-/*\r
- * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)\r
- * Copyright (C) 2011 J Procter, AM Waterhouse, J Engelhardt, LM Lui, 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 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.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
-import jalview.ws.dbsources.das.api.jalviewSourceI;\r
-import jalview.ws.seqfetcher.DbSourceProxy;\r
-\r
-import java.lang.reflect.Array;\r
-import java.util.ArrayList;\r
-import java.util.Enumeration;\r
-import java.util.Hashtable;\r
-import java.util.List;\r
-import java.util.StringTokenizer;\r
-import java.util.Vector;\r
-\r
-import uk.ac.ebi.picr.model.UPEntry;\r
-\r
-/**\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
-public class DBRefFetcher implements Runnable\r
-{\r
- SequenceI[] dataset;\r
-\r
- IProgressIndicator af;\r
-\r
- CutAndPasteTransfer output = new CutAndPasteTransfer();\r
-\r
- StringBuffer sbuffer = new StringBuffer();\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
- DbSourceProxy[] dbSources;\r
-\r
- SequenceFetcher sfetcher;\r
-\r
- private SequenceI[] alseqs;\r
-\r
- public DBRefFetcher()\r
- {\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
- */\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, DbSourceProxy[] 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
- ds[i] = seqs[i];\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
- if (sources == null)\r
- {\r
- // af.featureSettings_actionPerformed(null);\r
- String[] defdb = null, otherdb = sfetcher\r
- .getDbInstances(jalview.ws.dbsources.das.datamodel.DasSequenceSource.class);\r
- List<DbSourceProxy> selsources = new ArrayList<DbSourceProxy>();\r
- Vector dasselsrc = (af.featureSettings != null) ? af.featureSettings\r
- .getSelectedSources() : new jalview.gui.DasSourceBrowser()\r
- .getSelectedSources();\r
- Enumeration<jalviewSourceI> en = dasselsrc.elements();\r
- while (en.hasMoreElements())\r
- {\r
- jalviewSourceI src = en.nextElement();\r
- List<DbSourceProxy> sp=src.getSequenceSourceProxies();\r
- selsources.addAll(sp);\r
- if (sp.size()>1)\r
- {\r
- Cache.log.debug("Added many Db Sources for :"+src.getTitle());\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
- List<DbSourceProxy> srces=new ArrayList<DbSourceProxy>();\r
- for (String ddb:defdb) {\r
- srces.addAll(sfetcher.getSourceProxy(ddb));\r
- }\r
- \r
- // append the selected sequence sources to the default dbs\r
- srces.addAll(selsources);\r
- dbSources = srces.toArray(new DbSourceProxy[0]);\r
- }\r
- else\r
- {\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 DbSourceProxy[0];\r
- }\r
- // append additional sources\r
- DbSourceProxy[] otherdb=sfetcher\r
- .getDbSourceProxyInstances(jalview.ws.dbsources.das.datamodel.DasSequenceSource.class);\r
- if (otherdb != null && otherdb.length > 0)\r
- {\r
- DbSourceProxy[] newsrc = new DbSourceProxy[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
- /**\r
- * start the fetcher thread\r
- * \r
- * @param waitTillFinished\r
- * true to block until the fetcher has finished\r
- */\r
- public void fetchDBRefs(boolean waitTillFinished)\r
- {\r
- Thread thread = new Thread(this);\r
- thread.start();\r
- running = true;\r
-\r
- if (waitTillFinished)\r
- {\r
- while (running)\r
- {\r
- try\r
- {\r
- Thread.sleep(500);\r
- } catch (Exception ex)\r
- {\r
- }\r
- }\r
- }\r
- }\r
-\r
- /**\r
- * The sequence will be added to a vector of sequences belonging to key which\r
- * could be either seq name or dbref id\r
- * \r
- * @param seq\r
- * SequenceI\r
- * @param key\r
- * String\r
- */\r
- void addSeqId(SequenceI seq, String key)\r
- {\r
- key = key.toUpperCase();\r
-\r
- Vector seqs;\r
- if (seqRefs.containsKey(key))\r
- {\r
- seqs = (Vector) seqRefs.get(key);\r
-\r
- if (seqs != null && !seqs.contains(seq))\r
- {\r
- seqs.addElement(seq);\r
- }\r
- else if (seqs == null)\r
- {\r
- seqs = new Vector();\r
- seqs.addElement(seq);\r
- }\r
-\r
- }\r
- else\r
- {\r
- seqs = new Vector();\r
- seqs.addElement(seq);\r
- }\r
-\r
- seqRefs.put(key, seqs);\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- */\r
- public void run()\r
- {\r
- if (dbSources == null)\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
- 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
- sdataset.addElement(dataset[s]);\r
- }\r
- while (sdataset.size() > 0 && db < dbSources.length)\r
- {\r
- int maxqlen = 1; // default number of queries made to at one time\r
- System.err.println("Verifying against " + dbSources[db].getDbName());\r
