+/*
+ * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
+ * Copyright (C) $$Year-Rel$$ The Jalview Authors
+ *
+ * This file is part of Jalview.
+ *
+ * Jalview is free software: you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation, either version 3
+ * of the License, or (at your option) any later version.
+ *
+ * Jalview is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ * PURPOSE. See the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
+ * The Jalview Authors are detailed in the 'AUTHORS' file.
+ */
package jalview.ext.ensembl;
+import jalview.api.FeatureColourI;
+import jalview.api.FeatureSettingsModelI;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceI;
+import jalview.datamodel.features.SequenceFeatures;
import jalview.io.gff.SequenceOntologyFactory;
import jalview.io.gff.SequenceOntologyI;
+import jalview.schemes.FeatureColour;
+import jalview.schemes.FeatureSettingsAdapter;
import jalview.util.MapList;
-import java.io.IOException;
+import java.awt.Color;
+import java.io.UnsupportedEncodingException;
+import java.net.URLDecoder;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
{
private static final String GENE_PREFIX = "gene:";
- // TODO modify to accept other species e.g. ENSMUSGnnn
- private static final Regex ACCESSION_REGEX = new Regex(
- "(ENSG|ENST)[0-9]{11}$");
+ /*
+ * 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 final EnsemblFeatureType[] FEATURES_TO_FETCH = {
EnsemblFeatureType.gene, EnsemblFeatureType.transcript,
EnsemblFeatureType.exon, EnsemblFeatureType.cds,
EnsemblFeatureType.variation };
+ /**
+ * Default constructor (to use rest.ensembl.org)
+ */
+ public EnsemblGene()
+ {
+ super();
+ }
+
+ /**
+ * Constructor given the target domain to fetch data from
+ *
+ * @param d
+ */
+ public EnsemblGene(String d)
+ {
+ super(d);
+ }
+
@Override
public String getDbName()
{
- return "ENSEMBL (GENE)";
+ return "ENSEMBL";
}
@Override
return EnsemblSeqType.GENOMIC;
}
+ @Override
+ protected String getObjectType()
+ {
+ return OBJECT_TYPE_GENE;
+ }
+
/**
- * Builds an alignment of all transcripts for the requested 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
+ * gene name or external identifier, returns any related gene sequences found
+ * for model organisms. If only a single gene is queried for, then its
+ * transcripts are also retrieved and added to the alignment. <br>
+ * Method:
* <ul>
+ * <li>resolves a transcript identifier by looking up its parent gene id</li>
+ * <li>resolves an external identifier by looking up xref-ed gene ids</li>
* <li>fetches the gene sequence</li>
* <li>fetches features on the sequence</li>
- * <li>identifies "transcript" features whose Parent is the requested gene</li>
+ * <li>identifies "transcript" features whose Parent is the requested
+ * gene</li>
* <li>fetches the transcript sequence for each transcript</li>
* <li>makes a mapping from the gene to each transcript</li>
* <li>copies features from gene to transcript sequences</li>
* <li>aligns each transcript against the gene sequence based on the position
* mappings</li>
* </ul>
+ *
+ * @param query
+ * a single gene or transcript identifier or gene name
+ * @return an alignment containing a gene, and possibly transcripts, or null
*/
@Override
public AlignmentI getSequenceRecords(String query) throws Exception
{
- List<String> transcriptsWanted = null;
+ /*
+ * convert to a non-duplicated list of gene identifiers
+ */
+ List<String> geneIds = getGeneIds(query);
- if (isTranscriptIdentifier(query))
+ AlignmentI al = null;
+ for (String geneId : geneIds)
{
- transcriptsWanted = Arrays.asList(query
- .split(getAccessionSeparator()));
- query = getGeneForTranscript(query);
- if (query == null)
+ /*
+ * fetch the gene sequence(s) with features and xrefs
+ */
+ AlignmentI geneAlignment = super.getSequenceRecords(geneId);
+ if (geneAlignment == null)
{
- return null;
+ continue;
+ }
+ if (geneAlignment.getHeight() == 1)
+ {
+ // ensure id has 'correct' case for the Ensembl identifier
+ geneId = geneAlignment.getSequenceAt(0).getName();
+ getTranscripts(geneAlignment, geneId);
+ }
+ if (al == null)
+ {
+ al = geneAlignment;
+ }
+ else
+ {
+ al.append(geneAlignment);
}
}
-
- AlignmentI al = super.getSequenceRecords(query);
- if (al.getHeight() > 0)
- {
- getTranscripts(al, query, transcriptsWanted);
- }
-
return al;
}
/**
- * Gets the parent gene identifier for a given transcript identifier, by
- * retrieving 'transcript' features overlapping the transcript, and finding
- * the Parent property of the feature whose id is the given identifier.
