/*\r
- * Jalview - A Sequence Alignment Editor and Viewer\r
- * Copyright (C) 2007 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle\r
- *\r
- * This program is free software; you can redistribute it and/or\r
- * modify it under the terms of the GNU General Public License\r
- * as published by the Free Software Foundation; either version 2\r
- * of the License, or (at your option) any later version.\r
- *\r
- * This program is distributed in the hope that it will be useful,\r
- * but WITHOUT ANY WARRANTY; without even the implied warranty of\r
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\r
- * GNU General Public License for more details.\r
- *\r
- * You should have received a copy of the GNU General Public License\r
- * along with this program; if not, write to the Free Software\r
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA\r
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)\r
+ * Copyright (C) 2011 J Procter, AM Waterhouse, G Barton, M Clamp, S Searle\r
+ * \r
+ * This file is part of Jalview.\r
+ * \r
+ * Jalview is free software: you can redistribute it and/or\r
+ * modify it under the terms of the GNU General Public License \r
+ * as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.\r
+ * \r
+ * Jalview is distributed in the hope that it will be useful, but \r
+ * WITHOUT ANY WARRANTY; without even the implied warranty \r
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR \r
+ * PURPOSE. See the GNU General Public License for more details.\r
+ * \r
+ * You should have received a copy of the GNU General Public License along with Jalview. If not, see <http://www.gnu.org/licenses/>.\r
*/\r
package jalview.ws;\r
\r
-import java.io.*;\r
-import java.util.*;\r
-\r
-import org.exolab.castor.mapping.*;\r
-import org.exolab.castor.xml.*;\r
-import jalview.analysis.*;\r
-import jalview.datamodel.*;\r
+import jalview.analysis.AlignSeq;\r
+import jalview.bin.Cache;\r
+import jalview.datamodel.AlignmentI;\r
+import jalview.datamodel.DBRefEntry;\r
+import jalview.datamodel.DBRefSource;\r
import jalview.datamodel.Mapping;\r
-import jalview.gui.*;\r
-import jalview.ws.dbsources.Uniprot;\r
-import jalview.ws.ebi.EBIFetchClient;\r
+import jalview.datamodel.SequenceFeature;\r
+import jalview.datamodel.SequenceI;\r
+import jalview.gui.AlignFrame;\r
+import jalview.gui.CutAndPasteTransfer;\r
+import jalview.gui.Desktop;\r
+import jalview.gui.IProgressIndicator;\r
+import jalview.gui.OOMWarning;\r
+\r
+import java.lang.reflect.Array;\r
+import java.util.Enumeration;\r
+import java.util.Hashtable;\r
+import java.util.StringTokenizer;\r
+import java.util.Vector;\r
+\r
+import org.biojava.dasobert.dasregistry.DasSource;\r
+\r
+import uk.ac.ebi.picr.model.UPEntry;\r
\r
/**\r
- * Implements a runnable for validating a sequence \r
- * against external databases and then propagating \r
- * references and features onto the sequence(s)\r
+ * Implements a runnable for validating a sequence against external databases\r
+ * and then propagating references and features onto the sequence(s)\r
* \r
* @author $author$\r
* @version $Revision$\r
\r
boolean running = false;\r
\r
+ /**\r
+ * picr client instance\r
+ */\r
+ uk.ac.ebi.www.picr.AccessionMappingService.AccessionMapperInterface picrClient = null;\r
+\r
// /This will be a collection of Vectors of sequenceI refs.\r
