X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FMappingUtils.java;h=76967deb6d22ec2ab5c4184120b02032769b310b;hb=56b18f65d6718fc6cfd7eeae53fdec97fdc7a360;hp=d21eac39a3ed6183b93c376d422f210cf42394c7;hpb=5baf7de79f1a31bb199ba619e221a1cdac99ce7d;p=jalview.git diff --git a/src/jalview/util/MappingUtils.java b/src/jalview/util/MappingUtils.java index d21eac3..76967de 100644 --- a/src/jalview/util/MappingUtils.java +++ b/src/jalview/util/MappingUtils.java @@ -20,18 +20,28 @@ */ 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; @@ -39,13 +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. * @@ -78,7 +81,7 @@ public final class MappingUtils action = action.getUndoAction(); } // TODO write this - System.err.println("MappingUtils.mapCutOrPaste not yet implemented"); + Cache.error("MappingUtils.mapCutOrPaste not yet implemented"); } /** @@ -361,56 +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()) { - 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()) + // 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()) { - 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 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 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); @@ -449,20 +451,23 @@ 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()) { - if (seq2.getDatasetSequence() == mappedSeq) + /* + * 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) { mappedOrder.add(seq2); j++; break; } } - } } } @@ -524,7 +529,7 @@ public final class MappingUtils if (colsel == null) { - return; // mappedColumns; + return; } char fromGapChar = mapFrom.getAlignment().getGapCharacter(); @@ -542,12 +547,13 @@ public final class MappingUtils toSequences, fromGapChar); } - for (int[] hidden : hiddencols.getHiddenColumnsCopy()) + Iterator regions = hiddencols.iterator(); + while (regions.hasNext()) { - mapHiddenColumns(hidden, codonFrames, newHidden, fromSequences, - toSequences, fromGapChar); + mapHiddenColumns(regions.next(), codonFrames, newHidden, + fromSequences, toSequences, fromGapChar); } - return; // mappedColumns; + return; } /** @@ -665,7 +671,9 @@ public final class MappingUtils */ for (SequenceI toSeq : toSequences) { - if (toSeq.getDatasetSequence() == mappedSeq) + if (toSeq.getDatasetSequence() == mappedSeq + && mappedStartResidue >= toSeq.getStart() + && mappedEndResidue <= toSeq.getEnd()) { int mappedStartCol = toSeq.findIndex(mappedStartResidue); int mappedEndCol = toSeq.findIndex(mappedEndResidue); @@ -834,7 +842,7 @@ public final class MappingUtils { if (range.length % 2 != 0) { - System.err.println( + Cache.error( "Error unbalance start/end ranges: " + ranges.toString()); return 0; } @@ -963,8 +971,141 @@ public final class MappingUtils 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 ranges) + { + int toRemove = positions; + Iterator it = new ReverseListIterator<>(ranges); + while (toRemove > 0) + { + int[] endRange = it.next(); + if (endRange.length != 2) + { + /* + * not coded for [start1, end1, start2, end2, ...] + */ + Cache.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.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 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. + * + * 
+   * 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; + } }