+ * Converts a query, which may contain one or more gene, transcript, or
+ * external (to Ensembl) identifiers, into a non-redundant list of gene
+ * identifiers.
*
- * @param query
+ * @param accessions
* @return
*/
- protected String getGeneForTranscript(String transcriptId)
+ List<String> getGeneIds(String accessions)
{
- String geneId = null;
+ List<String> geneIds = new ArrayList<>();
- /*
- * reduce multiple transcripts (e.g. from Uniprot x-ref) to the first
- * one only as representative (they should all have the same gene)
- */
- transcriptId = transcriptId.split(getAccessionSeparator())[0];
-
- try
+ for (String acc : accessions.split(getAccessionSeparator()))
{
- EnsemblFeatureType[] geneFeature = new EnsemblFeatureType[] { EnsemblFeatureType.transcript };
- AlignmentI al = new EnsemblFeatures().getSequenceRecords(
- transcriptId, geneFeature);
- if (al != null && al.getHeight() > 0)
+ /*
+ * First try lookup as an Ensembl (gene or transcript) identifier
+ */
+ String geneId = new EnsemblLookup(getDomain()).getGeneId(acc);
+ if (geneId != null)
+ {
+ if (!geneIds.contains(geneId))
+ {
+ geneIds.add(geneId);
+ }
+ }
+ else
{
- SequenceFeature[] sfs = al.getSequenceAt(0).getSequenceFeatures();
- if (sfs != null)
+ /*
+ * if given a gene or other external name, lookup and fetch
+ * the corresponding gene for all model organisms
+ */
+ List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
+ getDbVersion()).getGeneIds(acc);
+ for (String id : ids)
{
- for (SequenceFeature sf : sfs)
+ if (!geneIds.contains(id))
{
- if (transcriptId.equals(getTranscriptId(sf)))
- {
- String parent = (String) sf.getValue(PARENT);
- if (parent != null && parent.startsWith(GENE_PREFIX))
- {
- geneId = parent.substring(5);
- }
- break;
- }
+ geneIds.add(id);
}
}
}
- return geneId;
- } catch (IOException e)
- {
- System.err.println("Error retrieving gene id for " + transcriptId
- + ": " + e.getMessage());
- return null;
}
+ return geneIds;
}
/**
*
* @param al
* @param accId
- * @param transcriptsWanted
- * optional list of transcript ids to filter by
* @throws Exception
*/
- protected void getTranscripts(AlignmentI al, String accId,
- List<String> transcriptsWanted)
+ protected void getTranscripts(AlignmentI al, String accId)
throws Exception
{
SequenceI gene = al.getSequenceAt(0);
List<SequenceFeature> transcriptFeatures = getTranscriptFeatures(accId,
- gene, transcriptsWanted);
+ gene);
for (SequenceFeature transcriptFeature : transcriptFeatures)
{
makeTranscript(transcriptFeature, al, gene);
}
+
+ clearGeneFeatures(gene);
+ }
+
+ /**
+ * Remove unwanted features (transcript, exon, CDS) from the gene sequence
+ * after we have used them to derive transcripts and transfer features
+ *
+ * @param gene
+ */
+ protected void clearGeneFeatures(SequenceI gene)
+ {
+ /*
+ * Note we include NMD_transcript_variant here because it behaves like
+ * 'transcript' in Ensembl, although strictly speaking it is not
+ * (it is a sub-type of sequence_variant)
+ */
+ String[] soTerms = new String[] {
+ SequenceOntologyI.NMD_TRANSCRIPT_VARIANT,
