X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FAlignmentUtils.java;h=ea96b3b29efeca02a5e236062ffc93a10e9daad7;hb=45e015aabe8f35a4a13be26e7630641ef8c94fbb;hp=b0a22693f57c7d0ac30dc54ab382a68d0a6bee28;hpb=4bf04707b9bb755a7cdaa829eeec02ab87b4a228;p=jalview.git diff --git a/src/jalview/analysis/AlignmentUtils.java b/src/jalview/analysis/AlignmentUtils.java index b0a2269..ea96b3b 100644 --- a/src/jalview/analysis/AlignmentUtils.java +++ b/src/jalview/analysis/AlignmentUtils.java @@ -60,6 +60,7 @@ import java.util.Map; import java.util.Map.Entry; import java.util.NoSuchElementException; import java.util.Set; +import java.util.SortedMap; import java.util.TreeMap; /** @@ -72,22 +73,26 @@ import java.util.TreeMap; public class AlignmentUtils { + private static final int CODON_LENGTH = 3; + private static final String SEQUENCE_VARIANT = "sequence_variant:"; + private static final String ID = "ID"; /** * A data model to hold the 'normal' base value at a position, and an optional * sequence variant feature */ - static class DnaVariant + static final class DnaVariant { - String base; + final String base; SequenceFeature variant; DnaVariant(String nuc) { base = nuc; + variant = null; } DnaVariant(String nuc, SequenceFeature var) @@ -95,6 +100,11 @@ public class AlignmentUtils base = nuc; variant = var; } + + public String getSource() + { + return variant == null ? null : variant.getFeatureGroup(); + } } /** @@ -427,7 +437,7 @@ public class AlignmentUtils /* * cdnaStart/End, proteinStartEnd are base 1 (for dataset sequence mapping) */ - final int mappedLength = 3 * aaSeqChars.length; + final int mappedLength = CODON_LENGTH * aaSeqChars.length; int cdnaLength = cdnaSeqChars.length; int cdnaStart = cdnaSeq.getStart(); int cdnaEnd = cdnaSeq.getEnd(); @@ -439,14 +449,14 @@ public class AlignmentUtils */ if (cdnaLength != mappedLength && cdnaLength > 2) { - String lastCodon = String.valueOf(cdnaSeqChars, cdnaLength - 3, 3) - .toUpperCase(); + String lastCodon = String.valueOf(cdnaSeqChars, + cdnaLength - CODON_LENGTH, CODON_LENGTH).toUpperCase(); for (String stop : ResidueProperties.STOP) { if (lastCodon.equals(stop)) { - cdnaEnd -= 3; - cdnaLength -= 3; + cdnaEnd -= CODON_LENGTH; + cdnaLength -= CODON_LENGTH; break; } } @@ -458,12 +468,12 @@ public class AlignmentUtils int startOffset = 0; if (cdnaLength != mappedLength && cdnaLength > 2 - && String.valueOf(cdnaSeqChars, 0, 3).toUpperCase() + && String.valueOf(cdnaSeqChars, 0, CODON_LENGTH).toUpperCase() .equals(ResidueProperties.START)) { - startOffset += 3; - cdnaStart += 3; - cdnaLength -= 3; + startOffset += CODON_LENGTH; + cdnaStart += CODON_LENGTH; + cdnaLength -= CODON_LENGTH; } if (translatesAs(cdnaSeqChars, startOffset, aaSeqChars)) @@ -472,7 +482,7 @@ public class AlignmentUtils * protein is translation of dna (+/- start/stop codons) */ MapList map = new MapList(new int[] { cdnaStart, cdnaEnd }, new int[] - { proteinStart, proteinEnd }, 3, 1); + { proteinStart, proteinEnd }, CODON_LENGTH, 1); return map; } @@ -502,10 +512,9 @@ public class AlignmentUtils int aaPos = 0; int dnaPos = cdnaStart; - for (; dnaPos < cdnaSeqChars.length - 2 - && aaPos < aaSeqChars.length; dnaPos += 3, aaPos++) + for (; dnaPos < cdnaSeqChars.length - 2 && aaPos < aaSeqChars.length; dnaPos += CODON_LENGTH, aaPos++) { - String codon = String.valueOf(cdnaSeqChars, dnaPos, 3); + String codon = String.valueOf(cdnaSeqChars, dnaPos, CODON_LENGTH); final String translated = ResidueProperties.codonTranslate(codon); /* @@ -541,9 +550,9 @@ public class AlignmentUtils { return true; } - if (dnaPos == cdnaSeqChars.length - 3) + if (dnaPos == cdnaSeqChars.length - CODON_LENGTH) { - String codon = String.valueOf(cdnaSeqChars, dnaPos, 3); + String codon = String.valueOf(cdnaSeqChars, dnaPos, CODON_LENGTH); if ("STOP".equals(ResidueProperties.codonTranslate(codon))) { return true; @@ -894,7 +903,8 @@ public class AlignmentUtils } width = Math.max(dnaSeq.getLength(), width); } - int oldwidth, diff; + int oldwidth; + int diff; for (SequenceI dnaSeq : dna.getSequences()) { oldwidth = dnaSeq.getLength(); @@ -928,15 +938,15 @@ public class AlignmentUtils .println("alignCdsSequenceAsProtein needs aligned