResidue = m.getEnd();
}
+ // locate end - allowing for adjustment of start range
+ sr = buildSearchResults(selected, endResiduePos, mapping);
+ for (SearchResultMatchI 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);
* the mappings available
* @return
*/
- public static CommandI mapOrderCommand(OrderCommand command,
- boolean undo, AlignmentI mapTo, List<AlignedCodonFrame> mappings)
+ public static CommandI mapOrderCommand(OrderCommand command, boolean undo,
+ AlignmentI mapTo, List<AlignedCodonFrame> mappings)
{
SequenceI[] sortOrder = command.getSequenceOrder(undo);
List<SequenceI> mappedOrder = new ArrayList<>();
AlignViewportI protein = targetIsNucleotide ? mapFrom : mapTo;
List<AlignedCodonFrame> codonFrames = protein.getAlignment()
.getCodonFrames();
- // ColumnSelection mappedColumns = new ColumnSelection();
if (colsel == null)
{
toSequences, fromGapChar);
}
- for (int[] hidden : hiddencols)
+ Iterator<int[]> regions = hiddencols.iterator();
+ while (regions.hasNext())
{
- mapHiddenColumns(hidden, codonFrames, newHidden, fromSequences,
- toSequences, fromGapChar);
+ mapHiddenColumns(regions.next(), codonFrames, newHidden,
+ fromSequences, toSequences, fromGapChar);
}
return; // mappedColumns;
}
* @param toSequences
* @param fromGapChar
*/
- protected static void mapColumn(int col,
- List<AlignedCodonFrame> mappings, ColumnSelection mappedColumns,
- List<SequenceI> fromSequences, List<SequenceI> toSequences,
- char fromGapChar)
+ protected static void mapColumn(int col, List<AlignedCodonFrame> mappings,
+ ColumnSelection mappedColumns, List<SequenceI> fromSequences,
+ List<SequenceI> toSequences, char fromGapChar)
{
int[] mappedTo = findMappedColumns(col, mappings, fromSequences,
toSequences, fromGapChar);
{
if (mapping.involvesSequence(seq))
{
- List<char[]> codons = mapping.getMappedCodons(
- seq.getDatasetSequence(), dsPos);
+ List<char[]> codons = mapping
+ .getMappedCodons(seq.getDatasetSequence(), dsPos);
if (codons != null)
{
result.addAll(codons);
SequenceI otherDataset = otherseq.getDatasetSequence();
if (otherseq == sequence
|| otherseq == sequence.getDatasetSequence()
- || (otherDataset != null && (otherDataset == sequence || otherDataset == sequence
- .getDatasetSequence())))
+ || (otherDataset != null && (otherDataset == sequence
+ || otherDataset == sequence
+ .getDatasetSequence())))
{
// skip sequences in subset which directly relate to sequence
continue;
{
if (range.length % 2 != 0)
{
- System.err.println("Error unbalance start/end ranges: "
- + ranges.toString());
+ Cache.log.error(
+ "Error unbalance start/end ranges: " + ranges.toString());
return 0;
}
for (int i = 0; i < range.length - 1; i += 2)
}
return copy;
}
+
+ /**
+ * Answers true if range's start-end positions include those of queryRange,
+ * where either range might be in reverse direction, else false
+ *
+ * @param range
+ * a start-end range
+ * @param queryRange
+ * a candidate subrange of range (start2-end2)
+ * @return
+ */
+ public static boolean rangeContains(int[] range, int[] queryRange)
+ {
+ if (range == null || queryRange == null || range.length != 2
+ || queryRange.length != 2)
+ {
+ /*
+ * invalid arguments
+ */
+ return false;
+ }
+
+ int min = Math.min(range[0], range[1]);
+ int max = Math.max(range[0], range[1]);
+
+ return (min <= queryRange[0] && max >= queryRange[0]
+ && min <= queryRange[1] && max >= queryRange[1]);
+ }
+
+ /**
+ * Removes the specified number of positions from the given ranges. Provided
+ * to allow a stop codon to be stripped from a CDS sequence so that it matches
+ * the peptide translation length.
+ *
+ * @param positions
+ * @param ranges
+ * a list of (single) [start, end] ranges
+ * @return
+ */
+ public static void removeEndPositions(int positions, List<int[]> ranges)
+ {
+ int toRemove = positions;
+ Iterator<int[]> it = new ReverseListIterator<>(ranges);
+ while (toRemove > 0)
+ {
+ int[] endRange = it.next();
+ if (endRange.length != 2)
+ {
+ /*
+ * not coded for [start1, end1, start2, end2, ...]
+ */
+ Cache.log.error(
+ "MappingUtils.removeEndPositions doesn't handle multiple ranges");
+ return;
+ }
+
+ int length = endRange[1] - endRange[0] + 1;
+ if (length <= 0)
+ {
+ /*
+ * not coded for a reverse strand range (end < start)
+ */
+ Cache.log.error(
+ "MappingUtils.removeEndPositions doesn't handle reverse strand");
+ return;
+ }
+ if (length > toRemove)
+ {
+ endRange[1] -= toRemove;
+ toRemove = 0;
+ }
+ else
+ {
+ toRemove -= length;
+ it.remove();
+ }
+ }
+ }
+
+ /**
+ * Converts a list of {@code start-end} ranges to a single array of
+ * {@code start1, end1, start2, ... } ranges
+ *
+ * @param ranges
+ * @return
+ */
+ public static int[] rangeListToArray(List<int[]> ranges)
+ {
+ int rangeCount = ranges.size();
+ int[] result = new int[rangeCount * 2];
+ int j = 0;
+ for (int i = 0; i < rangeCount; i++)
+ {
+ int[] range = ranges.get(i);
+ result[j++] = range[0];
+ result[j++] = range[1];
+ }
+ return result;
+ }
+
+ /*
+ * Returns the maximal start-end positions in the given (ordered) list of
+ * ranges which is overlapped by the given begin-end range, or null if there
+ * is no overlap.
+ *
+ * <pre>
+ * Examples:
+ * if ranges is {[4, 8], [10, 12], [16, 19]}
+ * then
+ * findOverlap(ranges, 1, 20) == [4, 19]
+ * findOverlap(ranges, 6, 11) == [6, 11]
+ * findOverlap(ranges, 9, 15) == [10, 12]
+ * findOverlap(ranges, 13, 15) == null
+ * </pre>
+ *
+ * @param ranges
+ * @param begin
+ * @param end
+ * @return
+ */
+ protected static int[] findOverlap(List<int[]> ranges, final int begin,
+ final int end)
+ {
+ boolean foundStart = false;
+ int from = 0;
+ int to = 0;
+
+ /*
+ * traverse the ranges to find the first position (if any) >= begin,
+ * and the last position (if any) <= end
+ */
+ for (int[] range : ranges)
+ {
+ if (!foundStart)
+ {
+ if (range[0] >= begin)
+ {
+ /*
+ * first range that starts with, or follows, begin
+ */
+ foundStart = true;
+ from = Math.max(range[0], begin);
+ }
+ else if (range[1] >= begin)
+ {
+ /*
+ * first range that contains begin
+ */
+ foundStart = true;
+ from = begin;
+ }
+ }
+
+ if (range[0] <= end)
+ {
+ to = Math.min(end, range[1]);
+ }
+ }
+
+ return foundStart && to >= from ? new int[] { from, to } : null;
+ }
}