{
String lastCodon = String.valueOf(cdnaSeqChars,
cdnaLength - CODON_LENGTH, CODON_LENGTH).toUpperCase();
- for (String stop : ResidueProperties.STOP)
+ for (String stop : ResidueProperties.STOP_CODONS)
{
if (lastCodon.equals(stop))
{
* allow * in protein to match untranslatable in dna
*/
final char aaRes = aaSeqChars[aaPos];
- if ((translated == null || "STOP".equals(translated)) && aaRes == '*')
+ if ((translated == null || ResidueProperties.STOP.equals(translated))
+ && aaRes == '*')
{
continue;
}
if (dnaPos == cdnaSeqChars.length - CODON_LENGTH)
{
String codon = String.valueOf(cdnaSeqChars, dnaPos, CODON_LENGTH);
- if ("STOP".equals(ResidueProperties.codonTranslate(codon)))
+ if (ResidueProperties.STOP
+ .equals(ResidueProperties.codonTranslate(codon)))
{
return true;
}
return false;
}
String name = seq2.getName();
- final DBRefEntry[] xrefs = seq1.getDBRefs();
+ final List<DBRefEntry> 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))
// need to
// synthesize an xref.
- for (DBRefEntry primRef : dnaDss.getPrimaryDBRefs())
+ List<DBRefEntry> 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
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'
.getInverse()));
proteinProduct.addDBRef(proteinToCdsRef);
}
-
/*
* transfer any features on dna that overlap the CDS
*/
* @param seqMappings
* the set of mappings involving dnaSeq
* @param aMapping
- * an initial candidate from seqMappings
+ * a transcript-to-peptide mapping
* @return
*/
static SequenceI findCdsForProtein(List<AlignedCodonFrame> mappings,
if (mappedFromLength == dnaLength
|| mappedFromLength == dnaLength - CODON_LENGTH)
{
- return seqDss;
+ /*
+ * if sequence has CDS features, this is a transcript with no UTR
+ * - do not take this as the CDS sequence! (JAL-2789)
+ */
+ if (seqDss.getFeatures().getFeaturesByOntology(SequenceOntologyI.CDS)
+ .isEmpty())
+ {
+ return seqDss;
+ }
}
/*
{
/*
* found a 3:1 mapping to the protein product which covers
- * the whole dna sequence i.e. is from CDS; finally check it
- * is from the dna start sequence
+ * the whole dna sequence i.e. is from CDS; finally check the CDS
+ * is mapped from the given dna start sequence
*/
SequenceI cdsSeq = map.getFromSeq();
+ // todo this test is weak if seqMappings contains multiple mappings;
+ // we get away with it if transcript:cds relationship is 1:1
List<AlignedCodonFrame> dnaToCdsMaps = MappingUtils
.findMappingsForSequence(cdsSeq, seqMappings);
if (!dnaToCdsMaps.isEmpty())
protected static List<DBRefEntry> propagateDBRefsToCDS(SequenceI cdsSeq,
SequenceI contig, SequenceI proteinProduct, Mapping mapping)
{
+
// gather direct refs from contig congruent with mapping
List<DBRefEntry> direct = new ArrayList<>();
HashSet<String> directSources = new HashSet<>();
- if (contig.getDBRefs() != null)
+
+ List<DBRefEntry> 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))
{
}
}
}
- DBRefEntry[] onSource = DBRefUtils.selectRefs(
+ List<DBRefEntry> onSource = DBRefUtils.selectRefs(
proteinProduct.getDBRefs(),
directSources.toArray(new String[0]));
List<DBRefEntry> 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(),
int mappedDnaLength = MappingUtils.getLength(ranges);
/*
- * if not a whole number of codons, something is wrong,
- * abort mapping
+ * if not a whole number of codons, truncate mapping
*/
- if (mappedDnaLength % CODON_LENGTH > 0)
+ int codonRemainder = mappedDnaLength % CODON_LENGTH;
+ if (codonRemainder > 0)
{
- return null;
+ mappedDnaLength -= codonRemainder;
+ MappingUtils.removeEndPositions(codonRemainder, ranges);
}
int proteinLength = proteinSeq.getLength();
int phase = 0;
try
{
- phase = Integer.parseInt(sf.getPhase());
+ String s = sf.getPhase();
+ if (s != null)
+ {
+ phase = Integer.parseInt(s);
+ }
} catch (NumberFormatException e)
{
- // ignore
+ // SwingJS -- need to avoid these.