sequence!"); return false; } - + List dnaMappings = MappingUtils .findMappingsForSequence(cdsSeq, mappings); for (AlignedCodonFrame mapping : dnaMappings) { SequenceI peptide = mapping.findAlignedSequence(cdsSeq, protein); - int peptideLength = peptide.getLength(); if (peptide != null) { + int peptideLength = peptide.getLength(); Mapping map = mapping.getMappingBetween(cdsSeq, peptide); if (map != null) { @@ -950,7 +960,8 @@ public class AlignmentUtils .getFromRanges()); int mappedToLength = MappingUtils .getLength(mapList.getToRanges()); - boolean addStopCodon = (cdsLength == mappedFromLength * 3 + 3) + boolean addStopCodon = (cdsLength == mappedFromLength + * CODON_LENGTH + CODON_LENGTH) || (peptide.getDatasetSequence().getLength() == mappedFromLength - 1); if (cdsLength != mappedToLength && !addStopCodon) { @@ -964,8 +975,8 @@ public class AlignmentUtils /* * pre-fill the aligned cds sequence with gaps */ - char[] alignedCds = new char[peptideLength * 3 - + (addStopCodon ? 3 : 0)]; + char[] alignedCds = new char[peptideLength * CODON_LENGTH + + (addStopCodon ? CODON_LENGTH : 0)]; Arrays.fill(alignedCds, gapChar); /* @@ -982,7 +993,7 @@ public class AlignmentUtils { if (Comparison.isGap(residue)) { - cdsCol += 3; + cdsCol += CODON_LENGTH; } else { @@ -991,7 +1002,7 @@ public class AlignmentUtils if (codon == null) { // e.g. incomplete start codon, X in peptide - cdsCol += 3; + cdsCol += CODON_LENGTH; } else { @@ -1009,7 +1020,7 @@ public class AlignmentUtils * append stop codon if not mapped from protein, * closing it up to the end of the mapped sequence */ - if (copiedBases == nucleotides.length - 3) + if (copiedBases == nucleotides.length - CODON_LENGTH) { for (int i = alignedCds.length - 1; i >= 0; i--) { @@ -1019,7 +1030,7 @@ public class AlignmentUtils break; } } - for (int i = nucleotides.length - 3; i < nucleotides.length; i++) + for (int i = nucleotides.length - CODON_LENGTH; i < nucleotides.length; i++) { alignedCds[cdsCol++] = nucleotides[i]; } @@ -1089,7 +1100,7 @@ public class AlignmentUtils // TODO resolve JAL-2022 so this fudge can be removed int mappedSequenceCount = protein.getHeight() - unmappedProtein.size(); addUnmappedPeptideStarts(alignedCodons, mappedSequenceCount); - + return alignedCodons; } @@ -1702,8 +1713,9 @@ public class AlignmentUtils */ List cdsRange = Collections.singletonList(new int[] { 1, cdsSeq.getLength() }); - MapList cdsToProteinMap = new MapList(cdsRange, mapList.getToRanges(), - mapList.getFromRatio(), mapList.getToRatio()); + MapList cdsToProteinMap = new MapList(cdsRange, + mapList.getToRanges(), mapList.getFromRatio(), + mapList.getToRatio()); AlignedCodonFrame cdsToProteinMapping = new AlignedCodonFrame(); cdsToProteinMapping.addMap(cdsSeqDss, proteinProduct, cdsToProteinMap); @@ -1723,8 +1735,7 @@ public class AlignmentUtils */ AlignedCodonFrame dnaToCdsMapping = new AlignedCodonFrame(); MapList dnaToCdsMap = new MapList(mapList.getFromRanges(), - cdsRange, 1, - 1); + cdsRange, 1, 1); dnaToCdsMapping.addMap(dnaSeq.getDatasetSequence(), cdsSeqDss, dnaToCdsMap); if (!mappings.contains(dnaToCdsMapping)) @@ -1739,16 +1750,16 @@ public class AlignmentUtils * same source and accession, so need a different accession for * the CDS from the dna sequence */ - + // specific use case: // Genomic contig ENSCHR:1, contains coding regions for ENSG01, // ENSG02, ENSG03, with transcripts and products similarly named. // cannot add distinct dbrefs mapping location on ENSCHR:1 to ENSG01 - + // JBPNote: ?? can't actually create an example that demonstrates we // need to // synthesize an xref. - + for (DBRefEntry primRef : dnaDss.getPrimaryDBRefs()) { // creates a complementary cross-reference to the source sequence's @@ -1778,7 +1789,7 @@ public class AlignmentUtils /* * transfer any features on dna that overlap the CDS */ - transferFeatures(dnaSeq, cdsSeq, cdsToProteinMap, null, + transferFeatures(dnaSeq, cdsSeq, dnaToCdsMap, null, SequenceOntologyI.CDS); } } @@ -1825,7 +1836,8 @@ public class AlignmentUtils int mappedFromLength = MappingUtils.getLength(aMapping.getMap() .getFromRanges()); int dnaLength = seqDss.getLength(); - if (mappedFromLength == dnaLength || mappedFromLength == dnaLength - 3) + if (mappedFromLength == dnaLength + || mappedFromLength == dnaLength - CODON_LENGTH) { return seqDss; } @@ -1841,7 +1853,8 @@ public class AlignmentUtils for (SequenceToSequenceMapping map : acf.getMappings()) { Mapping mapping = map.getMapping(); - if (mapping != aMapping && mapping.getMap().getFromRatio() == 3 + if (mapping != aMapping + && mapping.getMap().getFromRatio() == CODON_LENGTH && proteinProduct == mapping.getTo() && seqDss != map.getFromSeq()) { @@ -1909,7 +1922,7 @@ public class AlignmentUtils } } } - + /* * assign 'from id' held in the mapping if set (e.g. EMBL protein_id), * else generate a sequence name @@ -2163,7 +2176,7 @@ public class AlignmentUtils /* * dna length should map to protein (or protein plus stop codon) */ - int codesForResidues = mappedDnaLength / 3; + int codesForResidues = mappedDnaLength / CODON_LENGTH; if (codesForResidues == (proteinLength + 1)) { // assuming extra codon is for STOP and not in peptide @@ -2172,7 +2185,7 @@ public class AlignmentUtils if (codesForResidues == proteinLength) { proteinRange.add(new int[] { proteinStart, proteinEnd }); - return new MapList(ranges, proteinRange, 3, 1); + return new MapList(ranges, proteinRange, CODON_LENGTH, 1); } return null; } @@ -2448,7 +2461,7 @@ public class AlignmentUtils * are currently ignored here */ String trans = codon.contains("-") ? "-" - : (codon.length() > 3 ? null : ResidueProperties + : (codon.length() > CODON_LENGTH ? null : ResidueProperties .codonTranslate(codon)); if (trans != null && !trans.equals(residue)) { @@ -2460,7 +2473,7 @@ public class AlignmentUtils // set score to 0f so 'graduated colour' option is offered! JAL-2060 SequenceFeature sf = new SequenceFeature( SequenceOntologyI.SEQUENCE_VARIANT, desc, peptidePos, - peptidePos, 0f, "Jalview"); + peptidePos, 0f, var.getSource()); StringBuilder attributes = new StringBuilder(32); String id = (String) var.variant.getValue(ID); if (id != null) @@ -2471,11 +2484,13 @@ public class AlignmentUtils } sf.setValue(ID, id); attributes.append(ID).append("=").append(id); - // TODO handle other species variants + // TODO handle other species variants JAL-2064 StringBuilder link = new StringBuilder(32); try { - link.append(desc).append(" ").append(id) + link.append(desc) + .append(" ") + .append(id) .append("|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=") .append(URLEncoder.encode(id, "UTF-8")); sf.addLink(link.toString()); @@ -2484,8 +2499,7 @@ public class AlignmentUtils // as if } } - String clinSig = (String) var.variant - .getValue(CLINICAL_SIGNIFICANCE); + String clinSig = (String) var.variant.getValue(CLINICAL_SIGNIFICANCE); if (clinSig != null) { sf.setValue(CLINICAL_SIGNIFICANCE, clinSig); @@ -2510,6 +2524,7 @@ public class AlignmentUtils * @param dnaToProtein * @return */ + @SuppressWarnings("unchecked") static LinkedHashMap[]> buildDnaVariantsMap( SequenceI dnaSeq, MapList dnaToProtein) { @@ -2553,7 +2568,7 @@ public class AlignmentUtils List[] codonVariants = variants.get(peptidePosition); if (codonVariants == null) { - codonVariants = new ArrayList[3]; + codonVariants = new ArrayList[CODON_LENGTH]; codonVariants[0] = new ArrayList(); codonVariants[1] = new ArrayList(); codonVariants[2] = new ArrayList(); @@ -2587,7 +2602,7 @@ public class AlignmentUtils /* * save nucleotide (and any variant) for each codon position */ - for (int codonPos = 0; codonPos < 3; codonPos++) + for (int codonPos = 0; codonPos < CODON_LENGTH; codonPos++) { String nucleotide = String.valueOf( dnaSeq.getCharAt(codon[codonPos] - dnaStart)) @@ -2733,7 +2748,7 @@ public class AlignmentUtils } newCol++; } - + /* * trim trailing gaps */ @@ -2826,7 +2841,7 @@ public class AlignmentUtils * @param unmapped * @return */ - static Map> buildMappedColumnsMap( + static SortedMap> buildMappedColumnsMap( AlignmentI unaligned, AlignmentI aligned, List unmapped) { /* @@ -2834,7 +2849,7 @@ public class AlignmentUtils * {unalignedSequence, characterPerSequence} at that position. * TreeMap keeps the entries in ascending column order. */ - Map> map = new TreeMap>(); + SortedMap> map = new TreeMap>(); /* * record any sequences that have no mapping so can't be realigned