X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FAlignmentUtils.java;h=23c5d644433263dd518f19874bb3d08c67e2b74d;hb=8f08ed712138cea8987594a62b6053a0bfdba327;hp=55efaa51c71d83467b20609ef0284ffe96463d34;hpb=9b85e1552fa57f02cf6cd312cfbd7efdfd079ea3;p=jalview.git diff --git a/src/jalview/analysis/AlignmentUtils.java b/src/jalview/analysis/AlignmentUtils.java index 55efaa5..23c5d64 100644 --- a/src/jalview/analysis/AlignmentUtils.java +++ b/src/jalview/analysis/AlignmentUtils.java @@ -20,7 +20,7 @@ */ package jalview.analysis; -import static jalview.io.gff.GffConstants.CLINICAL_SIGNIFICANCE; +import java.util.Locale; import jalview.commands.RemoveGapColCommand; import jalview.datamodel.AlignedCodon; @@ -38,7 +38,6 @@ import jalview.datamodel.SequenceFeature; import jalview.datamodel.SequenceGroup; import jalview.datamodel.SequenceI; import jalview.datamodel.features.SequenceFeatures; -import jalview.io.gff.Gff3Helper; import jalview.io.gff.SequenceOntologyI; import jalview.schemes.ResidueProperties; import jalview.util.Comparison; @@ -46,10 +45,7 @@ import jalview.util.DBRefUtils; import jalview.util.IntRangeComparator; import jalview.util.MapList; import jalview.util.MappingUtils; -import jalview.util.StringUtils; -import java.io.UnsupportedEncodingException; -import java.net.URLEncoder; import java.util.ArrayList; import java.util.Arrays; import java.util.Collection; @@ -187,9 +183,9 @@ public class AlignmentUtils // TODO use Character.toLowerCase to avoid creating String objects? char[] upstream = new String(ds .getSequence(s.getStart() - 1 - ustream_ds, s.getStart() - 1)) - .toLowerCase().toCharArray(); + .toLowerCase(Locale.ROOT).toCharArray(); char[] downstream = new String( - ds.getSequence(s_end - 1, s_end + dstream_ds)).toLowerCase() + ds.getSequence(s_end - 1, s_end + dstream_ds)).toLowerCase(Locale.ROOT) .toCharArray(); char[] coreseq = s.getSequence(); char[] nseq = new char[offset + upstream.length + downstream.length @@ -468,7 +464,7 @@ public class AlignmentUtils if (cdnaLength != mappedLength && cdnaLength > 2) { String lastCodon = String.valueOf(cdnaSeqChars, - cdnaLength - CODON_LENGTH, CODON_LENGTH).toUpperCase(); + cdnaLength - CODON_LENGTH, CODON_LENGTH).toUpperCase(Locale.ROOT); for (String stop : ResidueProperties.STOP_CODONS) { if (lastCodon.equals(stop)) @@ -485,7 +481,7 @@ public class AlignmentUtils */ int startOffset = 0; if (cdnaLength != mappedLength && cdnaLength > 2 - && String.valueOf(cdnaSeqChars, 0, CODON_LENGTH).toUpperCase() + && String.valueOf(cdnaSeqChars, 0, CODON_LENGTH).toUpperCase(Locale.ROOT) .equals(ResidueProperties.START)) { startOffset += CODON_LENGTH; @@ -1603,11 +1599,12 @@ public class AlignmentUtils return false; } String name = seq2.getName(); - final DBRefEntry[] xrefs = seq1.getDBRefs(); + final List xrefs = seq1.getDBRefs(); if (xrefs != null) { - for (DBRefEntry xref : xrefs) + for (int ix = 0, nx = xrefs.size(); ix < nx; ix++) { + DBRefEntry xref = xrefs.get(ix); String xrefName = xref.getSource() + "|" + xref.getAccessionId(); // case-insensitive test, consistent with DBRefEntry.equalRef() if (xrefName.equalsIgnoreCase(name)) @@ -1737,15 +1734,8 @@ public class AlignmentUtils cdsSeqs.add(cdsSeq); - if (!dataset.getSequences().contains(cdsSeqDss)) - { - // check if this sequence is a newly created one - // so needs adding to the dataset - dataset.addSequence(cdsSeqDss); - } - /* - * add a mapping from CDS to the (unchanged) mapped to range + * build the mapping from CDS to protein */ List cdsRange = Collections .singletonList(new int[] @@ -1754,16 +1744,26 @@ public class AlignmentUtils MapList cdsToProteinMap = new MapList(cdsRange, mapList.getToRanges(), mapList.getFromRatio(), mapList.getToRatio()); - AlignedCodonFrame cdsToProteinMapping = new