// The key will be the seq name or accession id of the seq\r
Hashtable seqRefs;\r
\r
SequenceFetcher sfetcher;\r
\r
+ private SequenceI[] alseqs;\r
+\r
public DBRefFetcher()\r
{\r
}\r
\r
/**\r
- * Creates a new SequenceFeatureFetcher object.\r
+ * Creates a new SequenceFeatureFetcher object and fetches from the currently\r
+ * selected set of databases.\r
* \r
* @param seqs\r
- * fetch references for these sequences\r
+ * fetch references for these sequences\r
* @param af\r
- * the parent alignframe for progress bar monitoring.\r
+ * the parent alignframe for progress bar monitoring.\r
*/\r
public DBRefFetcher(SequenceI[] seqs, AlignFrame af)\r
{\r
+ this(seqs, af, null);\r
+ }\r
+\r
+ /**\r
+ * Creates a new SequenceFeatureFetcher object and fetches from the currently\r
+ * selected set of databases.\r
+ * \r
+ * @param seqs\r
+ * fetch references for these sequences\r
+ * @param af\r
+ * the parent alignframe for progress bar monitoring.\r
+ * @param sources\r
+ * array of database source strings to query references from\r
+ */\r
+ public DBRefFetcher(SequenceI[] seqs, AlignFrame af, String[] sources)\r
+ {\r
this.af = af;\r
+ alseqs = new SequenceI[seqs.length];\r
SequenceI[] ds = new SequenceI[seqs.length];\r
for (int i = 0; i < seqs.length; i++)\r
{\r
+ alseqs[i] = seqs[i];\r
if (seqs[i].getDatasetSequence() != null)\r
ds[i] = seqs[i].getDatasetSequence();\r
else\r
}\r
this.dataset = ds;\r
// TODO Jalview 2.5 lots of this code should be in the gui package!\r
- sfetcher = jalview.gui.SequenceFetcher.getSequenceFetcherSingleton(af); \r
- // select appropriate databases based on alignFrame context.\r
- if (af.getViewport().getAlignment().isNucleotide())\r
+ sfetcher = jalview.gui.SequenceFetcher.getSequenceFetcherSingleton(af);\r
+ if (sources == null)\r
{\r
- dbSources = DBRefSource.DNACODINGDBS;\r
+ // af.featureSettings_actionPerformed(null);\r
+ String[] defdb = null, otherdb = sfetcher\r
+ .getDbInstances(jalview.ws.dbsources.DasSequenceSource.class);\r
+ Vector selsources = new Vector(), dasselsrc = (af.featureSettings != null) ? af.featureSettings\r
+ .getSelectedSources() : new jalview.gui.DasSourceBrowser()\r
+ .getSelectedSources();\r
+ Enumeration en = dasselsrc.elements();\r
+ while (en.hasMoreElements())\r
+ {\r
+ DasSource src = (DasSource) en.nextElement();\r
+ selsources.addElement(src.getNickname());\r
+ }\r
+ int osel = 0;\r
+ for (int o = 0; otherdb != null && o < otherdb.length; o++)\r
+ {\r
+ if (!selsources.contains(otherdb[o]))\r
+ {\r
+ otherdb[o] = null;\r
+ }\r
+ else\r
+ {\r
+ osel++;\r
+ }\r
+ }\r
+ // select appropriate databases based on alignFrame context.\r
+ if (af.getViewport().getAlignment().isNucleotide())\r
+ {\r
+ defdb = DBRefSource.DNACODINGDBS;\r
+ }\r
+ else\r
+ {\r
+ defdb = DBRefSource.PROTEINDBS;\r
+ }\r
+ // append the selected sequence sources to the default dbs\r
+ dbSources = new String[defdb.length + osel];\r
+ System.arraycopy(defdb, 0, dbSources, 0, defdb.length);\r
+ for (int o = 0, op = defdb.length; otherdb != null\r
+ && o < otherdb.length; o++)\r
+ {\r
+ if (otherdb[o] != null)\r
+ {\r
+ dbSources[op++] = otherdb[o];\r
+ }\r
+ }\r
}\r
else\r
{\r
- dbSources = DBRefSource.PROTEINDBS;\r
+ // we assume the caller knows what they're doing and ensured that all the\r
+ // db source names are valid\r
+ dbSources = sources;\r
+ }\r
+ }\r
+\r
+ /**\r
