X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Futil%2FMappingUtils.java;h=915293e1da6f6963dc0650447930e79cee5654c6;hb=0a37e3b824b46b026916e124b42400590242d145;hp=1bbfc7337a158e363407b355423d5de8deb16c07;hpb=a6b324e3f5edac3df0b968f0037b1cc8b651598e;p=jalview.git diff --git a/src/jalview/util/MappingUtils.java b/src/jalview/util/MappingUtils.java index 1bbfc73..915293e 100644 --- a/src/jalview/util/MappingUtils.java +++ b/src/jalview/util/MappingUtils.java @@ -31,8 +31,10 @@ import jalview.datamodel.AlignedCodonFrame; 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.SearchResults.Match; +import jalview.datamodel.SearchResultsI; import jalview.datamodel.Sequence; import jalview.datamodel.SequenceGroup; import jalview.datamodel.SequenceI; @@ -106,7 +108,7 @@ 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(); @@ -194,7 +196,7 @@ public final class MappingUtils /* * Determine all mappings from this position to mapped sequences. */ - SearchResults sr = buildSearchResults(seq, seqpos, mappings); + SearchResultsI sr = buildSearchResults(seq, seqpos, mappings); if (!sr.isEmpty()) { @@ -217,8 +219,9 @@ 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); @@ -228,15 +231,15 @@ 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))); } } } @@ -266,10 +269,10 @@ public final class MappingUtils * @param seqmappings * @return */ - public static SearchResults buildSearchResults(SequenceI seq, int index, + public static SearchResultsI buildSearchResults(SequenceI seq, int index, List seqmappings) { - SearchResults results = new SearchResults(); + SearchResultsI results = new SearchResults(); addSearchResults(results, seq, index, seqmappings); return results; } @@ -283,7 +286,7 @@ public final class MappingUtils * @param index * @param seqmappings */ - public static void addSearchResults(SearchResults results, SequenceI seq, + public static void addSearchResults(SearchResultsI results, SequenceI seq, int index, List seqmappings) { if (index >= seq.getStart() && index <= seq.getEnd()) @@ -319,7 +322,7 @@ public final class MappingUtils * Copy group name, colours etc, but not sequences or sequence colour scheme */ SequenceGroup mappedGroup = new SequenceGroup(sg); - mappedGroup.cs = mapTo.getGlobalColourScheme(); + mappedGroup.setColourScheme(mapTo.getGlobalColourScheme()); mappedGroup.clear(); int minStartCol = -1; @@ -361,8 +364,9 @@ public final class MappingUtils for (AlignedCodonFrame acf : codonFrames) { - SequenceI mappedSequence = targetIsNucleotide ? acf - .getDnaForAaSeq(selected) : acf.getAaForDnaSeq(selected); + SequenceI mappedSequence = targetIsNucleotide + ? acf.getDnaForAaSeq(selected) + : acf.getAaForDnaSeq(selected); if (mappedSequence != null) { for (SequenceI seq : mapTo.getAlignment().getSequences()) @@ -374,16 +378,18 @@ public final class MappingUtils /* * Found a sequence mapping. Locate the start/end mapped residues. */ - List mapping = Arrays.asList(new AlignedCodonFrame[] { acf }); - SearchResults sr = buildSearchResults(selected, + List mapping = Arrays + .asList(new AlignedCodonFrame[] + { acf }); + SearchResultsI sr = buildSearchResults(selected, startResiduePos, mapping); - for (Match m : sr.getResults()) + for (SearchResultMatchI m : sr.getResults()) { mappedStartResidue = m.getStart(); mappedEndResidue = m.getEnd(); } sr = buildSearchResults(selected, endResiduePos, mapping); - for (Match m : sr.getResults()) + for (SearchResultMatchI m : sr.getResults()) { mappedStartResidue = Math.min(mappedStartResidue, m.getStart()); @@ -395,11 +401,11 @@ public final class MappingUtils * returns a base 1 position, SequenceGroup uses base 0 */ int mappedStartCol = seq.findIndex(mappedStartResidue) - 1; - minStartCol = minStartCol == -1 ? mappedStartCol : Math.min( - minStartCol, mappedStartCol); + minStartCol = minStartCol == -1 ? mappedStartCol + : Math.min(minStartCol, mappedStartCol); int mappedEndCol = seq.findIndex(mappedEndResidue) - 1; - maxEndCol = maxEndCol == -1 ? mappedEndCol : Math.max( - maxEndCol, mappedEndCol); + maxEndCol = maxEndCol == -1 ? mappedEndCol + : Math.max(maxEndCol, mappedEndCol); mappedGroup.addSequence(seq, false); break; } @@ -426,11 +432,11 @@ 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; /* @@ -506,18 +512,19 @@ 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; // mappedColumns; } char fromGapChar = mapFrom.getAlignment().getGapCharacter(); @@ -531,16 +538,18 @@ 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, + mapHiddenColumns(regions.next(), codonFrames, newHidden, + fromSequences, toSequences, fromGapChar); } - return mappedColumns; + return; // mappedColumns; } /** @@ -555,9 +564,9 @@ public final class MappingUtils * @param fromGapChar */ protected static void mapHiddenColumns(int[] hidden, - List mappings, - ColumnSelection mappedColumns, List fromSequences, - List toSequences, char fromGapChar) + List mappings, HiddenColumns mappedColumns, + List fromSequences, List toSequences, + char fromGapChar) { for (int col = hidden[0]; col <= hidden[1]; col++) { @@ -588,8 +597,7 @@ public final class MappingUtils * @param toSequences * @param fromGapChar */ - protected static void mapColumn(int col, - List mappings, + protected static void mapColumn(int col, List mappings, ColumnSelection mappedColumns, List fromSequences, List toSequences, char fromGapChar) { @@ -646,9 +654,8 @@ public final class MappingUtils * Get the residue position and find the mapped position. */ int residuePos = fromSeq.findPosition(col); - SearchResults sr = buildSearchResults(fromSeq, residuePos, - mappings); - for (Match m : sr.getResults()) + SearchResultsI sr = buildSearchResults(fromSeq, residuePos, mappings); + for (SearchResultMatchI m : sr.getResults()) { int mappedStartResidue = m.getStart(); int mappedEndResidue = m.getEnd(); @@ -692,14 +699,14 @@ 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); @@ -754,7 +761,24 @@ public final class MappingUtils public static List findMappingsForSequence( SequenceI sequence, List mappings) { - List result = new ArrayList(); + return findMappingsForSequenceAndOthers(sequence, mappings, null); + } + + /** + * Returns a list of any mappings that are from or to the given (aligned or + * dataset) sequence, optionally limited to mappings involving one of a given + * list of sequences. + * + * @param sequence + * @param mappings + * @param filterList + * @return + */ + public static List findMappingsForSequenceAndOthers( + SequenceI sequence, List mappings, + List filterList) + { + List result = new ArrayList<>(); if (sequence == null || mappings == null) { return result; @@ -763,9 +787,256 @@ public final class MappingUtils { if (mapping.involvesSequence(sequence)) { - result.add(mapping); + if (filterList != null) + { + for (SequenceI otherseq : filterList) + { + SequenceI otherDataset = otherseq.getDatasetSequence(); + if (otherseq == sequence + || otherseq == sequence.getDatasetSequence() + || (otherDataset != null && (otherDataset == sequence + || otherDataset == sequence + .getDatasetSequence()))) + { + // skip sequences in subset which directly relate to sequence + continue; + } + if (mapping.involvesSequence(otherseq)) + { + // selected a mapping contained in subselect alignment + result.add(mapping); + break; + } + } + } + else + { + result.add(mapping); + } } } return result; } + + /** + * Returns the total length of the supplied ranges, which may be as single + * [start, end] or multiple [start, end, start, end ...] + * + * @param ranges + * @return + */ + public static int getLength(List ranges) + { + if (ranges == null) + { + return 0; + } + int length = 0; + for (int[] range : ranges) + { + if (range.length % 2 != 0) + { + System.err.println( + "Error unbalance start/end ranges: " + ranges.toString()); + return 0; + } + for (int i = 0; i < range.length - 1; i += 2) + { + length += Math.abs(range[i + 1] - range[i]) + 1; + } + } + return length; + } + + /** + * Answers true if any range includes the given value + * + * @param ranges + * @param value + * @return + */ + public static boolean contains(List ranges, int value) + { + if (ranges == null) + { + return false; + } + for (int[] range : ranges) + { + if (range[1] >= range[0] && value >= range[0] && value <= range[1]) + { + /* + * value within ascending range + */ + return true; + } + if (range[1] < range[0] && value <= range[0] && value >= range[1]) + { + /* + * value within descending range + */ + return true; + } + } + return false; + } + + /** + * Removes a specified number of positions from the start of a ranges list. + * For example, could be used to adjust cds ranges to allow for an incomplete + * start codon. Subranges are removed completely, or their start positions + * adjusted, until the required number of positions has been removed from the + * range. Reverse strand ranges are supported. The input array is not + * modified. + * + * @param removeCount + * @param ranges + * an array of [start, end, start, end...] positions + * @return a new array with the first removeCount positions removed + */ + public static int[] removeStartPositions(int removeCount, + final int[] ranges) + { + if (removeCount <= 0) + { + return ranges; + } + + int[] copy = Arrays.copyOf(ranges, ranges.length); + int sxpos = -1; + int cdspos = 0; + for (int x = 0; x < copy.length && sxpos == -1; x += 2) + { + cdspos += Math.abs(copy[x + 1] - copy[x]) + 1; + if (removeCount < cdspos) + { + /* + * we have removed enough, time to finish + */ + sxpos = x; + + /* + * increment start of first exon, or decrement if reverse strand + */ + if (copy[x] <= copy[x + 1]) + { + copy[x] = copy[x + 1] - cdspos + removeCount + 1; + } + else + { + copy[x] = copy[x + 1] + cdspos - removeCount - 1; + } + break; + } + } + + if (sxpos > 0) + { + /* + * we dropped at least one entire sub-range - compact the array + */ + int[] nxon = new int[copy.length - sxpos]; + System.arraycopy(copy, sxpos, nxon, 0, copy.length - sxpos); + return nxon; + } + 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; + } }