X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fext%2Fensembl%2FEnsemblGene.java;h=63a6a6ca7a0589f44034d89dfd5508087128686e;hb=990ed4ffbaa7a95b2ebb6bf6ab0440310f6e83ab;hp=cdcfa9642c5f3f57e3a4a0ff976b7904fc624955;hpb=14193747f3831242bc7dfac12394eb20eb0ba480;p=jalview.git diff --git a/src/jalview/ext/ensembl/EnsemblGene.java b/src/jalview/ext/ensembl/EnsemblGene.java index cdcfa96..63a6a6c 100644 --- a/src/jalview/ext/ensembl/EnsemblGene.java +++ b/src/jalview/ext/ensembl/EnsemblGene.java @@ -34,6 +34,7 @@ import jalview.io.gff.SequenceOntologyI; import jalview.schemes.FeatureColour; import jalview.schemes.FeatureSettingsAdapter; import jalview.util.MapList; +import jalview.util.Platform; import java.awt.Color; import java.io.UnsupportedEncodingException; @@ -51,13 +52,11 @@ import com.stevesoft.pat.Regex; */ public class EnsemblGene extends EnsemblSeqProxy { - private static final String GENE_PREFIX = "gene:"; - /* * accepts anything as we will attempt lookup of gene or * transcript id or gene name */ - private static final Regex ACCESSION_REGEX = new Regex(".*"); + private static Regex ACCESSION_REGEX; private static final EnsemblFeatureType[] FEATURES_TO_FETCH = { EnsemblFeatureType.gene, EnsemblFeatureType.transcript, @@ -100,6 +99,12 @@ public class EnsemblGene extends EnsemblSeqProxy return EnsemblSeqType.GENOMIC; } + @Override + protected String getObjectType() + { + return OBJECT_TYPE_GENE; + } + /** * Returns an alignment containing the gene(s) for the given gene or * transcript identifier, or external identifier (e.g. Uniprot id). If given a @@ -134,7 +139,6 @@ public class EnsemblGene extends EnsemblSeqProxy * convert to a non-duplicated list of gene identifiers */ List geneIds = getGeneIds(query); - AlignmentI al = null; for (String geneId : geneIds) { @@ -149,6 +153,8 @@ public class EnsemblGene extends EnsemblSeqProxy if (geneAlignment.getHeight() == 1) { + // ensure id has 'correct' case for the Ensembl identifier + geneId = geneAlignment.getSequenceAt(0).getName(); findGeneLoci(geneAlignment.getSequenceAt(0), geneId); getTranscripts(geneAlignment, geneId); } @@ -235,7 +241,7 @@ public class EnsemblGene extends EnsemblSeqProxy */ List getGeneIds(String accessions) { - List geneIds = new ArrayList(); + List geneIds = new ArrayList<>(); for (String acc : accessions.split(getAccessionSeparator())) { @@ -358,7 +364,7 @@ public class EnsemblGene extends EnsemblSeqProxy * look for exon features of the transcript, failing that for CDS * (for example ENSG00000124610 has 1 CDS but no exon features) */ - String parentId = "transcript:" + accId; + String parentId = accId; List splices = findFeatures(gene, SequenceOntologyI.EXON, parentId); if (splices.isEmpty()) @@ -370,7 +376,7 @@ public class EnsemblGene extends EnsemblSeqProxy int transcriptLength = 0; final char[] geneChars = gene.getSequence(); int offset = gene.getStart(); // to convert to 0-based positions - List mappedFrom = new ArrayList(); + List mappedFrom = new ArrayList<>(); for (SequenceFeature sf : splices) { @@ -389,7 +395,7 @@ public class EnsemblGene extends EnsemblSeqProxy * Ensembl has gene name as transcript Name * EnsemblGenomes doesn't, but has a url-encoded description field */ - String description = (String) transcriptFeature.getValue(NAME); + String description = transcriptFeature.getDescription(); if (description == null) { description = (String) transcriptFeature.getValue(DESCRIPTION); @@ -412,7 +418,7 @@ public class EnsemblGene extends EnsemblSeqProxy * transfer features to the new sequence; we use EnsemblCdna to do this, * to filter out unwanted features types (see method retainFeature) */ - List mapTo = new ArrayList(); + List mapTo = new ArrayList<>(); mapTo.add(new int[] { 1, transcriptLength }); MapList mapping = new MapList(mappedFrom, mapTo, 1, 1); EnsemblCdna cdna = new EnsemblCdna(getDomain()); @@ -478,7 +484,7 @@ public class EnsemblGene extends EnsemblSeqProxy */ protected