+ * retrieve all the das sequence sources and add them to the list of db\r
+ * sources to retrieve from\r
+ */\r
+ public void appendAllDasSources()\r
+ {\r
+ if (dbSources == null)\r
+ {\r
+ dbSources = new String[]\r
+ {};\r
+ }\r
+ // append additional sources\r
+ String[] otherdb = sfetcher\r
+ .getDbInstances(jalview.ws.dbsources.DasSequenceSource.class);\r
+ if (otherdb != null && otherdb.length > 0)\r
+ {\r
+ String[] newsrc = new String[dbSources.length + otherdb.length];\r
+ System.arraycopy(dbSources, 0, newsrc, 0, dbSources.length);\r
+ System.arraycopy(otherdb, 0, newsrc, dbSources.length, otherdb.length);\r
+ dbSources = newsrc;\r
}\r
}\r
\r
* start the fetcher thread\r
* \r
* @param waitTillFinished\r
- * true to block until the fetcher has finished\r
+ * true to block until the fetcher has finished\r
*/\r
public void fetchDBRefs(boolean waitTillFinished)\r
{\r
* could be either seq name or dbref id\r
* \r
* @param seq\r
- * SequenceI\r
+ * SequenceI\r
* @param key\r
- * String\r
+ * String\r
*/\r
void addSeqId(SequenceI seq, String key)\r
{\r
{\r
throw new Error("Implementation error. Must initialise dbSources");\r
}\r
+ running = true;\r
long startTime = System.currentTimeMillis();\r
af.setProgressBar("Fetching db refs", startTime);\r
- running = true;\r
+ try\r
+ {\r
+ if (Cache.getDefault("DBREFFETCH_USEPICR", false))\r
+ {\r
+ picrClient = new uk.ac.ebi.www.picr.AccessionMappingService.AccessionMapperServiceLocator()\r
+ .getAccessionMapperPort();\r
+ }\r
+ } catch (Exception e)\r
+ {\r
+ System.err.println("Couldn't locate PICR service instance.\n");\r
+ e.printStackTrace();\r
+ }\r
int db = 0;\r
Vector sdataset = new Vector();\r
for (int s = 0; s < dataset.length; s++)\r
{\r
maxqlen = ((Integer) dbsource.getDbSourceProperties().get(\r
DBRefSource.MULTIACC)).intValue();\r
- } else {\r
- maxqlen=1;\r
+ }\r
+ else\r
+ {\r
+ maxqlen = 1;\r
}\r
// iterate through db for each remaining un-verified sequence\r
SequenceI[] currSeqs = new SequenceI[sdataset.size()];\r
{\r
// Still queries to make for current seqIndex\r
StringBuffer queryString = new StringBuffer("");\r
- int nqSize = (maxqlen > queries.size()) ? queries.size()\r
- : maxqlen;\r
- for (int nq = 0, numq = 0; nq < nqSize; nq++)\r
+ int numq = 0, nqSize = (maxqlen > queries.size()) ? queries\r
+ .size() : maxqlen;\r
+\r
+ while (queries.size() > 0 && numq < nqSize)\r
{\r
- String query = (String) queries.elementAt(nq);\r
+ String query = (String) queries.elementAt(0);\r
if (dbsource.isValidReference(query))\r
{\r
- queryString.append((nq == 0) ? "" : dbsource\r
+ queryString.append((numq == 0) ? "" : dbsource\r
.getAccessionSeparator());\r
queryString.append(query);\r
numq++;\r
}\r
- }\r
- for (int nq = 0; nq < nqSize; nq++)\r
- {\r
+ // remove the extracted query string\r
queries.removeElementAt(0);\r
}\r
// make the queries and process the response\r
AlignmentI retrieved = null;\r
try\r
{\r
+ if (jalview.bin.Cache.log.isDebugEnabled())\r
+ {\r
+ jalview.bin.Cache.log.debug("Querying "\r
+ + dbsource.getDbName() + " with : '"\r
+ + queryString.toString() + "'");\r
+ }\r
retrieved = dbsource.getSequenceRecords(queryString.toString());\r
} catch (Exception ex)\r
{\r
ex.printStackTrace();\r
+ } catch (OutOfMemoryError err)\r
+ {\r
+ new OOMWarning("retrieving database references ("\r
+ + queryString.toString() + ")", err);\r
}\r
