*/
public class AlignmentUtils
{
-
private static final int CODON_LENGTH = 3;
private static final String SEQUENCE_VARIANT = "sequence_variant:";
- private static final String ID = "ID";
+ /*
+ * the 'id' attribute is provided for variant features fetched from
+ * Ensembl using its REST service with JSON format
+ */
+ public static final String VARIANT_ID = "id";
/**
* A data model to hold the 'normal' base value at a position, and an optional
}
/**
- * Returns the index (zero-based position) of a sequence in an alignment, or
- * -1 if not found.
- *
- * @param al
- * @param seq
- * @return
- */
- public static int getSequenceIndex(AlignmentI al, SequenceI seq)
- {
- int result = -1;
- int pos = 0;
- for (SequenceI alSeq : al.getSequences())
- {
- if (alSeq == seq)
- {
- result = pos;
- break;
- }
- pos++;
- }
- return result;
- }
-
- /**
* Returns a map of lists of sequences in the alignment, keyed by sequence
* name. For use in mapping between different alignment views of the same
* sequences.
{
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;
}
* @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)
{
for (DBRefEntry dbr : contig.getDBRefs())
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();
{
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, var,
+ 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(VARIANT_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(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)
{
- 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;
}
/**