+/*
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.6)
+ * Copyright (C) 2010 J Procter, AM Waterhouse, G Barton, M Clamp, S Searle
+ *
+ * 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/>.
+ */
package jalview.datamodel.xdb.embl;
import jalview.datamodel.DBRefEntry;
+import jalview.datamodel.DBRefSource;
+import jalview.datamodel.FeatureProperties;
+import jalview.datamodel.Mapping;
import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceI;
* EMBL Feature support is limited. The text below is included for the benefit
* of any developer working on improving EMBL feature import in Jalview.
* Extract from EMBL feature specification see
- * http://www.embl-ebi.ac.uk/embl/Documentation/FT_definitions/feature_table.html
- * 3.5 Location 3.5.1 Purpose
+ * http://www.embl-ebi.ac.uk/embl/Documentation
+ * /FT_definitions/feature_table.html 3.5 Location 3.5.1 Purpose
*
* The location indicates the region of the presented sequence which
* corresponds to a feature.
*
* join(1..100,J00194.1:100..202) Joins region 1..100 of the existing entry
* with the region 100..202 of remote entry J00194
- *
*/
/**
* Recover annotated sequences from EMBL file
*/
public jalview.datamodel.SequenceI[] getSequences(boolean noNa,
boolean noPeptide, String sourceDb)
- {
+ { // TODO: ensure emblEntry.getSequences behaves correctly for returning all
+ // cases of noNa and noPeptide
Vector seqs = new Vector();
Sequence dna = null;
if (!noNa)
{
+ // In theory we still need to create this if noNa is set to avoid a null
+ // pointer exception
dna = new Sequence(sourceDb + "|" + accession, sequence.getSequence());
dna.setDescription(desc);
- dna.addDBRef(new DBRefEntry(sourceDb, version, accession));
- // TODO: add mapping for parentAccession attribute
+ DBRefEntry retrievedref = new DBRefEntry(sourceDb, version, accession);
+ dna.addDBRef(retrievedref);
+ // add map to indicate the sequence is a valid coordinate frame for the
+ // dbref
+ retrievedref.setMap(new Mapping(null, new int[]
+ { 1, dna.getLength() }, new int[]
+ { 1, dna.getLength() }, 1, 1));
// TODO: transform EMBL Database refs to canonical form
if (dbRefs != null)
for (Iterator i = dbRefs.iterator(); i.hasNext(); dna
{
for (Iterator i = features.iterator(); i.hasNext();)
{
- boolean nextFeature=false;
EmblFeature feature = (EmblFeature) i.next();
if (!noNa)
{
;
}
}
- if (feature.getName().equalsIgnoreCase("CDS"))
+ if (FeatureProperties.isCodingFeature(sourceDb, feature.getName()))
{
- // extract coding region(s)
- jalview.datamodel.Mapping map = null;
- int[] exon = null;
- if (feature.locations != null && feature.locations.size() > 0)
- {
- for (Enumeration locs = feature.locations.elements(); locs
- .hasMoreElements();)
- {
- EmblFeatureLocations loc = (EmblFeatureLocations) locs
- .nextElement();
- int[] se = loc.getElementRanges(accession);
- if (exon == null)
- {
- exon = se;
- }
- else
- {
- int[] t = new int[exon.length + se.length];
- System.arraycopy(exon, 0, t, 0, exon.length);
- System.arraycopy(se, 0, t, exon.length, se.length);
- exon = t;
- }
- }
- }
- String prseq = null;
- String prname = new String();
- String prid = null;
- Hashtable vals = new Hashtable();
- int prstart = 1;
- // get qualifiers
- if (feature.getQualifiers() != null
- && feature.getQualifiers().size() > 0)
- {
- for (Iterator quals = feature.getQualifiers().iterator(); quals
- .hasNext();)
- {
- Qualifier q = (Qualifier) quals.next();
- if (q.getName().equals("translation"))
- {
- prseq = q.getValues()[0];
- }
- else if (q.getName().equals("protein_id"))
- {
- prid = q.getValues()[0];
- }
- else if (q.getName().equals("codon_start"))
- {
- prstart = Integer.parseInt(q.getValues()[0]);
- }
- else if (q.getName().equals("product"))
- {
- prname = q.getValues()[0];
- }
- else
- {
- // throw anything else into the additional properties hash
- vals.put(q.getName(), q.getValues().toString());
- }
- }
- }
- Sequence product = null;
- if (prseq != null && prname != null && prid != null)
- {
- // extract proteins.
