X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FAlignmentUtils.java;h=55efaa51c71d83467b20609ef0284ffe96463d34;hb=9b85e1552fa57f02cf6cd312cfbd7efdfd079ea3;hp=8e0335fee4a161b0414377a99fe60992228f6f12;hpb=933ca77e913c56fc4c71cc0fe534df208bfd742d;p=jalview.git diff --git a/src/jalview/analysis/AlignmentUtils.java b/src/jalview/analysis/AlignmentUtils.java index 8e0335f..55efaa5 100644 --- a/src/jalview/analysis/AlignmentUtils.java +++ b/src/jalview/analysis/AlignmentUtils.java @@ -2389,172 +2389,6 @@ public class AlignmentUtils } /** - * Maps exon features from dna to protein, and computes variants in peptide - * product generated by variants in dna, and adds them as sequence_variant - * features on the protein sequence. Returns the number of variant features - * added. - * - * @param dnaSeq - * @param peptide - * @param dnaToProtein - */ - public static int computeProteinFeatures(SequenceI dnaSeq, - SequenceI peptide, MapList dnaToProtein) - { - while (dnaSeq.getDatasetSequence() != null) - { - dnaSeq = dnaSeq.getDatasetSequence(); - } - while (peptide.getDatasetSequence() != null) - { - peptide = peptide.getDatasetSequence(); - } - - transferFeatures(dnaSeq, peptide, dnaToProtein, SequenceOntologyI.EXON); - - /* - * compute protein variants from dna variants and codon mappings; - * NB - alternatively we could retrieve this using the REST service e.g. - * http://rest.ensembl.org/overlap/translation - * /ENSP00000288602?feature=transcript_variation;content-type=text/xml - * which would be a bit slower but possibly more reliable - */ - - /* - * build a map with codon variations for each potentially varying peptide - */ - LinkedHashMap[]> variants = buildDnaVariantsMap( - dnaSeq, dnaToProtein); - - /* - * scan codon variations, compute peptide variants and add to peptide sequence - */ - int count = 0; - for (Entry[]> variant : variants.entrySet()) - { - int peptidePos = variant.getKey(); - List[] codonVariants = variant.getValue(); - count += computePeptideVariants(peptide, peptidePos, codonVariants); - } - - return count; - } - - /** - * Computes non-synonymous peptide variants from codon variants and adds them as - * sequence_variant features on the protein sequence (one feature per allele - * variant). Selected attributes (variant id, clinical significance) are copied - * over to the new features. - * - * @param peptide - * the protein dataset (ungapped) sequence - * @param peptidePos - * the position to compute peptide variants for - * @param codonVariants - * a list of dna variants per codon position - * @return the number of features added - */ - static int computePeptideVariants(SequenceI peptide, int peptidePos, - List[] codonVariants) - { - String residue = String - .valueOf(peptide.getCharAt(peptidePos - peptide.getStart())); - int count = 0; - String base1 = codonVariants[0].get(0).base; - String base2 = codonVariants[1].get(0).base; - String base3 = codonVariants[2].get(0).base; - - /* - * variants in first codon base - */ - for (DnaVariant var : codonVariants[0]) - { - if (var.variant != null) - { - String alleles = (String) var.variant.getValue(Gff3Helper.ALLELES); - if (alleles != null) - { - for (String base : alleles.split(",")) - { - if (!base1.equalsIgnoreCase(base)) - { - 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++; - } - } - } - } - } - } - - /* - * variants in second codon base - */ - for (DnaVariant var : codonVariants[1]) - { - if (var.variant != null) - { - String alleles = (String) var.variant.getValue(Gff3Helper.ALLELES); - if (alleles != null) - { - for (String base : alleles.split(",")) - { - if (!base2.equalsIgnoreCase(base)) - { - 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++; - } - } - } - } - } - } - - /* - * variants