+/*
+ * 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.analysis.AlignmentUtils;
+import jalview.analysis.Dna;
+import jalview.bin.Cache;
import jalview.datamodel.Alignment;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.DBRefEntry;
import jalview.datamodel.Mapping;
import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceI;
+import jalview.datamodel.features.SequenceFeatures;
import jalview.exceptions.JalviewException;
import jalview.io.FastaFile;
import jalview.io.FileParse;
import jalview.io.gff.SequenceOntologyFactory;
import jalview.io.gff.SequenceOntologyI;
+import jalview.util.Comparison;
import jalview.util.DBRefUtils;
+import jalview.util.IntRangeComparator;
import jalview.util.MapList;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
-import java.util.Comparator;
import java.util.List;
+import java.util.Map;
+import java.util.Map.Entry;
/**
* Base class for Ensembl sequence fetchers
*/
public abstract class EnsemblSeqProxy extends EnsemblRestClient
{
- private static final List<String> CROSS_REFERENCES = Arrays
- .asList(new String[] { "CCDS", "Uniprot/SWISSPROT",
- "Uniprot/SPTREMBL" });
-
- protected static final String CONSEQUENCE_TYPE = "consequence_type";
+ private static final String ALLELES = "alleles";
protected static final String PARENT = "Parent";
GENOMIC("genomic"),
/**
- * type=cdna to fetch dna including UTRs
+ * type=cdna to fetch coding dna including UTRs
*/
CDNA("cdna"),
}
/**
- * A comparator to sort ranges into ascending start position order
- */
- private class RangeSorter implements Comparator<int[]>
- {
- boolean forwards;
-
- RangeSorter(boolean forward)
- {
- forwards = forward;
- }
-
- @Override
- public int compare(int[] o1, int[] o2)
- {
- return (forwards ? 1 : -1) * Integer.compare(o1[0], o2[0]);
- }
-
- }
-
- /**
* Default constructor (to use rest.ensembl.org)
*/
public EnsemblSeqProxy()
+ " chunks. Unexpected problem (" + r.getLocalizedMessage()
+ ")";
System.err.println(msg);
+ r.printStackTrace();
break;
}
}
EnsemblFeatureType[] features = getFeaturesToFetch();
AlignmentI geneFeatures = gffFetcher.getSequenceRecords(accId,
features);
- if (geneFeatures.getHeight() > 0)
+ if (geneFeatures != null && geneFeatures.getHeight() > 0)
{
genomicSequence = geneFeatures.getSequenceAt(0);
}
proteinSeq.createDatasetSequence();
querySeq.createDatasetSequence();
- MapList mapList = AlignmentUtils.mapCdsToProtein(querySeq, proteinSeq);
+ MapList mapList = AlignmentUtils
+ .mapCdsToProtein(querySeq, proteinSeq);
if (mapList != null)
{
// clunky: ensure Uniprot xref if we have one is on mapped sequence
SequenceI ds = proteinSeq.getDatasetSequence();
- ds.setSourceDBRef(proteinSeq.getSourceDBRef());
+ // TODO: Verify ensp primary ref is on proteinSeq.getDatasetSequence()
Mapping map = new Mapping(ds, mapList);
- DBRefEntry dbr = new DBRefEntry(getDbSource(), getDbVersion(),
- accId, map);
+ DBRefEntry dbr = new DBRefEntry(getDbSource(),
+ getEnsemblDataVersion(), proteinSeq.getName(), map);
querySeq.getDatasetSequence().addDBRef(dbr);
-
+ DBRefEntry[] uprots = DBRefUtils.selectRefs(ds.getDBRefs(),
+ new String[] { DBRefSource.UNIPROT });
+ DBRefEntry[] upxrefs = DBRefUtils.selectRefs(querySeq.getDBRefs(),
+ new String[] { DBRefSource.UNIPROT });
+ if (uprots != null)
+ {
+ for (DBRefEntry up : uprots)
+ {
+ // locate local uniprot ref and map
+ List<DBRefEntry> upx = DBRefUtils.searchRefs(upxrefs,
+ up.getAccessionId());
+ DBRefEntry upxref;
+ if (upx.size() != 0)
+ {
+ upxref = upx.get(0);
+
+ if (upx.size() > 1)
+ {
+ Cache.log
+ .warn("Implementation issue - multiple uniprot acc on product sequence.");
+ }
+ }
+ else
+ {
+ upxref = new DBRefEntry(DBRefSource.UNIPROT,
+ getEnsemblDataVersion(), up.getAccessionId());
+ }
+
+ Mapping newMap = new Mapping(ds, mapList);
+ upxref.setVersion(getEnsemblDataVersion());
+ upxref.setMap(newMap);
