X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FMappingUtils.java;h=cf90bf94453f34982b152ac5d45649b1e79b54fe;hb=e132642765c503cb0c93dc47f304007d0527c2cc;hp=2e30132c0105fd0fde8576434688b2d86db877eb;hpb=8e1be43e250107a4d86898bd554cf03098fa5957;p=jalview.git diff --git a/src/jalview/util/MappingUtils.java b/src/jalview/util/MappingUtils.java index 2e30132..cf90bf9 100644 --- a/src/jalview/util/MappingUtils.java +++ b/src/jalview/util/MappingUtils.java @@ -28,9 +28,11 @@ 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.SearchResultMatchI; import jalview.datamodel.SearchResults; import jalview.datamodel.SearchResultsI; @@ -107,7 +109,7 @@ public final class MappingUtils * Cache a copy of the target sequences so we can mimic successive edits on * them. This lets us compute mappings for all edits in the set. */ - Map targetCopies = new HashMap(); + Map targetCopies = new HashMap<>(); for (SequenceI seq : mapTo.getSequences()) { SequenceI ds = seq.getDatasetSequence(); @@ -218,8 +220,9 @@ public final class MappingUtils * 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); @@ -229,15 +232,15 @@ public final class MappingUtils */ 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))); } } } @@ -362,50 +365,45 @@ public final class MappingUtils for (AlignedCodonFrame acf : codonFrames) { - SequenceI mappedSequence = targetIsNucleotide ? acf - .getDnaForAaSeq(selected) : acf.getAaForDnaSeq(selected); - if (mappedSequence != null) + for (SequenceI seq : mapTo.getAlignment().getSequences()) { - for (SequenceI seq : mapTo.getAlignment().getSequences()) + SequenceI peptide = targetIsNucleotide ? selected : seq; + SequenceI cds = targetIsNucleotide ? seq : selected; + SequenceToSequenceMapping s2s = acf.getCoveringMapping(cds, + peptide); + if (s2s == null) { - int mappedStartResidue = 0; - int mappedEndResidue = 0; - if (seq.getDatasetSequence() == mappedSequence) - { - /* - * 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; - } + 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()); } + + /* + * 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; } } } @@ -428,11 +426,11 @@ public final class MappingUtils * the mappings available * @return */ - public static CommandI mapOrderCommand(OrderCommand command, - boolean undo, AlignmentI mapTo, List mappings) + public static CommandI mapOrderCommand(OrderCommand command, boolean undo, + AlignmentI mapTo, List mappings) { SequenceI[] sortOrder = command.getSequenceOrder(undo); - List mappedOrder = new ArrayList(); + List mappedOrder = new ArrayList<>(); int j = 0; /* @@ -508,18 +506,19 @@ public final class MappingUtils * @param mapTo * @return */ - public static ColumnSelection mapColumnSelection(ColumnSelection colsel, - AlignViewportI mapFrom, AlignViewportI mapTo) + public static void mapColumnSelection(ColumnSelection colsel, + HiddenColumns hiddencols, AlignViewportI mapFrom, + AlignViewportI mapTo, ColumnSelection newColSel, + HiddenColumns newHidden) { boolean targetIsNucleotide = mapTo.isNucleotide(); AlignViewportI protein = targetIsNucleotide ? mapFrom : mapTo; List codonFrames = protein.getAlignment() .getCodonFrames(); - ColumnSelection mappedColumns = new ColumnSelection(); if (colsel == null) { - return mappedColumns; + return; } char fromGapChar = mapFrom.getAlignment().getGapCharacter(); @@ -533,16 +532,17 @@ public final class MappingUtils for (Integer sel : colsel.getSelected()) { - mapColumn(sel.intValue(), codonFrames, mappedColumns, fromSequences, + mapColumn(sel.intValue(), codonFrames, newColSel, fromSequences, toSequences, fromGapChar); } - for (int[] hidden : colsel.getHiddenColumns()) + Iterator regions = hiddencols.iterator(); + while (regions.hasNext()) { - mapHiddenColumns(hidden, codonFrames, mappedColumns, fromSequences, - toSequences, fromGapChar); + mapHiddenColumns(regions.next(), codonFrames, newHidden, + fromSequences, toSequences, fromGapChar); } - return mappedColumns; + return; } /** @@ -557,7 +557,7 @@ public final class MappingUtils * @param fromGapChar */ protected static void mapHiddenColumns(int[] hidden, - List mappings, ColumnSelection mappedColumns, + List mappings, HiddenColumns mappedColumns, List fromSequences, List toSequences, char fromGapChar) { @@ -590,10 +590,9 @@ public final class MappingUtils * @param toSequences * @param fromGapChar */ - protected static void mapColumn(int col, - List mappings, ColumnSelection mappedColumns, - List fromSequences, List toSequences, - char fromGapChar) + protected static void mapColumn(int col, List mappings, + ColumnSelection mappedColumns, List fromSequences, + List toSequences, char fromGapChar) { int[] mappedTo = findMappedColumns(col, mappings, fromSequences, toSequences, fromGapChar); @@ -661,7 +660,9 @@ 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); @@ -693,14 +694,14 @@ public final class MappingUtils public static List findCodonsFor(SequenceI seq, int col, List mappings) { - List result = new ArrayList(); + List result = new ArrayList<>(); int dsPos = seq.findPosition(col); for (AlignedCodonFrame mapping : mappings) { if (mapping.involvesSequence(seq)) { - List codons = mapping.getMappedCodons( - seq.getDatasetSequence(), dsPos); + List codons = mapping + .getMappedCodons(seq.getDatasetSequence(), dsPos); if (codons != null) { result.addAll(codons); @@ -772,7 +773,7 @@ public final class MappingUtils SequenceI sequence, List mappings, List filterList) { - List result = new ArrayList(); + List result = new ArrayList<>(); if (sequence == null || mappings == null) { return result; @@ -788,8 +789,9 @@ public final class MappingUtils 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; @@ -829,8 +831,8 @@ public final class MappingUtils { if (range.length % 2 != 0) { - System.err.println("Error unbalance start/end ranges: " - + ranges.toString()); + System.err.println( + "Error unbalance start/end ranges: " + ranges.toString()); return 0; } for (int i = 0; i < range.length - 1; i += 2) @@ -934,4 +936,101 @@ public final class MappingUtils } 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 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, ...] + */ + System.err.println( + "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) + */ + System.err.println( + "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 [start, end] ranges to a single array of [start, end, + * start, end ...] + * + * @param ranges + * @return + */ + public static int[] listToArray(List ranges) + { + int[] result = new int[ranges.size() * 2]; + int i = 0; + for (int[] range : ranges) + { + result[i++] = range[0]; + result[i++] = range[1]; + } + return result; + } }