- if (!noPeptide)
- {
- product = new Sequence(sourceDb + "|" + "EMBLCDS|" + prid
- + "|" + prname, prseq, prstart, prstart
- + prseq.length() - 1);
- product.setDescription("Protein Product from " + sourceDb);
- seqs.add(product);
- }
- // we have everything - create the mapping and perhaps the protein
- // sequence
- map = new jalview.datamodel.Mapping(product, exon, new int[]
- { prstart, prstart + prseq.length() - 1 }, 3, 1);
- // add cds feature to dna seq - this may include the stop codon
- for (int xint = 0; xint < exon.length; xint += 2)
- {
- SequenceFeature sf = new SequenceFeature();
- sf.setBegin(exon[xint]);
- sf.setEnd(exon[xint + 1]);
- sf.setType(feature.getName());
- sf.setFeatureGroup(jalview.datamodel.DBRefSource.EMBL);
- sf.setDescription("Exon " + (1 + xint) + " for protein '"
- + prname + "' EMBLCDS:" + prid);
- if (vals != null && vals.size() > 0)
- {
- Enumeration kv = vals.elements();
- while (kv.hasMoreElements())
- {
- Object key = kv.nextElement();
- if (key != null)
- sf.setValue(key.toString(), vals.get(key));
- }
- }
- dna.addSequenceFeature(sf);
- }
- }
- // add dbRefs to sequence
- if (feature.dbRefs != null && feature.dbRefs.size() > 0)
- {
- for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext();)
- {
- DBRefEntry ref = (DBRefEntry) dbr.next();
- ref.setSource(jalview.util.DBRefUtils.getCanonicalName(ref
- .getSource()));
- if (ref.getSource().equals(
- jalview.datamodel.DBRefSource.UNIPROT))
- {
- ref.setMap(map);
- }
- if (product != null)
- {
- DBRefEntry pref = new DBRefEntry(ref.getSource(), ref
- .getVersion(), ref.getAccessionId());
- pref.setMap(null); // reference is direct
- }
- dna.addDBRef(ref);
- }
- }
-
+ parseCodingFeature(feature, sourceDb, seqs, dna, noPeptide);
}
else
{
} catch (Exception e)
{
System.err.println("EMBL Record Features parsing error!");
- System.err.println("Please report the following to help@jalview.org :");
- System.err.println("EMBL Record "+accession);
- System.err.println("Resulted in exception: "+e.getMessage());
+ System.err
+ .println("Please report the following to help@jalview.org :");
+ System.err.println("EMBL Record " + accession);
+ System.err.println("Resulted in exception: " + e.getMessage());
e.printStackTrace(System.err);
}
- if (!noNa && dna!=null)
+ if (!noNa && dna != null)
{
seqs.add(dna);
}
}
return sqs;
}
+
+ /**
+ * attempt to extract coding region and product from a feature and properly
+ * decorate it with annotations.