if (retrieved != null)\r
{\r
{ dbSources[db] }); // jalview.datamodel.DBRefSource.UNIPROT\r
// });\r
// check for existing dbrefs to use\r
- if (uprefs != null)\r
+ if (uprefs != null && uprefs.length > 0)\r
{\r
for (int j = 0; j < uprefs.length; j++)\r
{\r
addSeqId(sequence, uprefs[j].getAccessionId());\r
- queries\r
- .addElement(uprefs[j].getAccessionId()\r
- .toUpperCase());\r
+ queries.addElement(uprefs[j].getAccessionId().toUpperCase());\r
}\r
}\r
else\r
while (st.hasMoreTokens())\r
{\r
String token = st.nextToken();\r
- addSeqId(sequence, token);\r
- queries.addElement(token.toUpperCase());\r
+ UPEntry[] presp = null;\r
+ if (picrClient != null)\r
+ {\r
+ // resolve the string against PICR to recover valid IDs\r
+ try\r
+ {\r
+ presp = picrClient\r
+ .getUPIForAccession(token, null,\r
+ picrClient.getMappedDatabaseNames(),\r
+ null, true);\r
+ } catch (Exception e)\r
+ {\r
+ System.err.println("Exception with Picr for '" + token\r
+ + "'\n");\r
+ e.printStackTrace();\r
+ }\r
+ }\r
+ if (presp != null && presp.length > 0)\r
+ {\r
+ for (int id = 0; id < presp.length; id++)\r
+ {\r
+ // construct sequences from response if sequences are\r
+ // present, and do a transferReferences\r
+ // otherwise transfer non sequence x-references directly.\r
+ }\r
+ System.out\r
+ .println("Validated ID against PICR... (for what its worth):"\r
+ + token);\r
+ addSeqId(sequence, token);\r
+ queries.addElement(token.toUpperCase());\r
+ }\r
+ else\r
+ {\r
+ // if ()\r
+ // System.out.println("Not querying source with token="+token+"\n");\r
+ addSeqId(sequence, token);\r
+ queries.addElement(token.toUpperCase());\r
+ }\r
}\r
}\r
}\r
} // all databases have been queries.\r
if (sbuffer.length() > 0)\r
{\r
- output\r
- .setText("Your sequences have been verified against known sequence databases. Some of the ids have been\n"\r
- + "altered, most likely the start/end residue will have been updated.\n"\r
- + "Save your alignment to maintain the updated id.\n\n"\r
- + sbuffer.toString());\r
+ output.setText("Your sequences have been verified against known sequence databases. Some of the ids have been\n"\r
+ + "altered, most likely the start/end residue will have been updated.\n"\r
+ + "Save your alignment to maintain the updated id.\n\n"\r
+ + sbuffer.toString());\r
Desktop.addInternalFrame(output, "Sequence names updated ", 600, 300);\r
// The above is the dataset, we must now find out the index\r
// of the viewed sequence\r
AlignmentI retrievedAl) // File\r
// file)\r
{\r
-\r
if (retrievedAl == null || retrievedAl.getHeight() == 0)\r
{\r
return;\r
}\r
- SequenceI[] retrieved = retrievedAl.getSequencesArray();\r
+ SequenceI[] retrieved = recoverDbSequences(retrievedAl\r
+ .getSequencesArray());\r
SequenceI sequence = null;\r
+ boolean transferred = false;\r
+ StringBuffer messages = new StringBuffer();\r
\r
// Vector entries = new Uniprot().getUniprotEntries(file);\r
\r
// taking into account all accessionIds and names in the file\r
Vector sequenceMatches = new Vector();\r
// look for corresponding accession ids\r
- DBRefEntry[] entryRefs = jalview.util.DBRefUtils.selectRefs(entry\r
- .getDBRef(), new String[]\r
- { dbSource });\r
+ DBRefEntry[] entryRefs = jalview.util.DBRefUtils.selectRefs(\r
+ entry.getDBRef(), new String[]\r
+ { dbSource });\r
+ if (entryRefs == null)\r
+ {\r
+ System.err\r
+ .println("Dud dbSource string ? no entryrefs selected for "\r
