continue;
}
- parseChromosomeLocations(geneAlignment);
-
if (geneAlignment.getHeight() == 1)
{
+ findGeneLoci(geneAlignment.getSequenceAt(0), geneId);
getTranscripts(geneAlignment, geneId);
}
if (al == null)
}
/**
+ * Calls the /lookup/id REST service, parses the response for gene
+ * coordinates, and if successful, adds these to the sequence. If this fails,
+ * fall back on trying to parse the sequence description in case it is in
+ * Ensembl-gene format e.g. chromosome:GRCh38:17:45051610:45109016:1.
+ *
+ * @param seq
+ * @param geneId
+ */
+ void findGeneLoci(SequenceI seq, String geneId)
+ {
+ GeneLociI geneLoci = new EnsemblLookup(getDomain()).getGeneLoci(geneId);
+ if (geneLoci != null)
+ {
+ seq.setGeneLoci(geneLoci.getSpeciesId(), geneLoci.getAssemblyId(),
+ geneLoci.getChromosomeId(), geneLoci.getMap());
+ }
+ else
+ {
+ parseChromosomeLocations(seq);
+ }
+ }
+
+ /**
* Parses and saves fields of an Ensembl-style description e.g.
* chromosome:GRCh38:17:45051610:45109016:1
*
- * @param alignment
+ * @param seq
*/
- private void parseChromosomeLocations(AlignmentI alignment)
+ boolean parseChromosomeLocations(SequenceI seq)
{
- for (SequenceI seq : alignment.getSequences())
+ String description = seq.getDescription();
+ if (description == null)
+ {
+ return false;
+ }
+ String[] tokens = description.split(":");
+ if (tokens.length == 6 && tokens[0].startsWith(DBRefEntry.CHROMOSOME))
{
- String description = seq.getDescription();
- if (description == null)
+ String ref = tokens[1];
+ String chrom = tokens[2];
+ try
{
- continue;
- }
- String[] tokens = description.split(":");
- if (tokens.length == 6 && tokens[0].startsWith(DBRefEntry.CHROMOSOME))
+ int chStart = Integer.parseInt(tokens[3]);
+ int chEnd = Integer.parseInt(tokens[4]);
+ boolean forwardStrand = "1".equals(tokens[5]);
+ String species = ""; // not known here
+ int[] from = new int[] { seq.getStart(), seq.getEnd() };
+ int[] to = new int[] { forwardStrand ? chStart : chEnd,
+ forwardStrand ? chEnd : chStart };
+ MapList map = new MapList(from, to, 1, 1);
+ seq.setGeneLoci(species, ref, chrom, map);
+ return true;
+ } catch (NumberFormatException e)
{
- String ref = tokens[1];
- String chrom = tokens[2];
- try
- {
- int chStart = Integer.parseInt(tokens[3]);
- int chEnd = Integer.parseInt(tokens[4]);
- boolean forwardStrand = "1".equals(tokens[5]);
- String species = ""; // dunno yet!
- int[] from = new int[] { seq.getStart(), seq.getEnd() };
- int[] to = new int[] { forwardStrand ? chStart : chEnd,
- forwardStrand ? chEnd : chStart };
- MapList map = new MapList(from, to, 1, 1);
- seq.setGeneLoci(species, ref, chrom, map);
- } catch (NumberFormatException e)
- {
- System.err.println("Bad integers in description " + description);
- }
+ System.err.println("Bad integers in description " + description);
}
}
+ return false;
}
/**
- * Converts a query, which may contain one or more gene or transcript
- * identifiers, into a non-redundant list of gene identifiers.
+ * 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 accessions
* @return
for (String acc : accessions.split(getAccessionSeparator()))
{
- if (isGeneIdentifier(acc))
- {
- if (!geneIds.contains(acc))
- {
- geneIds.add(acc);
- }
- }
-
/*
- * if given a transcript id, look up its gene parent
+ * First try lookup as an Ensembl (gene or transcript) identifier
*/
- else if (isTranscriptIdentifier(acc))
+ String geneId = new EnsemblLookup(getDomain()).getGeneId(acc);
+ if (geneId != null)
{
- String geneId = new EnsemblLookup(getDomain()).getParent(acc);
- if (geneId != null && !geneIds.contains(geneId))
+ if (!geneIds.contains(geneId))
{
geneIds.add(geneId);
}
}
- else if (isProteinIdentifier(acc))
- {
- String tscriptId = new EnsemblLookup(getDomain()).getParent(acc);
- if (tscriptId != null)
- {
- String geneId = new EnsemblLookup(getDomain())
- .getParent(tscriptId);
-
- if (geneId != null && !geneIds.contains(geneId))
- {
- geneIds.add(geneId);
- }
- }
- // NOTE - acc is lost if it resembles an ENS.+ ID but isn't actually
- // resolving to one... e.g. ENSMICP00000009241
- }
- /*
- * if given a gene or other external name, lookup and fetch
- * the corresponding gene for all model organisms
- */
else
{
+ /*
+ * 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()).getIds(acc);
- for (String geneId : ids)
+ getDbVersion()).getGeneIds(acc);
+ for (String id : ids)
{
- if (!geneIds.contains(geneId))
+ if (!geneIds.contains(id))
{
- geneIds.add(geneId);
+ geneIds.add(id);
}
}
}
}
/**
- * Attempts to get Ensembl stable identifiers for model organisms for a gene
- * name by calling the xrefs symbol REST service to resolve the gene name.
- *
- * @param query
- * @return
- */
- protected String getGeneIdentifiersForName(String query)
- {
- List<String> ids = new EnsemblSymbol(getDomain(), getDbSource(),
- getDbVersion()).getIds(query);
- if (ids != null)
- {
- for (String id : ids)
- {
- if (isGeneIdentifier(id))
- {
- return id;
- }
- }
- }
- return null;
- }
-
- /**
* Constructs all transcripts for the gene, as identified by "transcript"
* features whose Parent is the requested gene. The coding transcript
* sequences (i.e. with introns omitted) are added to the alignment.
git://source.jalview.org / jalview.git / blobdiff