+ if (upx.size() == 0)
+ {
+ // add the new uniprot ref
+ querySeq.getDatasetSequence().addDBRef(upxref);
+ }
+
+ }
+ }
+
/*
* copy exon features to protein, compute peptide variants from dna
* variants and add as features on the protein sequence ta-da
*/
- AlignmentUtils.computeProteinFeatures(querySeq, proteinSeq, mapList);
+ AlignmentUtils
+ .computeProteinFeatures(querySeq, proteinSeq, mapList);
}
} catch (Exception e)
{
seq = seq.getDatasetSequence();
}
- EnsemblXref xrefFetcher = new EnsemblXref(getDomain());
- List<DBRefEntry> xrefs = xrefFetcher.getCrossReferences(seq.getName(),
- getCrossReferenceDatabases());
+ EnsemblXref xrefFetcher = new EnsemblXref(getDomain(), getDbSource(),
+ getEnsemblDataVersion());
+ List<DBRefEntry> xrefs = xrefFetcher.getCrossReferences(seq.getName());
for (DBRefEntry xref : xrefs)
{
seq.addDBRef(xref);
- /*
- * Save any Uniprot xref to be the reference for SIFTS mapping
- */
- if (DBRefSource.UNIPROT.equals(xref.getSource()))
- {
- seq.setSourceDBRef(xref);
- }
}
- }
- /**
- * Returns a list of database names to be used when fetching cross-references.
- * Specifically, the names are used to filter data returned by the Ensembl
- * xrefs REST service on the value in field 'dbname'.
- *
- * @return
- */
- protected List<String> getCrossReferenceDatabases()
- {
- return CROSS_REFERENCES;
+ /*
+ * and add a reference to itself
+ */
+ DBRefEntry self = new DBRefEntry(getDbSource(),
+ getEnsemblDataVersion(), seq.getName());
+ seq.addDBRef(self);
}
/**
throw new JalviewException("ENSEMBL Rest API not available.");
}
FileParse fp = getSequenceReader(ids);
+ if (fp == null)
+ {
+ return alignment;
+ }
+
FastaFile fr = new FastaFile(fp);
if (fr.hasWarningMessage())
{
if (fr.getSeqs().size() > 0)
{
- AlignmentI seqal = new Alignment(
- fr.getSeqsAsArray());
- for (SequenceI sq:seqal.getSequences())
+ AlignmentI seqal = new Alignment(fr.getSeqsAsArray());
+ for (SequenceI sq : seqal.getSequences())
{
if (sq.getDescription() == null)
{
if (ids.contains(name)
|| ids.contains(name.replace("ENSP", "ENST")))
{
- DBRefUtils.parseToDbRef(sq, DBRefSource.ENSEMBL, "0", name);
+ DBRefEntry dbref = DBRefUtils.parseToDbRef(sq, getDbSource(),
+ getEnsemblDataVersion(), name);
+ sq.addDBRef(dbref);
}
}
if (alignment == null)
urlstring.append("?type=").append(getSourceEnsemblType().getType());
urlstring.append(("&Accept=text/x-fasta"));
+ Map<String, String> params = getAdditionalParameters();
+ if (params != null)
+ {
+ for (Entry<String, String> entry : params.entrySet())
+ {
+ urlstring.append("&").append(entry.getKey()).append("=")
+ .append(entry.getValue());
+ }
+ }
+
URL url = new URL(urlstring.toString());
return url;
}
/**
+ * Override this method to add any additional x=y URL parameters needed
+ *
+ * @return
+ */
+ protected Map<String, String> getAdditionalParameters()
+ {
+ return null;
+ }
+
+ /**
* A sequence/id POST request currently allows up to 50 queries
*
* @see http://rest.ensembl.org/documentation/info/sequence_id_post
protected MapList getGenomicRangesFromFeatures(SequenceI sourceSequence,
String accId, int start)
{
- SequenceFeature[] sfs = sourceSequence.getSequenceFeatures();
- if (sfs == null)
+ // SequenceFeature[] sfs = sourceSequence.getSequenceFeatures();
+ List<SequenceFeature> sfs = sourceSequence.getFeatures()
+ .getPositionalFeatures();
+ if (sfs.isEmpty())
{
return null;
}
int mappedLength = 0;
int direction = 1; // forward
boolean directionSet = false;
-
+
for (SequenceFeature sf : sfs)
{
/*
// abort - mix of forward and backward
System.err.println("Error: forward and backward strand for "
+ accId);
- return null;
- }
- direction = strand;
- directionSet = true;
-
- /*
- * add to CDS ranges, semi-sorted forwards/backwards
- */
- if (strand < 0)
- {
- regions.add(0, new int[] { sf.getEnd(), sf.getBegin() });
- }
- else
- {
+ return null;
+ }