+ + dbSource + " on " + entry.getName());\r
+ continue;\r
+ }\r
for (int j = 0; j < entryRefs.length; j++)\r
{\r
String accessionId = entryRefs[j].getAccessionId(); // .getAccession().elementAt(j).toString();\r
}\r
}\r
}\r
- if (sequenceMatches.size()==0)\r
+ if (sequenceMatches.size() == 0)\r
{\r
- // failed to match directly on accessionId==query so just compare all sequences to entry\r
+ // failed to match directly on accessionId==query so just compare all\r
+ // sequences to entry\r
Enumeration e = seqRefs.keys();\r
while (e.hasMoreElements())\r
{\r
Vector sqs = (Vector) seqRefs.get(e.nextElement());\r
- if (sqs!=null && sqs.size()>0)\r
+ if (sqs != null && sqs.size() > 0)\r
{\r
Enumeration sqe = sqs.elements();\r
while (sqe.hasMoreElements())\r
for (int m = 0; m < sequenceMatches.size(); m++)\r
{\r
sequence = (SequenceI) sequenceMatches.elementAt(m);\r
- // only update start and end positions and shift features if there are no existing references\r
- // TODO: test for legacy where uniprot or EMBL refs exist but no mappings are made (but content matches retrieved set)\r
- boolean updateRefFrame = sequence.getDBRef()==null || sequence.getDBRef().length==0;\r
+ // only update start and end positions and shift features if there are\r
+ // no existing references\r
+ // TODO: test for legacy where uniprot or EMBL refs exist but no\r
+ // mappings are made (but content matches retrieved set)\r
+ boolean updateRefFrame = sequence.getDBRef() == null\r
+ || sequence.getDBRef().length == 0;\r
// verify sequence against the entry sequence\r
\r
String nonGapped = AlignSeq.extractGaps("-. ",\r
absStart = nonGapped.indexOf(entrySeq);\r
if (absStart == -1)\r
{ // verification failed.\r
- sbuffer.append(sequence.getName()\r
+ messages.append(sequence.getName()\r
+ " SEQUENCE NOT %100 MATCH \n");\r
continue;\r
}\r
- \r
+ transferred = true;\r
sbuffer.append(sequence.getName() + " HAS " + absStart\r
- + " PREFIXED RESIDUES COMPARED TO " + dbSource+"\n");\r
+ + " PREFIXED RESIDUES COMPARED TO " + dbSource + "\n");\r
//\r
- // + " - ANY SEQUENCE FEATURES"\r
- // + " HAVE BEEN ADJUSTED ACCORDINGLY \n");\r
+ // + " - ANY SEQUENCE FEATURES"\r
+ // + " HAVE BEEN ADJUSTED ACCORDINGLY \n");\r
// absStart = 0;\r
// create valid mapping between matching region of local sequence and\r
// the mapped sequence\r
mp = new Mapping(null, new int[]\r
- { sequence.getStart()+absStart, sequence.getStart()+absStart+entrySeq.length()-1 }, new int[]\r
- { entry.getStart(),\r
- entry.getStart() + entrySeq.length() - 1 }, 1, 1);\r
- updateRefFrame=false; // mapping is based on current start/end so don't modify start and end\r
+ { sequence.getStart() + absStart,\r
+ sequence.getStart() + absStart + entrySeq.length() - 1 },\r
+ new int[]\r
+ { entry.getStart(),\r
+ entry.getStart() + entrySeq.length() - 1 }, 1, 1);\r
+ updateRefFrame = false; // mapping is based on current start/end so\r
+ // don't modify start and end\r
}\r
else\r
{\r
+ transferred = true;\r
// update start and end of local sequence to place it in entry's\r
// reference frame.\r
// apply identity map map from whole of local sequence to matching\r
// absStart+sequence.getStart()+entrySeq.length()-1},\r
// new int[] { entry.getStart(), entry.getEnd() }, 1, 1);\r
// relocate local features for updated start\r