String getTranscriptId(SequenceFeature feature) { - return (String) feature.getValue("transcript_id"); + return (String) feature.getValue(JSON_ID); } /** @@ -498,9 +504,9 @@ public class EnsemblGene extends EnsemblSeqProxy protected List getTranscriptFeatures(String accId, SequenceI geneSequence) { - List transcriptFeatures = new ArrayList(); + List transcriptFeatures = new ArrayList<>(); - String parentIdentifier = GENE_PREFIX + accId; + String parentIdentifier = accId; List sfs = geneSequence.getFeatures() .getFeaturesByOntology(SequenceOntologyI.TRANSCRIPT); @@ -510,7 +516,7 @@ public class EnsemblGene extends EnsemblSeqProxy for (SequenceFeature sf : sfs) { String parent = (String) sf.getValue(PARENT); - if (parentIdentifier.equals(parent)) + if (parentIdentifier.equalsIgnoreCase(parent)) { transcriptFeatures.add(sf); } @@ -531,29 +537,34 @@ public class EnsemblGene extends EnsemblSeqProxy @Override public String getTestQuery() { - return "ENSG00000157764"; // BRAF, 5 transcripts, reverse strand + return Platform.isJS() ? "ENSG00000123569" : "ENSG00000157764"; + // ENSG00000123569 // H2BFWT histone, 2 transcripts, reverse strand + // ENSG00000157764 // BRAF, 5 transcripts, reverse strand // ENSG00000090266 // NDUFB2, 15 transcripts, forward strand // ENSG00000101812 // H2BFM histone, 3 transcripts, forward strand - // ENSG00000123569 // H2BFWT histone, 2 transcripts, reverse strand } /** - * Answers true for a feature of type 'gene' (or a sub-type of gene in the - * Sequence Ontology), whose ID is the accession we are retrieving + * Answers a list of sequence features (if any) whose type is 'gene' (or a + * subtype of gene in the Sequence Ontology), and whose ID is the accession we + * are retrieving */ @Override - protected boolean identifiesSequence(SequenceFeature sf, String accId) + protected List getIdentifyingFeatures(SequenceI seq, + String accId) { - if (SequenceOntologyFactory.getInstance().isA(sf.getType(), - SequenceOntologyI.GENE)) + List result = new ArrayList<>(); + List sfs = seq.getFeatures() + .getFeaturesByOntology(SequenceOntologyI.GENE); + for (SequenceFeature sf : sfs) { - String id = (String) sf.getValue(ID); - if ((GENE_PREFIX + accId).equals(id)) + String id = (String) sf.getValue(JSON_ID); + if (accId.equalsIgnoreCase(id)) { - return true; + result.add(sf); } } - return false; + return result; } /** @@ -566,7 +577,7 @@ public class EnsemblGene extends EnsemblSeqProxy @Override protected boolean retainFeature(SequenceFeature sf, String accessionId) { - SequenceOntologyI so = SequenceOntologyFactory.getInstance(); + SequenceOntologyI so = SequenceOntologyFactory.getSequenceOntology(); String type = sf.getType(); if (so.isA(type, SequenceOntologyI.GENE)) { @@ -575,7 +586,7 @@ public class EnsemblGene extends EnsemblSeqProxy if (isTranscript(type)) { String parent = (String) sf.getValue(PARENT); - if (!(GENE_PREFIX + accessionId).equals(parent)) + if (!accessionId.equalsIgnoreCase(parent)) { return false; } @@ -584,17 +595,6 @@ public class EnsemblGene extends EnsemblSeqProxy } /** - * Answers false. This allows an optimisation - a single 'gene' feature is all - * that is needed to identify the positions of the gene on the genomic - * sequence. - */ - @Override - protected boolean isSpliceable() - { - return false; - } - - /** * Override to do nothing as Ensembl doesn't return a protein sequence for a * gene identifier */ @@ -606,6 +606,10 @@ public class EnsemblGene extends EnsemblSeqProxy @Override public Regex getAccessionValidator() { + if (ACCESSION_REGEX == null) + { + ACCESSION_REGEX = Platform.newRegex(".*", null); + } return ACCESSION_REGEX; } @@ -624,7 +628,7 @@ public class EnsemblGene extends EnsemblSeqProxy { return new FeatureSettingsAdapter() { - SequenceOntologyI so = SequenceOntologyFactory.getInstance(); + SequenceOntologyI so = SequenceOntologyFactory.getSequenceOntology(); @Override public boolean isFeatureDisplayed(String type)