- boolean dn = false;\r
-\r
- // iterate through db for each remaining un-verified sequence\r
- SequenceI[] currSeqs = new SequenceI[sdataset.size()];\r
- sdataset.copyInto(currSeqs);// seqs that are to be validated against\r
- // dbSources[db]\r
- Vector queries = new Vector(); // generated queries curSeq\r
- seqRefs = new Hashtable();\r
-\r
- int seqIndex = 0;\r
-\r
- jalview.ws.seqfetcher.DbSourceProxy dbsource = dbSources[db];\r
- {\r
- // for moment, we dumbly iterate over all retrieval sources for a particular database\r
- // TODO: introduce multithread multisource queries and logic to remove a query from other sources if any source for a database returns a record\r
- if (dbsource.getDbSourceProperties().containsKey(\r
- DBRefSource.MULTIACC))\r
- {\r
- maxqlen = ((Integer) dbsource.getDbSourceProperties().get(\r
- DBRefSource.MULTIACC)).intValue();\r
- }\r
- else\r
- {\r
- maxqlen = 1;\r
- }\r
- while (queries.size() > 0 || seqIndex < currSeqs.length)\r
- {\r
- if (queries.size() > 0)\r
- {\r
- // Still queries to make for current seqIndex\r
- StringBuffer queryString = new StringBuffer("");\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(0);\r
- if (dbsource.isValidReference(query))\r
- {\r
- queryString.append((numq == 0) ? "" : dbsource\r
- .getAccessionSeparator());\r
- queryString.append(query);\r
- numq++;\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\r
- .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
- transferReferences(sdataset, dbsource.getDbSource(), retrieved);\r
- }\r
- }\r
- else\r
- {\r
- // make some more strings for use as queries\r
- for (int i = 0; (seqIndex < dataset.length) && (i < 50); seqIndex++, i++)\r
- {\r
- SequenceI sequence = dataset[seqIndex];\r
- DBRefEntry[] uprefs = jalview.util.DBRefUtils.selectRefs(\r
- sequence.getDBRef(), new String[]\r
- { dbsource.getDbSource() }); // jalview.datamodel.DBRefSource.UNIPROT\r
- // });\r
- // check for existing dbrefs to use\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.addElement(uprefs[j].getAccessionId()\r
- .toUpperCase());\r
- }\r
- }\r
- else\r
- {\r
- // generate queries from sequence ID string\r
- StringTokenizer st = new StringTokenizer(\r
- sequence.getName(), "|");\r
- while (st.hasMoreTokens())\r
- {\r
- String token = st.nextToken();\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.getUPIForAccession(token, null,\r
- picrClient.getMappedDatabaseNames(), null,\r
- true);\r
- } catch (Exception e)\r
- {\r
- System.err.println("Exception with Picr for '"\r
- + token + "'\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
- }\r
- }\r
- }\r
- // advance to next database\r
- db++;\r
- } // all databases have been queries.\r
- if (sbuffer.length() > 0)\r
- {\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
-\r
- }\r
-\r
- af.setProgressBar("DBRef search completed", startTime);\r
- // promptBeforeBlast();\r
-\r
- running = false;\r
-\r
- }\r
-\r
- /**\r
- * Verify local sequences in seqRefs against the retrieved sequence database\r
- * records.\r
- * \r
- */\r
- void transferReferences(Vector sdataset, String dbSource,\r
- AlignmentI retrievedAl) // File\r
- // file)\r
- {\r
- if (retrievedAl == null || retrievedAl.getHeight() == 0)\r
- {\r
- return;\r
- }\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
- int i, iSize = retrieved.length; // entries == null ? 0 : entries.size();\r
- // UniprotEntry entry;\r
- for (i = 0; i < iSize; i++)\r
- {\r
- SequenceI entry = retrieved[i]; // (UniprotEntry) entries.elementAt(i);\r
-\r
- // Work out which sequences this sequence matches,\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(\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
- // match up on accessionId\r
- if (seqRefs.containsKey(accessionId.toUpperCase()))\r
- {\r
- Vector seqs = (Vector) seqRefs.get(accessionId);\r
- for (int jj = 0; jj < seqs.size(); jj++)\r
- {\r
- sequence = (SequenceI) seqs.elementAt(jj);\r
- if (!sequenceMatches.contains(sequence))\r
- {\r
- sequenceMatches.addElement(sequence);\r
- }\r
- }\r
- }\r
- }\r
- if (sequenceMatches.size() == 0)\r
- {\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
- {\r
- Enumeration sqe = sqs.elements();\r
- while (sqe.hasMoreElements())\r
- {\r
- sequenceMatches.addElement(sqe.nextElement());\r
- }\r
- }\r
- }\r
- }\r
- // look for corresponding names\r
- // this is uniprot specific ?\r
- // could be useful to extend this so we try to find any 'significant'\r