in third codon base - */ - for (DnaVariant var : codonVariants[2]) - { - if (var.variant != null) - { - String alleles = (String) var.variant.getValue(Gff3Helper.ALLELES); - if (alleles != null) - { - for (String base : alleles.split(",")) - { - if (!base3.equalsIgnoreCase(base)) - { - 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++; - } - } - } - } - } - } - - return count; - } - - /** * 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. @@ -2646,137 +2480,6 @@ public class AlignmentUtils } /** - * Builds a map whose key is position in the protein sequence, and value is a - * list of the base and all variants for each corresponding codon position. - *

- * This depends on dna variants being held as a comma-separated list as - * property "alleles" on variant features. - * - * @param dnaSeq - * @param dnaToProtein - * @return - */ - @SuppressWarnings("unchecked") - static LinkedHashMap[]> buildDnaVariantsMap( - SequenceI dnaSeq, MapList dnaToProtein) - { - /* - * map from peptide position to all variants of the codon which codes for it - * LinkedHashMap ensures we keep the peptide features in sequence order - */ - LinkedHashMap[]> variants = new LinkedHashMap<>(); - - List dnaFeatures = dnaSeq.getFeatures() - .getFeaturesByOntology(SequenceOntologyI.SEQUENCE_VARIANT); - if (dnaFeatures.isEmpty()) - { - return variants; - } - - int dnaStart = dnaSeq.getStart(); - int[] lastCodon = null; - int lastPeptidePostion = 0; - - /* - * build a map of codon variations for peptides - */ - for (SequenceFeature sf : dnaFeatures) - { - int dnaCol = sf.getBegin(); - if (dnaCol != sf.getEnd()) - { - // not handling multi-locus variant features - continue; - } - - /* - * ignore variant if not a SNP - */ - String alls = (String) sf.getValue(Gff3Helper.ALLELES); - if (alls == null) - { - continue; // non-SNP VCF variant perhaps - can't process this - } - - String[] alleles = alls.toUpperCase().split(","); - boolean isSnp = true; - for (String allele : alleles) - { - if (allele.trim().length() > 1) - { - isSnp = false; - } - } - if (!isSnp) - { - continue; - } - - int[] mapsTo = dnaToProtein.locateInTo(dnaCol, dnaCol); - if (mapsTo == null) - { - // feature doesn't lie within coding region - continue; - } - int peptidePosition = mapsTo[0]; - List[] codonVariants = variants.get(peptidePosition); - if (codonVariants == null) - { - codonVariants = new ArrayList[CODON_LENGTH]; - codonVariants[0] = new ArrayList<>(); - codonVariants[1] = new ArrayList<>(); - codonVariants[2] = new ArrayList<>(); - variants.put(peptidePosition, codonVariants); - } - - /* - * get this peptide's codon positions e.g. [3, 4, 5] or [4, 7, 10] - */ - int[] codon = peptidePosition == lastPeptidePostion ? lastCodon - : MappingUtils.flattenRanges(dnaToProtein.locateInFrom( - peptidePosition, peptidePosition)); - lastPeptidePostion = peptidePosition; - lastCodon = codon; - - /* - * save nucleotide (and any variant) for each codon position - */ - for (int codonPos = 0; codonPos < CODON_LENGTH; codonPos++) - { - String nucleotide = String.valueOf( - dnaSeq.getCharAt(codon[codonPos] - dnaStart)).toUpperCase(); - List codonVariant = codonVariants[codonPos]; - if (codon[codonPos] == dnaCol) - { - if (!codonVariant.isEmpty() - && codonVariant.get(0).variant == null) - { - /* - * already recorded base value, add this variant - */ - codonVariant.get(0).variant = sf; - } - else - { - /* - * add variant with base value - */ - codonVariant.add(new DnaVariant(nucleotide, sf)); - } - } - else if (codonVariant.isEmpty()) - { - /* - * record (possibly non-varying) base value - */ - codonVariant.add(new DnaVariant(nucleotide)); - } - } - } - return variants; - } - - /** * 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.