From: gmungoc Date: Fri, 26 Aug 2016 11:09:27 +0000 (+0100) Subject: JAL-2049 set Ensembl variant source to ENSEMBL if '.' (not provided) and X-Git-Tag: Release_2_10_0~75 X-Git-Url: http://source.jalview.org/gitweb/?a=commitdiff_plain;h=6b5f772a654104d856da1651c3d411981be463df;p=jalview.git JAL-2049 set Ensembl variant source to ENSEMBL if '.' (not provided) and in computed peptide variants --- diff --git a/src/jalview/analysis/AlignmentUtils.java b/src/jalview/analysis/AlignmentUtils.java index d93f42f..d1cd5a3 100644 --- a/src/jalview/analysis/AlignmentUtils.java +++ b/src/jalview/analysis/AlignmentUtils.java @@ -73,6 +73,7 @@ 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"; @@ -80,15 +81,16 @@ public class AlignmentUtils * 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) @@ -96,6 +98,11 @@ public class AlignmentUtils base = nuc; variant = var; } + + public String getSource() + { + return variant == null ? null : variant.getFeatureGroup(); + } } /** @@ -428,7 +435,7 @@ 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(); @@ -440,14 +447,14 @@ public class AlignmentUtils */ if (cdnaLength != mappedLength && cdnaLength > 2) { - String lastCodon = String.valueOf(cdnaSeqChars, cdnaLength - 3, 3) + 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; } } @@ -459,12 +466,12 @@ 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)) @@ -473,7 +480,7 @@ 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; } @@ -504,9 +511,9 @@ public class AlignmentUtils int aaPos = 0; int dnaPos = cdnaStart; for (; dnaPos < cdnaSeqChars.length - 2 - && aaPos < aaSeqChars.length; dnaPos += 3, aaPos++) + && 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); /* @@ -542,9 +549,9 @@ 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; @@ -895,7 +902,8 @@ public class AlignmentUtils } width = Math.max(dnaSeq.getLength(), width); } - int oldwidth, diff; + int oldwidth; + int diff; for (SequenceI dnaSeq : dna.getSequences()) { oldwidth = dnaSeq.getLength(); @@ -935,9 +943,9 @@ public class AlignmentUtils 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) { @@ -951,7 +959,7 @@ 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) { @@ -965,8 +973,8 @@ 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); /* @@ -983,7 +991,7 @@ public class AlignmentUtils { if (Comparison.isGap(residue)) { - cdsCol += 3; + cdsCol += CODON_LENGTH; } else { @@ -992,7 +1000,7 @@ public class AlignmentUtils if (codon == null) { // e.g. incomplete start codon, X in peptide - cdsCol += 3; + cdsCol += CODON_LENGTH; } else { @@ -1010,7 +1018,7 @@ 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--) { @@ -1020,7 +1028,7 @@ 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]; } @@ -1806,7 +1814,7 @@ 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; } @@ -1822,7 +1830,7 @@ 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()) { @@ -2027,7 +2035,7 @@ 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 @@ -2036,7 +2044,7 @@ 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; } @@ -2312,7 +2320,7 @@ 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)) { @@ -2324,7 +2332,7 @@ 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) @@ -2335,7 +2343,7 @@ 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 { @@ -2374,6 +2382,7 @@ public class AlignmentUtils * @param dnaToProtein * @return */ + @SuppressWarnings("unchecked") static LinkedHashMap[]> buildDnaVariantsMap( SequenceI dnaSeq, MapList dnaToProtein) { @@ -2417,7 +2426,7 @@ 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(); @@ -2451,7 +2460,7 @@ 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)) diff --git a/src/jalview/ext/ensembl/EnsemblSeqProxy.java b/src/jalview/ext/ensembl/EnsemblSeqProxy.java index 31552af..cc002e1 100644 --- a/src/jalview/ext/ensembl/EnsemblSeqProxy.java +++ b/src/jalview/ext/ensembl/EnsemblSeqProxy.java @@ -613,6 +613,10 @@ public abstract class EnsemblSeqProxy extends EnsemblRestClient SequenceFeature copy = new SequenceFeature(sf); copy.setBegin(Math.min(mappedRange[0], mappedRange[1])); copy.setEnd(Math.max(mappedRange[0], mappedRange[1])); + if (".".equals(copy.getFeatureGroup())) + { + copy.setFeatureGroup(getDbSource()); + } targetSequence.addSequenceFeature(copy); /* diff --git a/test/jalview/analysis/AlignmentUtilsTests.java b/test/jalview/analysis/AlignmentUtilsTests.java index 22bb680..a856231 100644 --- a/test/jalview/analysis/AlignmentUtilsTests.java +++ b/test/jalview/analysis/AlignmentUtilsTests.java @@ -1937,13 +1937,15 @@ public class AlignmentUtilsTests public void testComputePeptideVariants() { /* - * scenario: AAATTTCCC codes for KFP, with variants - * GAA -> E - * CAA -> Q - * AAG synonymous - * AAT -> N - * TTC synonymous - * CAC,CGC -> H,R (as one variant) + * scenario: AAATTTCCC codes for KFP + * variants: + * GAA -> E source: Ensembl + * CAA -> Q source: dbSNP + * AAG synonymous source: COSMIC + * AAT -> N source: Ensembl + * ...TTC synonymous source: dbSNP + * ......CAC,CGC -> H,R source: COSMIC + * (one variant with two alleles) */ SequenceI peptide = new Sequence("pep/10-12", "KFP"); @@ -1951,32 +1953,35 @@ public class AlignmentUtilsTests * two distinct variants for codon 1 position 1 * second one has clinical significance */ + String ensembl = "Ensembl"; + String dbSnp = "dbSNP"; + String cosmic = "COSMIC"; SequenceFeature sf1 = new SequenceFeature("sequence_variant", "", 1, 1, - 0f, null); + 0f, ensembl); sf1.setValue("alleles", "A,G"); // GAA -> E sf1.setValue("ID", "var1.125A>G"); SequenceFeature sf2 = new SequenceFeature("sequence_variant", "", 1, 1, - 0f, null); + 0f, dbSnp); sf2.setValue("alleles", "A,C"); // CAA -> Q sf2.setValue("ID", "var2"); sf2.setValue("clinical_significance", "Dodgy"); SequenceFeature sf3 = new SequenceFeature("sequence_variant", "", 3, 3, - 0f, null); + 0f, cosmic); sf3.setValue("alleles", "A,G"); // synonymous sf3.setValue("ID", "var3"); sf3.setValue("clinical_significance", "None"); SequenceFeature sf4 = new SequenceFeature("sequence_variant", "", 3, 3, - 0f, null); + 0f, ensembl); sf4.setValue("alleles", "A,T"); // AAT -> N sf4.setValue("ID", "sequence_variant:var4"); // prefix gets stripped off sf4.setValue("clinical_significance", "Benign"); SequenceFeature sf5 = new SequenceFeature("sequence_variant", "", 6, 6, - 0f, null); + 0f, dbSnp); sf5.setValue("alleles", "T,C"); // synonymous sf5.setValue("ID", "var5"); sf5.setValue("clinical_significance", "Bad"); SequenceFeature sf6 = new SequenceFeature("sequence_variant", "", 8, 8, - 0f, null); + 0f, cosmic); sf6.setValue("alleles", "C,A,G"); // CAC,CGC -> H,R sf6.setValue("ID", "var6"); sf6.setValue("clinical_significance", "Good"); @@ -2024,14 +2029,15 @@ public class AlignmentUtilsTests /* * verify added sequence features for - * var1 K -> E - * var2 K -> Q - * var4 K -> N - * var6 P -> H - * var6 P -> R + * var1 K -> E Ensembl + * var2 K -> Q dbSNP + * var4 K -> N Ensembl + * var6 P -> H COSMIC + * var6 P -> R COSMIC */ SequenceFeature[] sfs = peptide.getSequenceFeatures(); assertEquals(5, sfs.length); + SequenceFeature sf = sfs[0]; assertEquals(1, sf.getBegin()); assertEquals(1, sf.getEnd()); @@ -2044,7 +2050,8 @@ public class AlignmentUtilsTests assertEquals( "p.Lys1Glu var1.125A>G|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var1.125A%3EG", sf.links.get(0)); - assertEquals("Jalview", sf.getFeatureGroup()); + assertEquals(ensembl, sf.getFeatureGroup()); + sf = sfs[1]; assertEquals(1, sf.getBegin()); assertEquals(1, sf.getEnd()); @@ -2056,7 +2063,8 @@ public class AlignmentUtilsTests assertEquals( "p.Lys1Gln var2|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var2", sf.links.get(0)); - assertEquals("Jalview", sf.getFeatureGroup()); + assertEquals(dbSnp, sf.getFeatureGroup()); + sf = sfs[2]; assertEquals(1, sf.getBegin()); assertEquals(1, sf.getEnd()); @@ -2068,7 +2076,9 @@ public class AlignmentUtilsTests assertEquals( "p.Lys1Asn var4|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var4", sf.links.get(0)); - assertEquals("Jalview", sf.getFeatureGroup()); + assertEquals(ensembl, sf.getFeatureGroup()); + + // var5 generates two distinct protein variant features sf = sfs[3]; assertEquals(3, sf.getBegin()); assertEquals(3, sf.getEnd()); @@ -2080,8 +2090,8 @@ public class AlignmentUtilsTests assertEquals( "p.Pro3His var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6", sf.links.get(0)); - // var5 generates two distinct protein variant features - assertEquals("Jalview", sf.getFeatureGroup()); + assertEquals(cosmic, sf.getFeatureGroup()); + sf = sfs[4]; assertEquals(3, sf.getBegin()); assertEquals(3, sf.getEnd()); @@ -2093,7 +2103,7 @@ public class AlignmentUtilsTests assertEquals( "p.Pro3Arg var6|http://www.ensembl.org/Homo_sapiens/Variation/Summary?v=var6", sf.links.get(0)); - assertEquals("Jalview", sf.getFeatureGroup()); + assertEquals(cosmic, sf.getFeatureGroup()); } /**