}
/*
* phase > 0 on first codon means 5' incomplete - skip to the start
/*
* variants in first codon base
*/
- for (DnaVariant var : codonVariants[0])
+ for (DnaVariant dnavar : codonVariants[0])
{
- if (var.variant != null)
+ if (dnavar.variant != null)
{
- String alleles = (String) var.variant.getValue(Gff3Helper.ALLELES);
+ String alleles = (String) dnavar.variant.getValue(Gff3Helper.ALLELES);
if (alleles != null)
{
for (String base : alleles.split(","))
{
- String codon = base + base2 + base3;
- if (addPeptideVariant(peptide, peptidePos, residue, var, codon))
+ if (!base1.equalsIgnoreCase(base))
{
- count++;
+ String codon = base.toUpperCase() + base2.toLowerCase()
+ + base3.toLowerCase();
+ String canonical = base1.toUpperCase() + base2.toLowerCase()
+ + base3.toLowerCase();
+ if (addPeptideVariant(peptide, peptidePos, residue, dnavar,
+ codon, canonical))
+ {
+ count++;
+ }
}
}
}
{
for (String base : alleles.split(","))
{
- String codon = base1 + base + base3;
- if (addPeptideVariant(peptide, peptidePos, residue, var, codon))
+ if (!base2.equalsIgnoreCase(base))
{
- count++;
+ String codon = base1.toLowerCase() + base.toUpperCase()
+ + base3.toLowerCase();
+ String canonical = base1.toLowerCase() + base2.toUpperCase()
+ + base3.toLowerCase();
+ if (addPeptideVariant(peptide, peptidePos, residue, var,
+ codon, canonical))
+ {
+ count++;
+ }
}
}
}
{
for (String base : alleles.split(","))
{
- String codon = base1 + base2 + base;
- if (addPeptideVariant(peptide, peptidePos, residue, var, codon))
+ if (!base3.equalsIgnoreCase(base))
{
- count++;
+ String codon = base1.toLowerCase() + base2.toLowerCase()
+ + base.toUpperCase();
+ String canonical = base1.toLowerCase() + base2.toLowerCase()
+ + base3.toUpperCase();
+ if (addPeptideVariant(peptide, peptidePos, residue, var,
+ codon, canonical))
+ {
+ count++;
+ }
}
}
}
}
/**
- * Helper method that adds a peptide variant feature, provided the given codon
- * translates to a value different to the current residue (is a non-synonymous
- * variant). ID and clinical_significance attributes of the dna variant (if
- * present) are copied to the new feature.
+ * 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 residue, DnaVariant var, String codon, String canonical)
{
/*
* get peptide translation of codon e.g. GAT -> D
* e.g. multibase variants or HGMD_MUTATION etc
* are currently ignored here
*/
- String trans = codon.contains("-") ? "-"
+ String trans = codon.contains("-") ? null
: (codon.length() > CODON_LENGTH ? null
: ResidueProperties.codonTranslate(codon));
- if (trans != null && !trans.equals(residue))
+ 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));
- String desc = "p." + residue3Char + peptidePos + trans3Char;
- SequenceFeature sf = new SequenceFeature(
- SequenceOntologyI.SEQUENCE_VARIANT, desc, peptidePos,
- peptidePos, var.getSource());
- StringBuilder attributes = new StringBuilder(32);
- String id = (String) var.variant.getValue(ID);
- if (id != null)
- {
- if (id.startsWith(SEQUENCE_VARIANT))
- {
- id = id.substring(SEQUENCE_VARIANT.length());
- }
- sf.setValue(ID, id);
- attributes.append(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)
+ 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(ID);
+ if (id != null)
+ {
+ if (id.startsWith(SEQUENCE_VARIANT))
{
- sf.setValue(CLINICAL_SIGNIFICANCE, clinSig);
- attributes.append(";").append(CLINICAL_SIGNIFICANCE).append("=")
- .append(clinSig);
+ id = id.substring(SEQUENCE_VARIANT.length());
}
- peptide.addSequenceFeature(sf);
- if (attributes.length() > 0)
+ sf.setValue(ID, id);
+ attributes.append(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)
{
- sf.setAttributes(attributes.toString());
+ // as if
}
- return true;
}
- return false;
+ 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;
}
/**
SequenceIdMatcher matcher = new SequenceIdMatcher(seqs);
if (xrefs != null)
{
- for (SequenceI xref : xrefs)
+ // BH 2019.01.25 streamlined this triply nested loop to remove all iterators
+
+ for (int ix = 0, nx = xrefs.length; ix < nx; ix++)
{
- DBRefEntry[] dbrefs = xref.getDBRefs();
+ SequenceI xref = xrefs[ix];
+ List<DBRefEntry> 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)
{