+ direction = strand;
+ directionSet = true;
+
+ /*
+ * add to CDS ranges, semi-sorted forwards/backwards
+ */
+ if (strand < 0)
+ {
+ regions.add(0, new int[] { sf.getEnd(), sf.getBegin() });
+ }
+ else
+ {
regions.add(new int[] { sf.getBegin(), sf.getEnd() });
}
mappedLength += Math.abs(sf.getEnd() - sf.getBegin() + 1);
}
}
}
-
+
if (regions.isEmpty())
{
System.out.println("Failed to identify target sequence for " + accId
* a final sort is needed since Ensembl returns CDS sorted within source
* (havana / ensembl_havana)
*/
- Collections.sort(regions, new RangeSorter(direction == 1));
-
+ Collections.sort(regions, direction == 1 ? IntRangeComparator.ASCENDING
+ : IntRangeComparator.DESCENDING);
+
List<int[]> to = Arrays.asList(new int[] { start,
start + mappedLength - 1 });
-
+
return new MapList(regions, to, 1, 1);
}
* @param mapping
* mapping from the sequence feature's coordinates to the target
* sequence
+ * @param forwardStrand
*/
protected void transferFeature(SequenceFeature sf,
- SequenceI targetSequence, MapList mapping)
+ SequenceI targetSequence, MapList mapping, boolean forwardStrand)
{
int start = sf.getBegin();
int end = sf.getEnd();
int[] mappedRange = mapping.locateInTo(start, end);
-
+
if (mappedRange != null)
{
- SequenceFeature copy = new SequenceFeature(sf);
- copy.setBegin(Math.min(mappedRange[0], mappedRange[1]));
- copy.setEnd(Math.max(mappedRange[0], mappedRange[1]));
+ String group = sf.getFeatureGroup();
+ if (".".equals(group))
+ {
+ group = getDbSource();
+ }
+ int newBegin = Math.min(mappedRange[0], mappedRange[1]);
+ int newEnd = Math.max(mappedRange[0], mappedRange[1]);
+ SequenceFeature copy = new SequenceFeature(sf, newBegin, newEnd,
+ group, sf.getScore());
targetSequence.addSequenceFeature(copy);
/*
- * for sequence_variant, make an additional feature with consequence
+ * for sequence_variant on reverse strand, have to convert the allele
+ * values to their complements
*/
- // if (SequenceOntologyFactory.getInstance().isA(sf.getType(),
- // SequenceOntologyI.SEQUENCE_VARIANT))
- // {
- // String consequence = (String) sf.getValue(CONSEQUENCE_TYPE);
- // if (consequence != null)
- // {
- // SequenceFeature sf2 = new SequenceFeature("consequence",
- // consequence, copy.getBegin(), copy.getEnd(), 0f,
- // null);
- // targetSequence.addSequenceFeature(sf2);
- // }
- // }
+ if (!forwardStrand
+ && SequenceOntologyFactory.getInstance().isA(sf.getType(),
+ SequenceOntologyI.SEQUENCE_VARIANT))
+ {
+ reverseComplementAlleles(copy);
+ }
+ }
+ }
+
+ /**
+ * Change the 'alleles' value of a feature by converting to complementary
+ * bases, and also update the feature description to match
+ *
+ * @param sf
+ */
+ static void reverseComplementAlleles(SequenceFeature sf)
+ {
+ final String alleles = (String) sf.getValue(ALLELES);
+ if (alleles == null)
+ {
+ return;
+ }
+ StringBuilder complement = new StringBuilder(alleles.length());
+ for (String allele : alleles.split(","))
+ {
+ reverseComplementAllele(complement, allele);
+ }
+ String comp = complement.toString();
+ sf.setValue(ALLELES, comp);
+ sf.setDescription(comp);
+
+ /*
+ * replace value of "alleles=" in sf.ATTRIBUTES as well
+ * so 'output as GFF' shows reverse complement alleles
+ */
+ String atts = sf.getAttributes();
+ if (atts != null)
+ {
+ atts = atts.replace(ALLELES + "=" + alleles, ALLELES + "=" + comp);
+ sf.setAttributes(atts);
+ }
+ }
+
+ /**
+ * Makes the 'reverse complement' of the given allele and appends it to the
+ * buffer, after a comma separator if not the first
+ *
+ * @param complement
+ * @param allele
+ */
+ static void reverseComplementAllele(StringBuilder complement,
+ String allele)
+ {
+ if (complement.length() > 0)
+ {
+ complement.append(",");
+ }
+
+ /*
+ * some 'alleles' are actually descriptive terms
+ * e.g. HGMD_MUTATION, PhenCode_variation
+ * - we don't want to 'reverse complement' these