+ *
+ * @param feature
+ * coding feature
+ * @param sourceDb
+ * source database for the EMBLXML
+ * @param seqs
+ * place where sequences go
+ * @param dna
+ * parent dna sequence for this record
+ * @param noPeptide
+ * flag for generation of Peptide sequence objects
+ */
+ private void parseCodingFeature(EmblFeature feature, String sourceDb,
+ Vector seqs, Sequence dna, boolean noPeptide)
+ {
+ boolean isEmblCdna = sourceDb.equals(DBRefSource.EMBLCDS);
+ // extract coding region(s)
+ jalview.datamodel.Mapping map = null;
+ int[] exon = null;
+ if (feature.locations != null && feature.locations.size() > 0)
+ {
+ for (Enumeration locs = feature.locations.elements(); locs
+ .hasMoreElements();)
+ {
+ EmblFeatureLocations loc = (EmblFeatureLocations) locs
+ .nextElement();
+ int[] se = loc.getElementRanges(accession);
+ if (exon == null)
+ {
+ exon = se;
+ }
+ else
+ {
+ int[] t = new int[exon.length + se.length];
+ System.arraycopy(exon, 0, t, 0, exon.length);
+ System.arraycopy(se, 0, t, exon.length, se.length);
+ exon = t;
+ }
+ }
+ }
+ String prseq = null;
+ String prname = new String();
+ String prid = null;
+ Hashtable vals = new Hashtable();
+ int prstart = 1;
+ // get qualifiers
+ if (feature.getQualifiers() != null
+ && feature.getQualifiers().size() > 0)
+ {
+ for (Iterator quals = feature.getQualifiers().iterator(); quals
+ .hasNext();)
+ {
+ Qualifier q = (Qualifier) quals.next();
+ if (q.getName().equals("translation"))
+ {
+ StringBuffer prsq = new StringBuffer(q.getValues()[0]);
+ int p = prsq.indexOf(" ");
+ while (p > -1)
+ {
+ prsq.deleteCharAt(p);
+ p = prsq.indexOf(" ", p);
+ }
+ prseq = prsq.toString();
+ prsq = null;
+
+ }
+ else if (q.getName().equals("protein_id"))
+ {
+ prid = q.getValues()[0];
+ }
+ else if (q.getName().equals("codon_start"))
+ {
+ prstart = Integer.parseInt(q.getValues()[0]);
+ }
+ else if (q.getName().equals("product"))
+ {
+ prname = q.getValues()[0];
+ }
+ else
+ {
+ // throw anything else into the additional properties hash
+ String[] s = q.getValues();
+ StringBuffer sb = new StringBuffer();
+ if (s != null)
+ {
+ for (int i = 0; i < s.length; i++)
+ {
+ sb.append(s[i]);
+ sb.append("\n");
+ }
+ }
+ vals.put(q.getName(), sb.toString());
+ }
+ }
+ }
+ Sequence product = null;
+ exon = adjustForPrStart(prstart, exon);
+
+ if (prseq != null && prname != null && prid != null)
+ {
+ // extract proteins.
+ product = new Sequence(prid, prseq, 1, prseq.length());
+ product
+ .setDescription(((prname.length() == 0) ? "Protein Product from "
+ + sourceDb
+ : prname));
+ if (!noPeptide)
+ {
+ // Protein is also added to vector of sequences returned
+ seqs.add(product);
+ }
+ // we have everything - create the mapping and perhaps the protein
+ // sequence
+ if (exon == null || exon.length == 0)
+ {
+ System.err
+ .println("Implementation Notice: EMBLCDS records not properly supported yet - Making up the CDNA region of this sequence... may be incorrect ("
+ + sourceDb + ":" + getAccession() + ")");
+ if (prseq.length() * 3 == (1 - prstart + dna.getSequence().length))
+ {
+ System.err
+ .println("Not allowing for additional stop codon at end of cDNA fragment... !");
+ // this might occur for CDS sequences where no features are
+ // marked.