AlignedCodonFrame(); - cdsToProteinMapping.addMap(cdsSeqDss, proteinProduct, - cdsToProteinMap); - /* - * guard against duplicating the mapping if repeating this action - */ - if (!mappings.contains(cdsToProteinMapping)) + if (!dataset.getSequences().contains(cdsSeqDss)) { - mappings.add(cdsToProteinMapping); + /* + * if this sequence is a newly created one, add it to the dataset + * and made a CDS to protein mapping (if sequence already exists, + * CDS-to-protein mapping _is_ the transcript-to-protein mapping) + */ + dataset.addSequence(cdsSeqDss); + AlignedCodonFrame cdsToProteinMapping = new AlignedCodonFrame(); + cdsToProteinMapping.addMap(cdsSeqDss, proteinProduct, + cdsToProteinMap); + + /* + * guard against duplicating the mapping if repeating this action + */ + if (!mappings.contains(cdsToProteinMapping)) + { + mappings.add(cdsToProteinMapping); + } } propagateDBRefsToCDS(cdsSeqDss, dnaSeq.getDatasetSequence(), @@ -1805,8 +1805,10 @@ public class AlignmentUtils // need to // synthesize an xref. - for (DBRefEntry primRef : dnaDss.getPrimaryDBRefs()) + List primrefs = dnaDss.getPrimaryDBRefs(); + for (int ip = 0, np = primrefs.size(); ip < np; ip++) { + DBRefEntry primRef = primrefs.get(ip); /* * create a cross-reference from CDS to the source sequence's * primary reference and vice versa @@ -1820,7 +1822,6 @@ public class AlignmentUtils dnaSeq.addDBRef(new DBRefEntry(source, version, cdsSeq .getName(), new Mapping(cdsSeqDss, dnaToCdsMap))); - // problem here is that the cross-reference is synthesized - // cdsSeq.getName() may be like 'CDS|dnaaccession' or // 'CDS|emblcdsacc' @@ -1833,7 +1834,6 @@ public class AlignmentUtils .getInverse())); proteinProduct.addDBRef(proteinToCdsRef); } - /* * transfer any features on dna that overlap the CDS */ @@ -2099,13 +2099,15 @@ public class AlignmentUtils List direct = new ArrayList<>(); HashSet directSources = new HashSet<>(); - if (contig.getDBRefs() != null) + List refs = contig.getDBRefs(); + if (refs != null) { - for (DBRefEntry dbr : contig.getDBRefs()) + for (int ib = 0, nb = refs.size(); ib < nb; ib++) { - if (dbr.hasMap() && dbr.getMap().getMap().isTripletMap()) + DBRefEntry dbr = refs.get(ib); + MapList map; + if (dbr.hasMap() && (map = dbr.getMap().getMap()).isTripletMap()) { - MapList map = dbr.getMap().getMap(); // check if map is the CDS mapping if (mapping.getMap().equals(map)) { @@ -2115,21 +2117,22 @@ public class AlignmentUtils } } } - DBRefEntry[] onSource = DBRefUtils.selectRefs( + List onSource = DBRefUtils.selectRefs( proteinProduct.getDBRefs(), directSources.toArray(new String[0])); List propagated = new ArrayList<>(); // and generate appropriate mappings - for (DBRefEntry cdsref : direct) + for (int ic = 0, nc = direct.size(); ic < nc; ic++) { + DBRefEntry cdsref = direct.get(ic); + Mapping m = cdsref.getMap(); // clone maplist and mapping MapList cdsposmap = new MapList( Arrays.asList(new int[][] { new int[] { cdsSeq.getStart(), cdsSeq.getEnd() } }), - cdsref.getMap().getMap().getToRanges(), 3, 1); - Mapping cdsmap = new Mapping(cdsref.getMap().getTo(), - cdsref.getMap().getMap()); + m.getMap().getToRanges(), 3, 1); + Mapping cdsmap = new Mapping(m.getTo(), m.getMap()); // create dbref DBRefEntry newref = new DBRefEntry(cdsref.getSource(), @@ -2352,10 +2355,14 @@ public class AlignmentUtils int phase = 0; try { - phase = Integer.parseInt(sf.getPhase()); + String s = sf.getPhase(); + if (s != null) + { + phase = Integer.parseInt(s); + } } catch (NumberFormatException e) { - // ignore + // leave as zero } /* * phase > 0 on first codon means 5' incomplete - skip to the start @@ -2389,97 +2396,6 @@ public class AlignmentUtils } /** - * Helper method that adds a peptide variant feature. ID and - * clinical_significance attributes of the dna variant (if