+ */
+ if (!Comparison.isNucleotideSequence(allele, true))
+ {
+ complement.append(allele);
+ }
+ else
+ {
+ for (int i = allele.length() - 1; i >= 0; i--)
+ {
+ complement.append(Dna.getComplement(allele.charAt(i)));
+ }
}
}
return false;
}
- // long start = System.currentTimeMillis();
- SequenceFeature[] sfs = sourceSequence.getSequenceFeatures();
- MapList mapping = getGenomicRangesFromFeatures(sourceSequence, accessionId,
- targetSequence.getStart());
+ long start = System.currentTimeMillis();
+ // SequenceFeature[] sfs = sourceSequence.getSequenceFeatures();
+ List<SequenceFeature> sfs = sourceSequence.getFeatures()
+ .getPositionalFeatures();
+ MapList mapping = getGenomicRangesFromFeatures(sourceSequence,
+ accessionId, targetSequence.getStart());
if (mapping == null)
{
return false;
boolean result = transferFeatures(sfs, targetSequence, mapping,
accessionId);
- // System.out.println("transferFeatures (" + (sfs.length) + " --> "
- // + targetSequence.getSequenceFeatures().length + ") to "
- // + targetSequence.getName()
- // + " took " + (System.currentTimeMillis() - start) + "ms");
+ System.out.println("transferFeatures (" + (sfs.size()) + " --> "
+ + targetSequence.getFeatures().getFeatureCount(true) + ") to "
+ + targetSequence.getName() + " took "
+ + (System.currentTimeMillis() - start) + "ms");
return result;
}
* converted using the mapping. Features which do not overlap are ignored.
* Features whose parent is not the specified identifier are also ignored.
*
- * @param features
+ * @param sfs
* @param targetSequence
* @param mapping
* @param parentId
* @return
*/
- protected boolean transferFeatures(SequenceFeature[] features,
+ protected boolean transferFeatures(List<SequenceFeature> sfs,
SequenceI targetSequence, MapList mapping, String parentId)
{
final boolean forwardStrand = mapping.isFromForwardStrand();
/*
- * sort features by start position (descending if reverse strand)
- * before transferring (in forwards order) to the target sequence
+ * sort features by start position (which corresponds to end
+ * position descending if reverse strand) so as to add them in
+ * 'forwards' order to the target sequence
*/
- Arrays.sort(features, new Comparator<SequenceFeature>()
- {
- @Override
- public int compare(SequenceFeature o1, SequenceFeature o2)
- {
- int c = Integer.compare(o1.getBegin(), o2.getBegin());
- return forwardStrand ? c : -c;
- }
- });
+ SequenceFeatures.sortFeatures(sfs, forwardStrand);
boolean transferred = false;
- for (SequenceFeature sf : features)
+ for (SequenceFeature sf : sfs)
{
if (retainFeature(sf, parentId))
{
- transferFeature(sf, targetSequence, mapping);
+ transferFeature(sf, targetSequence, mapping, forwardStrand);
transferred = true;
}
}
/**
* Returns a (possibly empty) list of features on the sequence which have the
- * specified sequence ontology type (or a sub-type of it), and the given
+ * specified sequence ontology term (or a sub-type of it), and the given
* identifier as parent
*
* @param sequence
- * @param type
+ * @param term
* @param parentId
* @return
*/
protected List<SequenceFeature> findFeatures(SequenceI sequence,
- String type, String parentId)
+ String term, String parentId)
{
List<SequenceFeature> result = new ArrayList<SequenceFeature>();
-
- SequenceFeature[] sfs = sequence.getSequenceFeatures();
- if (sfs != null) {
- SequenceOntologyI so = SequenceOntologyFactory.getInstance();
- for (SequenceFeature sf :sfs) {
- if (so.isA(sf.getType(), type))
- {
- String parent = (String) sf.getValue(PARENT);
- if (parent.equals(parentId))
- {
- result.add(sf);
- }
- }
+
+ List<SequenceFeature> sfs = sequence.getFeatures()
+ .getFeaturesByOntology(term);
+ for (SequenceFeature sf : sfs)
+ {
+ String parent = (String) sf.getValue(PARENT);
+ if (parent != null && parent.equals(parentId))
+ {
+ result.add(sf);
}
}
+
return result;
}