X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FMappingUtils.java;h=dbbf8a06f379946455f9c69a22fcbda3fd669ed7;hb=4a3def9f59cefe629c9a33d87483283aee085928;hp=cf90bf94453f34982b152ac5d45649b1e79b54fe;hpb=e132642765c503cb0c93dc47f304007d0527c2cc;p=jalview.git diff --git a/src/jalview/util/MappingUtils.java b/src/jalview/util/MappingUtils.java index cf90bf9..dbbf8a0 100644 --- a/src/jalview/util/MappingUtils.java +++ b/src/jalview/util/MappingUtils.java @@ -20,8 +20,16 @@ */ 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.Console; import jalview.commands.CommandI; import jalview.commands.EditCommand; import jalview.commands.EditCommand.Action; @@ -33,6 +41,7 @@ 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; @@ -40,13 +49,6 @@ import jalview.datamodel.Sequence; 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. * @@ -79,7 +81,7 @@ public final class MappingUtils action = action.getUndoAction(); } // TODO write this - System.err.println("MappingUtils.mapCutOrPaste not yet implemented"); + Console.error("MappingUtils.mapCutOrPaste not yet implemented"); } /** @@ -362,48 +364,55 @@ public final class MappingUtils */ int startResiduePos = selected.findPosition(firstUngappedPos); int endResiduePos = selected.findPosition(lastUngappedPos); - - for (AlignedCodonFrame acf : codonFrames) + for (SequenceI seq : mapTo.getAlignment().getSequences()) { - for (SequenceI seq : mapTo.getAlignment().getSequences()) + int mappedStartResidue = 0; + int mappedEndResidue = 0; + for (AlignedCodonFrame acf : codonFrames) { - SequenceI peptide = targetIsNucleotide ? selected : seq; - SequenceI cds = targetIsNucleotide ? seq : selected; - SequenceToSequenceMapping s2s = acf.getCoveringMapping(cds, - peptide); - if (s2s == null) + // rather than use acf.getCoveringMapping() we iterate through all + // mappings to make sure all CDS are selected for a protein + for (SequenceToSequenceMapping map : acf.getMappings()) { - continue; - } - int mappedStartResidue = 0; - int mappedEndResidue = 0; - List mapping = Arrays.asList(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()); - } + if (map.covers(selected) && map.covers(seq)) + { + /* + * Found a sequence mapping. Locate the start/end mapped residues. + */ + List mapping = Arrays + .asList(new AlignedCodonFrame[] + { acf }); + // locate start + SearchResultsI sr = buildSearchResults(selected, + startResiduePos, mapping); + for (SearchResultMatchI m : sr.getResults()) + { + 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; + /* + * 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; + } + } } } } @@ -443,18 +452,21 @@ public final class MappingUtils { for (AlignedCodonFrame acf : mappings) { - SequenceI mappedSeq = mappingToNucleotide ? acf.getDnaForAaSeq(seq) - : acf.getAaForDnaSeq(seq); - if (mappedSeq != null) + for (SequenceI seq2 : mapTo.getSequences()) { - for (SequenceI seq2 : mapTo.getSequences()) + /* + * the corresponding peptide / CDS is the one for which there is + * a complete ('covering') mapping to 'seq' + */ + SequenceI peptide = mappingToNucleotide ? seq2 : seq; + SequenceI cds = mappingToNucleotide ? seq : seq2; + SequenceToSequenceMapping s2s = acf.getCoveringMapping(cds, + peptide); + if (s2s != null) { - if (seq2.getDatasetSequence() == mappedSeq) - { - mappedOrder.add(seq2); - j++; - break; - } + mappedOrder.add(seq2); + j++; + break; } } } @@ -518,7 +530,7 @@ public final class MappingUtils if (colsel == null) { - return; + return; } char fromGapChar = mapFrom.getAlignment().getGapCharacter(); @@ -542,7 +554,7 @@ public final class MappingUtils mapHiddenColumns(regions.next(), codonFrames, newHidden, fromSequences, toSequences, fromGapChar); } - return; + return; } /** @@ -831,7 +843,7 @@ public final class MappingUtils { if (range.length % 2 != 0) { - System.err.println( + Console.error( "Error unbalance start/end ranges: " + ranges.toString()); return 0; } @@ -987,7 +999,7 @@ public final class MappingUtils /* * not coded for [start1, end1, start2, end2, ...] */ - System.err.println( + Console.error( "MappingUtils.removeEndPositions doesn't handle multiple ranges"); return; } @@ -998,7 +1010,7 @@ public final class MappingUtils /* * not coded for a reverse strand range (end < start) */ - System.err.println( + Console.error( "MappingUtils.removeEndPositions doesn't handle reverse strand"); return; } @@ -1016,21 +1028,102 @@ public final class MappingUtils } /** - * Converts a list of [start, end] ranges to a single array of [start, end, - * start, end ...] + * Converts a list of {@code start-end} ranges to a single array of + * {@code start1, end1, start2, ... } ranges * * @param ranges * @return */ - public static int[] listToArray(List ranges) + public static int[] rangeListToArray(List ranges) { - int[] result = new int[ranges.size() * 2]; - int i = 0; - for (int[] range : ranges) + int rangeCount = ranges.size(); + int[] result = new int[rangeCount * 2]; + int j = 0; + for (int i = 0; i < rangeCount; i++) { - result[i++] = range[0]; - result[i++] = range[1]; + 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. + * + * 
+   * 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
+   *
+ * + * @param ranges + * @param begin + * @param end + * @return + */ + protected static int[] findOverlap(List 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; + } + + public static Map putWithDuplicationCheck(Map map, K key, + V value) + { + if (!map.containsKey(key)) + { + map.put(key, value); + } + else + { + Console.warn( + "Attempt to add duplicate entry detected for map with key: " + + key.toString() + " and value: " + value.toString()); + } + + return map; + } }