+ exon = new int[]
+ { dna.getStart() + (prstart - 1), dna.getEnd() };
+ map = new jalview.datamodel.Mapping(product, exon, new int[]
+ { 1, prseq.length() }, 3, 1);
+ }
+ if ((prseq.length() + 1) * 3 == (1 - prstart + dna.getSequence().length))
+ {
+ System.err
+ .println("Allowing for additional stop codon at end of cDNA fragment... will probably cause an error in VAMSAs!");
+ exon = new int[]
+ { dna.getStart() + (prstart - 1), dna.getEnd() - 3 };
+ map = new jalview.datamodel.Mapping(product, exon, new int[]
+ { 1, prseq.length() }, 3, 1);
+ }
+ }
+ else
+ {
+ // Trim the exon mapping if necessary - the given product may only be a
+ // fragment of a larger protein. (EMBL:AY043181 is an example)
+
+ if (isEmblCdna)
+ {
+ // TODO: Add a DbRef back to the parent EMBL sequence with the exon
+ // map
+ // if given a dataset reference, search dataset for parent EMBL
+ // sequence if it exists and set its map
+ // make a new feature annotating the coding contig
+ }
+ else
+ {
+ // final product length trunctation check
+
+ map = new jalview.datamodel.Mapping(product,
+ adjustForProteinLength(prseq.length(), exon), new int[]
+ { 1, prseq.length() }, 3, 1);
+ // reconstruct the EMBLCDS entry
+ // TODO: this is only necessary when there codon annotation is
+ // complete (I think JBPNote)
+ DBRefEntry pcdnaref = new DBRefEntry();
+ pcdnaref.setAccessionId(prid);
+ pcdnaref.setSource(DBRefSource.EMBLCDS);
+ pcdnaref.setVersion(getVersion()); // same as parent EMBL version.
+ jalview.util.MapList mp = new jalview.util.MapList(new int[]
+ { 1, prseq.length() }, new int[]
+ { 1 + (prstart - 1), (prstart - 1) + 3 * prseq.length() }, 1, 3);
+ // { 1 + (prstart - 1) * 3,
+ // 1 + (prstart - 1) * 3 + prseq.length() * 3 - 1 }, new int[]
+ // { 1prstart, prstart + prseq.length() - 1 }, 3, 1);
+ pcdnaref.setMap(new Mapping(mp));
+ if (product != null)
+ product.addDBRef(pcdnaref);
+
+ }
+ }
+ // add cds feature to dna seq - this may include the stop codon
+ for (int xint = 0; exon != null && xint < exon.length; xint += 2)
+ {
+ SequenceFeature sf = new SequenceFeature();
+ sf.setBegin(exon[xint]);
+ sf.setEnd(exon[xint + 1]);
+ sf.setType(feature.getName());
+ sf.setFeatureGroup(sourceDb);
+ sf.setDescription("Exon " + (1 + (int) (xint / 2))
+ + " for protein '" + prname + "' EMBLCDS:" + prid);
+ sf.setValue(FeatureProperties.EXONPOS, new Integer(1 + xint));
+ sf.setValue(FeatureProperties.EXONPRODUCT, prname);
+ if (vals != null && vals.size() > 0)
+ {
+ Enumeration kv = vals.elements();
+ while (kv.hasMoreElements())
+ {
+ Object key = kv.nextElement();
+ if (key != null)
+ sf.setValue(key.toString(), vals.get(key));
+ }
+ }
+ dna.addSequenceFeature(sf);
+ }
+ }
+ // add dbRefs to sequence
+ if (feature.dbRefs != null && feature.dbRefs.size() > 0)
+ {
+ for (Iterator dbr = feature.dbRefs.iterator(); dbr.hasNext();)
+ {
+ DBRefEntry ref = (DBRefEntry) dbr.next();
+ ref.setSource(jalview.util.DBRefUtils.getCanonicalName(ref
+ .getSource()));
+ // Hard code the kind of protein product accessions that EMBL cite
+ if (ref.getSource().equals(jalview.datamodel.DBRefSource.UNIPROT))
+ {
+ ref.setMap(map);
+ if (map != null && map.getTo() != null)
+ {
+ map.getTo().addDBRef(
+ new DBRefEntry(ref.getSource(), ref.getVersion(), ref
+ .getAccessionId())); // don't copy map over.
