package jalview.ext.ensembl;
import jalview.api.FeatureColourI;
-import jalview.api.FeatureSettingsI;
+import jalview.api.FeatureSettingsModelI;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceI;
import jalview.io.gff.SequenceOntologyFactory;
import jalview.io.gff.SequenceOntologyI;
-import jalview.schemes.FeatureColourAdapter;
+import jalview.schemes.FeatureColour;
import jalview.schemes.FeatureSettingsAdapter;
import jalview.util.MapList;
-import jalview.util.StringUtils;
import java.awt.Color;
import java.io.UnsupportedEncodingException;
*/
public class EnsemblGene extends EnsemblSeqProxy
{
- private static final List<String> CROSS_REFERENCES = Arrays
- .asList(new String[] { "CCDS" });
-
private static final String GENE_PREFIX = "gene:";
/*
* </ul>
*
* @param query
- * one or more identifiers separated by a space
- * @return an alignment containing one or more genes, and possibly
- * transcripts, or null
+ * 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
{
- // todo: tidy up handling of one or multiple accession ids
- String[] queries = query.split(getAccessionSeparator());
-
/*
- * if given a transcript id, look up its gene parent
+ * convert to a non-duplicated list of gene identifiers
*/
- if (isTranscriptIdentifier(query))
+ List<String> geneIds = getGeneIds(query);
+
+ AlignmentI al = null;
+ for (String geneId : geneIds)
{
- // we are assuming all transcripts have the same gene parent here
- query = new EnsemblLookup(getDomain()).getParent(queries[0]);
- if (query == null)
+ /*
+ * fetch the gene sequence(s) with features and xrefs
+ */
+ AlignmentI geneAlignment = super.getSequenceRecords(geneId);
+
+ if (geneAlignment.getHeight() == 1)
{
- return null;
+ getTranscripts(geneAlignment, geneId);
+ }
+ if (al == null)
+ {
+ al = geneAlignment;
+ }
+ else
+ {
+ al.append(geneAlignment);
}
}
+ return al;
+ }
- /*
- * if given a gene or other external name, lookup and fetch
- * the corresponding gene for all model organisms
- */
- if (!isGeneIdentifier(query))
+ /**
+ * Converts a query, which may contain one or more gene or transcript
+ * identifiers, into a non-redundant list of gene identifiers.
+ *
+ * @param accessions
+ * @return
+ */
+ List<String> getGeneIds(String accessions)
+ {
+ List<String> geneIds = new ArrayList<String>();
+
+ for (String acc : accessions.split(getAccessionSeparator()))
{
- List<String> geneIds = new EnsemblSymbol(getDomain()).getIds(query);
- if (geneIds.isEmpty())
+ if (isGeneIdentifier(acc))
{
- return null;
+ if (!geneIds.contains(acc))
+ {
+ geneIds.add(acc);
+ }
}
- String theIds = StringUtils.listToDelimitedString(geneIds,
- getAccessionSeparator());
- return getSequenceRecords(theIds);
- }
- /*
- * fetch the gene sequence(s) with features and xrefs
- */
- AlignmentI al = super.getSequenceRecords(query);
+ /*
+ * if given a transcript id, look up its gene parent
+ */
+ else if (isTranscriptIdentifier(acc))
+ {
+ String geneId = new EnsemblLookup(getDomain()).getParent(acc);
+ if (geneId != null && !geneIds.contains(geneId))
+ {
+ geneIds.add(geneId);
+ }
+ }
- /*
- * if we retrieved a single gene, get its transcripts as well
- */
- if (al.getHeight() == 1)
- {
- getTranscripts(al, query);
+ /*
+ * if given a gene or other external name, lookup and fetch
+ * the corresponding gene for all model organisms
+ */
+ else
+ {
+ List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
+ getDbVersion()).getIds(acc);
+ for (String geneId : ids)
+ {
+ if (!geneIds.contains(geneId))
+ {
+ geneIds.add(geneId);
+ }
+ }
+ }
}
-
- return al;
+ return geneIds;
}
/**
*/
protected String getGeneIdentifiersForName(String query)
{
- List<String> ids = new EnsemblSymbol(getDomain()).getIds(query);
+ List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
+ getDbVersion()).getIds(query);
if (ids != null)
{
for (String id : ids)
List<int[]> mapTo = new ArrayList<int[]>();
mapTo.add(new int[] { 1, transcriptLength });
MapList mapping = new MapList(mappedFrom, mapTo, 1, 1);
- new EnsemblCdna(getDomain()).transferFeatures(
- gene.getSequenceFeatures(), transcript.getDatasetSequence(),
- mapping, parentId);
+ EnsemblCdna cdna = new EnsemblCdna(getDomain());
+ cdna.transferFeatures(gene.getSequenceFeatures(),
+ transcript.getDatasetSequence(), mapping, parentId);
/*
* fetch and save cross-references
*/
- new EnsemblCdna(getDomain()).getCrossReferences(transcript);
+ cdna.getCrossReferences(transcript);
/*
* and finally fetch the protein product and save as a cross-reference
*/
- new EnsemblCdna(getDomain()).addProteinProduct(transcript);
+ cdna.addProteinProduct(transcript);
return transcript;
}
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 CROSS_REFERENCES;
- }
-
/**
* 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 FeatureSettingsI getFeatureColourScheme()
+ public FeatureSettingsModelI getFeatureColourScheme()
{
return new FeatureSettingsAdapter()
{
{
if (so.isA(type, SequenceOntologyI.EXON))
{
- return new FeatureColourAdapter()
+ return new FeatureColour()
{
@Override
public boolean isColourByLabel()
}
if (so.isA(type, SequenceOntologyI.SEQUENCE_VARIANT))
{
- return new FeatureColourAdapter()
+ return new FeatureColour()
{
@Override