present) are copied - * to the new feature. - * - * @param peptide - * @param peptidePos - * @param residue - * @param var - * @param codon - * the variant codon e.g. aCg - * @param canonical - * the 'normal' codon e.g. aTg - * @return true if a feature was added, else false - */ - static boolean addPeptideVariant(SequenceI peptide, int peptidePos, - String residue, DnaVariant var, String codon, String canonical) - { - /* - * get peptide translation of codon e.g. GAT -> D - * note that variants which are not single alleles, - * e.g. multibase variants or HGMD_MUTATION etc - * are currently ignored here - */ - String trans = codon.contains("-") ? null - : (codon.length() > CODON_LENGTH ? null - : ResidueProperties.codonTranslate(codon)); - if (trans == null) - { - return false; - } - String desc = canonical + "/" + codon; - String featureType = ""; - if (trans.equals(residue)) - { - featureType = SequenceOntologyI.SYNONYMOUS_VARIANT; - } - else if (ResidueProperties.STOP.equals(trans)) - { - featureType = SequenceOntologyI.STOP_GAINED; - } - else - { - String residue3Char = StringUtils - .toSentenceCase(ResidueProperties.aa2Triplet.get(residue)); - String trans3Char = StringUtils - .toSentenceCase(ResidueProperties.aa2Triplet.get(trans)); - desc = "p." + residue3Char + peptidePos + trans3Char; - featureType = SequenceOntologyI.NONSYNONYMOUS_VARIANT; - } - SequenceFeature sf = new SequenceFeature(featureType, desc, peptidePos, - peptidePos, var.getSource()); - - StringBuilder attributes = new StringBuilder(32); - String id = (String) var.variant.getValue(VARIANT_ID); - if (id != null) - { - if (id.startsWith(SEQUENCE_VARIANT)) - { - id = id.substring(SEQUENCE_VARIANT.length()); - } - sf.setValue(VARIANT_ID, id); - attributes.append(VARIANT_ID).append("=").append(id); - // TODO handle other species variants JAL-2064 - StringBuilder link = new StringBuilder(32); - try - { - 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()); - } catch (UnsupportedEncodingException e) - { - // as if - } - } - String clinSig = (String) var.variant.getValue(CLINICAL_SIGNIFICANCE); - if (clinSig != null) - { - sf.setValue(CLINICAL_SIGNIFICANCE, clinSig); - attributes.append(";").append(CLINICAL_SIGNIFICANCE).append("=") - .append(clinSig); - } - peptide.addSequenceFeature(sf); - if (attributes.length() > 0) - { - sf.setAttributes(attributes.toString()); - } - return true; - } - - /** * Makes an alignment with a copy of the given sequences, adding in any * non-redundant sequences which are mapped to by the cross-referenced * sequences. @@ -2499,19 +2415,25 @@ public class AlignmentUtils SequenceIdMatcher matcher = new SequenceIdMatcher(seqs); if (xrefs != null) { - for (SequenceI xref : xrefs) + // BH 2019.01.25 recoded to remove iterators + + for (int ix = 0, nx = xrefs.length; ix < nx; ix++) { - DBRefEntry[] dbrefs = xref.getDBRefs(); + SequenceI xref = xrefs[ix]; + List dbrefs = xref.getDBRefs(); if (dbrefs != null) { - for (DBRefEntry dbref : dbrefs) + for (int ir = 0, nir = dbrefs.size(); ir < nir; ir++) { - if (dbref.getMap() == null || dbref.getMap().getTo() == null - || dbref.getMap().getTo().isProtein() != isProtein) + DBRefEntry dbref = dbrefs.get(ir); + Mapping map = dbref.getMap(); + SequenceI mto; + if (map == null || (mto = map.getTo()) == null + || mto.isProtein() != isProtein) { continue; } - SequenceI mappedTo = dbref.getMap().getTo(); + SequenceI mappedTo = mto; SequenceI match = matcher.findIdMatch(mappedTo); if (match == null) { @@ -2667,6 +2589,13 @@ public class AlignmentUtils { List alignedSequences = alignedDatasets .get(seq.getDatasetSequence()); + if (alignedSequences.isEmpty()) + { + /* + * defensive check - shouldn't happen! (JAL-3536) + */ + continue; + } SequenceI alignedSeq = alignedSequences.get(0); /*