+ SequenceOntologyI.TRANSCRIPT, SequenceOntologyI.EXON,
+ SequenceOntologyI.CDS };
+ List<SequenceFeature> sfs = gene.getFeatures().getFeaturesByOntology(
+ soTerms);
+ for (SequenceFeature sf : sfs)
+ {
+ gene.deleteFeature(sf);
+ }
}
/**
* the parent gene sequence, with features
* @return
*/
- SequenceI makeTranscript(SequenceFeature transcriptFeature,
- AlignmentI al, SequenceI gene)
+ SequenceI makeTranscript(SequenceFeature transcriptFeature, AlignmentI al,
+ SequenceI gene)
{
String accId = getTranscriptId(transcriptFeature);
if (accId == null)
{
splices = findFeatures(gene, SequenceOntologyI.CDS, parentId);
}
+ SequenceFeatures.sortFeatures(splices, true);
int transcriptLength = 0;
final char[] geneChars = gene.getSequence();
int offset = gene.getStart(); // to convert to 0-based positions
- List<int[]> mappedFrom = new ArrayList<int[]>();
+ List<int[]> mappedFrom = new ArrayList<>();
for (SequenceFeature sf : splices)
{
mappedFrom.add(new int[] { sf.getBegin(), sf.getEnd() });
}
- Sequence transcript = new Sequence(accId, seqChars, 1, transcriptLength);
- String geneName = (String) transcriptFeature.getValue(NAME);
- if (geneName != null)
+ Sequence transcript = new Sequence(accId, seqChars, 1,
+ transcriptLength);
+
+ /*
+ * Ensembl has gene name as transcript Name
+ * EnsemblGenomes doesn't, but has a url-encoded description field
+ */
+ String description = (String) transcriptFeature.getValue(NAME);
+ if (description == null)
{
- transcript.setDescription(geneName);
+ description = (String) transcriptFeature.getValue(DESCRIPTION);
+ }
+ if (description != null)
+ {
+ try
+ {
+ transcript.setDescription(URLDecoder.decode(description, "UTF-8"));
+ } catch (UnsupportedEncodingException e)
+ {
+ e.printStackTrace(); // as if
+ }
}
transcript.createDatasetSequence();
* transfer features to the new sequence; we use EnsemblCdna to do this,
* to filter out unwanted features types (see method retainFeature)
*/
- List<int[]> mapTo = new ArrayList<int[]>();
+ List<int[]> mapTo = new ArrayList<>();
mapTo.add(new int[] { 1, transcriptLength });
MapList mapping = new MapList(mappedFrom, mapTo, 1, 1);
- new EnsemblCdna().transferFeatures(gene.getSequenceFeatures(),
+ EnsemblCdna cdna = new EnsemblCdna(getDomain());
+ cdna.transferFeatures(gene.getFeatures().getPositionalFeatures(),
transcript.getDatasetSequence(), mapping, parentId);
/*
+ * fetch and save cross-references
+ */
+ cdna.getCrossReferences(transcript);
+
+ /*
* and finally fetch the protein product and save as a cross-reference
*/
- new EnsemblCdna().addProteinProduct(transcript);
+ cdna.addProteinProduct(transcript);
return transcript;
}
/**
* Returns a list of the transcript features on the sequence whose Parent is
* the gene for the accession id.
+ * <p>
+ * Transcript features are those of type "transcript", or any of its sub-types
+ * in the Sequence Ontology e.g. "mRNA", "processed_transcript". We also
+ * include "NMD_transcript_variant", because this type behaves like a
+ * transcript identifier in Ensembl, although strictly speaking it is not in
+ * the SO.