- if (updateRefFrame && sequence.getSequenceFeatures() != null)\r
+ if (updateRefFrame)\r
{\r
- SequenceFeature[] sf = sequence.getSequenceFeatures();\r
- int start = sequence.getStart();\r
- int end = sequence.getEnd();\r
- int startShift = 1-absStart-start; // how much the features are to be shifted by\r
- for (int sfi = 0; sfi < sf.length; sfi++)\r
+ if (sequence.getSequenceFeatures() != null)\r
{\r
- if (sf[sfi].getBegin() >= start && sf[sfi].getEnd() <= end)\r
+ SequenceFeature[] sf = sequence.getSequenceFeatures();\r
+ int start = sequence.getStart();\r
+ int end = sequence.getEnd();\r
+ int startShift = 1 - absStart - start; // how much the features\r
+ // are\r
+ // to be shifted by\r
+ for (int sfi = 0; sfi < sf.length; sfi++)\r
{\r
- // shift feature along by absstart\r
- sf[sfi].setBegin(sf[sfi].getBegin() + startShift);\r
- sf[sfi].setEnd(sf[sfi].getEnd() + startShift);\r
+ if (sf[sfi].getBegin() >= start && sf[sfi].getEnd() <= end)\r
+ {\r
+ // shift feature along by absstart\r
+ sf[sfi].setBegin(sf[sfi].getBegin() + startShift);\r
+ sf[sfi].setEnd(sf[sfi].getEnd() + startShift);\r
+ }\r
}\r
}\r
}\r
// finally, update local sequence reference frame if we're allowed\r
sequence.setStart(absStart);\r
sequence.setEnd(absEnd);\r
+ // search for alignment sequences to update coordinate frame for\r
+ for (int alsq = 0; alsq < alseqs.length; alsq++)\r
+ {\r
+ if (alseqs[alsq].getDatasetSequence() == sequence)\r
+ {\r
+ String ngAlsq = AlignSeq.extractGaps("-. ",\r
+ alseqs[alsq].getSequenceAsString()).toUpperCase();\r
+ int oldstrt = alseqs[alsq].getStart();\r
+ alseqs[alsq].setStart(sequence.getSequenceAsString()\r
+ .toUpperCase().indexOf(ngAlsq)\r
+ + sequence.getStart());\r
+ if (oldstrt != alseqs[alsq].getStart())\r
+ {\r
+ alseqs[alsq].setEnd(ngAlsq.length()\r
+ + alseqs[alsq].getStart() - 1);\r
+ }\r
+ }\r
+ }\r
+ // TODO: search for all other references to this dataset sequence, and\r
+ // update start/end\r
+ // TODO: update all AlCodonMappings which involve this alignment\r
+ // sequence (e.g. Q30167 cdna translation from exon2 product (vamsas\r
+ // demo)\r
}\r
// and remove it from the rest\r
// TODO: decide if we should remove annotated sequence from set\r
sdataset.remove(sequence);\r
+ // TODO: should we make a note of sequences that have received new DB\r
+ // ids, so we can query all enabled DAS servers for them ?\r
}\r
}\r
+ if (!transferred)\r
+ {\r
+ // report the ID/sequence mismatches\r
+ sbuffer.append(messages);\r
+ }\r
+ }\r
+\r
+ /**\r
+ * loop thru and collect additional sequences in Map.\r
+ * \r
+ * @param sequencesArray\r
+ * @return\r
+ */\r
+ private SequenceI[] recoverDbSequences(SequenceI[] sequencesArray)\r
+ {\r
+ Vector nseq = new Vector();\r
+ for (int i = 0; sequencesArray != null && i < sequencesArray.length; i++)\r
+ {\r
+ nseq.addElement(sequencesArray[i]);\r
+ DBRefEntry dbr[] = sequencesArray[i].getDBRef();\r
+ jalview.datamodel.Mapping map = null;\r
+ for (int r = 0; (dbr != null) && r < dbr.length; r++)\r
+ {\r
+ if ((map = dbr[r].getMap()) != null)\r
+ {\r
+ if (map.getTo() != null && !nseq.contains(map.getTo()))\r
+ {\r
+ nseq.addElement(map.getTo());\r
+ }\r
+ }\r
+ }\r
+ }\r
+ if (nseq.size() > 0)\r
+ {\r
+ sequencesArray = new SequenceI[nseq.size()];\r
+ nseq.toArray(sequencesArray);\r
+ }\r
+ return sequencesArray;\r
}\r
}\r
git://source.jalview.org / jalview.git / blobdiff