+ if (map.getTo().getName().indexOf(prid) == 0)
+ {
+ map.getTo().setName(
+ jalview.datamodel.DBRefSource.UNIPROT + "|"
+ + ref.getAccessionId());
+ }
+ }
+ }
+ if (product != null)
+ {
+ DBRefEntry pref = new DBRefEntry(ref.getSource(), ref
+ .getVersion(), ref.getAccessionId());
+ pref.setMap(null); // reference is direct
+ product.addDBRef(pref);
+ // Add converse mapping reference
+ if (map != null)
+ {
+ Mapping pmap = new Mapping(dna, map.getMap().getInverse());
+ pref = new DBRefEntry(sourceDb, getVersion(), this
+ .getAccession());
+ pref.setMap(pmap);
+ if (map.getTo() != null)
+ {
+ map.getTo().addDBRef(pref);
+ }
+ }
+ }
+ dna.addDBRef(ref);
+ }
+ }
+ }
+
+ private int[] adjustForPrStart(int prstart, int[] exon)
+ {
+
+ int origxon[], sxpos = -1;
+ int sxstart, sxstop; // unnecessary variables used for debugging
+ // first adjust range for codon start attribute
+ if (prstart > 1)
+ {
+ origxon = new int[exon.length];
+ System.arraycopy(exon, 0, origxon, 0, exon.length);
+ int cdspos = 0;
+ for (int x = 0; x < exon.length && sxpos == -1; x += 2)
+ {
+ cdspos += exon[x + 1] - exon[x] + 1;
+ if (prstart <= cdspos)
+ {
+ sxpos = x;
+ sxstart = exon[x];
+ sxstop = exon[x + 1];
+ // and adjust start boundary of first exon.
+ exon[x] = exon[x + 1] - cdspos + prstart;
+ break;
+ }
+ }
+
+ if (sxpos > 0)
+ {
+ int[] nxon = new int[exon.length - sxpos];
+ System.arraycopy(exon, sxpos, nxon, 0, exon.length - sxpos);
+ exon = nxon;
+ }
+ }
+ return exon;
+ }
+
+ /**
+ * truncate the last exon interval to the prlength'th codon
+ *
+ * @param prlength
+ * @param exon
+ * @return new exon
+ */
+ private int[] adjustForProteinLength(int prlength, int[] exon)
+ {
+
+ int origxon[], sxpos = -1, endxon = 0, cdslength = prlength * 3;
+ int sxstart, sxstop; // unnecessary variables used for debugging
+ // first adjust range for codon start attribute
+ if (prlength >= 1 && exon != null)
+ {
+ origxon = new int[exon.length];
+ System.arraycopy(exon, 0, origxon, 0, exon.length);
+ int cdspos = 0;
+ for (int x = 0; x < exon.length && sxpos == -1; x += 2)
+ {
+ cdspos += exon[x + 1] - exon[x] + 1;
+ if (cdslength <= cdspos)
+ {
+ // advanced beyond last codon.
+ sxpos = x;
+ sxstart = exon[x];
+ sxstop = exon[x + 1];
+ if (cdslength != cdspos)
+ {
+ System.err
+ .println("Truncating final exon interval on region by "
+ + (cdspos - cdslength));
+ }
+ // locate the new end boundary of final exon as endxon
+ endxon = exon[x + 1] - cdspos + cdslength;
+ break;
+ }
+ }
+
+ if (sxpos != -1)
+ {
+ // and trim the exon interval set if necessary
+ int[] nxon = new int[sxpos + 2];
+ System.arraycopy(exon, 0, nxon, 0, sxpos + 2);
+ nxon[sxpos + 1] = endxon; // update the end boundary for the new exon
+ // set
+ exon = nxon;
+ }
+ }
+ return exon;
+ }
}