*
* @param accId
* @param geneSequence
- * @param transcriptsWanted
- * optional list of ids to filter on
* @return
*/
protected List<SequenceFeature> getTranscriptFeatures(String accId,
- SequenceI geneSequence, List<String> transcriptsWanted)
+ SequenceI geneSequence)
{
- List<SequenceFeature> transcriptFeatures = new ArrayList<SequenceFeature>();
+ List<SequenceFeature> transcriptFeatures = new ArrayList<>();
String parentIdentifier = GENE_PREFIX + accId;
- SequenceFeature[] sfs = geneSequence.getSequenceFeatures();
- if (sfs != null)
+ List<SequenceFeature> sfs = geneSequence.getFeatures()
+ .getFeaturesByOntology(SequenceOntologyI.TRANSCRIPT);
+ sfs.addAll(geneSequence.getFeatures().getPositionalFeatures(
+ SequenceOntologyI.NMD_TRANSCRIPT_VARIANT));
+
+ for (SequenceFeature sf : sfs)
{
- for (SequenceFeature sf : sfs)
+ String parent = (String) sf.getValue(PARENT);
+ if (parentIdentifier.equalsIgnoreCase(parent))
{
- if (isTranscript(sf.getType()))
- {
- if (transcriptsWanted != null)
- {
- String transcriptId = (String) sf.getValue("transcript_id");
- if (!transcriptsWanted.contains(transcriptId))
- {
- // continue;
- }
- }
- String parent = (String) sf.getValue(PARENT);
- if (parentIdentifier.equals(parent))
- {
- transcriptFeatures.add(sf);
- }
- }
+ transcriptFeatures.add(sf);
}
}
if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
SequenceOntologyI.GENE))
{
- String id = (String) sf.getValue(ID);
- if ((GENE_PREFIX + accId).equals(id))
+ // NB features as gff use 'ID'; rest services return as 'id'
+ String id = (String) sf.getValue("ID");
+ if ((GENE_PREFIX + accId).equalsIgnoreCase(id))
{
return true;
}
@Override
protected boolean retainFeature(SequenceFeature sf, String accessionId)
{
- if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
- SequenceOntologyI.GENE))
+ SequenceOntologyI so = SequenceOntologyFactory.getInstance();
+ String type = sf.getType();
+ if (so.isA(type, SequenceOntologyI.GENE))
{
return false;
}
-
- if (isTranscript(sf.getType()))
+ if (isTranscript(type))
{
String parent = (String) sf.getValue(PARENT);
- if (!(GENE_PREFIX + accessionId).equals(parent))
+ if (!(GENE_PREFIX + accessionId).equalsIgnoreCase(parent))
{
return false;
}
return false;
}
- @Override
- protected List<String> getCrossReferenceDatabases()
- {
- // found these for ENSG00000157764 on 30/01/2016:
- // return new String[] {"Vega_gene", "OTTG", "ENS_LRG_gene", "ArrayExpress",
- // "EntrezGene", "HGNC", "MIM_GENE", "MIM_MORBID", "WikiGene"};
- return super.getCrossReferenceDatabases();
- }
-
/**
* Override to do nothing as Ensembl doesn't return a protein sequence for a
* gene identifier
return ACCESSION_REGEX;
}
+ /**
+ * Returns a descriptor for suitable feature display settings with
+ * <ul>
+ * <li>only exon or sequence_variant features (or their subtypes in the
+ * Sequence Ontology) visible</li>
+ * <li>variant features coloured red</li>
+ * <li>exon features coloured by label (exon name)</li>
+ * <li>variants displayed above (on top of) exons</li>
+ * </ul>
+ */
+ @Override
+ public FeatureSettingsModelI getFeatureColourScheme()
+ {
+ return new FeatureSettingsAdapter()
+ {
+ SequenceOntologyI so = SequenceOntologyFactory.getInstance();
+
+ @Override
+ public boolean isFeatureDisplayed(String type)
+ {
+ return (so.isA(type, SequenceOntologyI.EXON)
+ || so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT));
+ }
+
+ @Override
+ public FeatureColourI getFeatureColour(String type)
+ {
+ if (so.isA(type, SequenceOntologyI.EXON))
+ {
+ return new FeatureColour()
+ {
+ @Override
+ public boolean isColourByLabel()
+ {
+ return true;
+ }
+ };
+ }
+ if (so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT))
+ {
+ return new FeatureColour()
+ {
+
+ @Override
+ public Color getColour()
+ {
+ return Color.RED;
+ }
+ };
+ }
+ return null;
+ }
+
+ /**
+ * order to render sequence_variant after exon after the rest
+ */
+ @Override
+ public int compare(String feature1, String feature2)
+ {
+ if (so.isA(feature1, SequenceOntologyI.SEQUENCE_VARIANT))
+ {
+ return +1;
+ }
+ if (so.isA(feature2, SequenceOntologyI.SEQUENCE_VARIANT))
+ {
+ return -1;
+ }
+ if (so.isA(feature1, SequenceOntologyI.EXON))
+ {
+ return +1;
+ }
+ if (so.isA(feature2, SequenceOntologyI.EXON))
+ {
+ return -1;
+ }
+ return 0;
+ }
+ };
+ }
+
}