- // information in common between two sequence objects.\r
- /*\r
- * DBRefEntry[] entryRefs =\r
- * jalview.util.DBRefUtils.selectRefs(entry.getDBRef(), new String[] {\r
- * dbSource }); for (int j = 0; j < entry.getName().size(); j++) { String\r
- * name = entry.getName().elementAt(j).toString(); if\r
- * (seqRefs.containsKey(name)) { Vector seqs = (Vector) seqRefs.get(name);\r
- * for (int jj = 0; jj < seqs.size(); jj++) { sequence = (SequenceI)\r
- * seqs.elementAt(jj); if (!sequenceMatches.contains(sequence)) {\r
- * sequenceMatches.addElement(sequence); } } } }\r
- */\r
- // sequenceMatches now contains the set of all sequences associated with\r
- // the returned db record\r
- String entrySeq = entry.getSequenceAsString().toUpperCase();\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\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
- sequence.getSequenceAsString()).toUpperCase();\r
-\r
- int absStart = entrySeq.indexOf(nonGapped);\r
- int mapStart = entry.getStart();\r
- jalview.datamodel.Mapping mp;\r
-\r
- if (absStart == -1)\r
- {\r
- // Is local sequence contained in dataset sequence?\r
- absStart = nonGapped.indexOf(entrySeq);\r
- if (absStart == -1)\r
- { // verification failed.\r
- messages.append(sequence.getName()\r
- + " SEQUENCE NOT %100 MATCH \n");\r
- continue;\r
- }\r
- transferred = true;\r
- sbuffer.append(sequence.getName() + " HAS " + absStart\r
- + " PREFIXED RESIDUES COMPARED TO " + dbSource + "\n");\r
- //\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,\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
- // region of database\r
- // sequence\r
- mp = null; // Mapping.getIdentityMap();\r
- // new Mapping(null,\r
- // new int[] { absStart+sequence.getStart(),\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)\r
- {\r
- if (sequence.getSequenceFeatures() != null)\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\r
- // are\r
- // to be shifted by\r
- for (int sfi = 0; sfi < sf.length; sfi++)\r
- {\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
- }\r
-\r
- System.out.println("Adding dbrefs to " + sequence.getName()\r
- + " from " + dbSource + " sequence : " + entry.getName());\r
- sequence.transferAnnotation(entry, mp);\r
- // unknownSequences.remove(sequence);\r
- int absEnd = absStart + nonGapped.length();\r
- absStart += 1;\r
- if (updateRefFrame)\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
+/*
+ * 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.ws;
+
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Enumeration;
+import java.util.HashMap;
+import java.util.Hashtable;
+import java.util.List;
+import java.util.Map;
+import java.util.StringTokenizer;
+import java.util.Vector;
+
+import jalview.analysis.AlignSeq;
+import jalview.api.FeatureSettingsModelI;
+import jalview.bin.Cache;
+import jalview.datamodel.AlignmentI;
+import jalview.datamodel.DBRefEntry;
+import jalview.datamodel.DBRefSource;
+import jalview.datamodel.Mapping;
+import jalview.datamodel.SequenceI;
+import jalview.gui.CutAndPasteTransfer;
+import jalview.gui.Desktop;
+import jalview.gui.FeatureSettings;
+import jalview.gui.IProgressIndicator;
+import jalview.gui.OOMWarning;
+import jalview.util.DBRefUtils;
+import jalview.util.MessageManager;
+import jalview.ws.seqfetcher.DbSourceProxy;
+import uk.ac.ebi.picr.model.UPEntry;
+import uk.ac.ebi.www.picr.AccessionMappingService.AccessionMapperServiceLocator;
+
+/**
+ * Implements a runnable for validating a sequence against external databases
+ * and then propagating references and features onto the sequence(s)
+ *
+ * @author $author$
+ * @version $Revision$
+ */
+public class DBRefFetcher implements Runnable
+{
+ private static final String NEWLINE = System.lineSeparator();
+
+ public static final String TRIM_RETRIEVED_SEQUENCES = "TRIM_FETCHED_DATASET_SEQS";
+
+ public interface FetchFinishedListenerI
+ {
+ void finished();
+ }
+
+ SequenceI[] dataset;
+
+ IProgressIndicator progressWindow;
+
+ CutAndPasteTransfer output = new CutAndPasteTransfer();
+
+ /**
+ * picr client instance
+ */
+ uk.ac.ebi.www.picr.AccessionMappingService.AccessionMapperInterface picrClient = null;
+
+ // This will be a collection of Vectors of sequenceI refs.
+ // The key will be the seq name or accession id of the seq
+ Hashtable<String, Vector<SequenceI>> seqRefs;
+
+ DbSourceProxy[] dbSources;
+
+ SequenceFetcher sfetcher;
+
+ private List<FetchFinishedListenerI> listeners;
+
+ private SequenceI[] alseqs;
+
+ /*
+ * when true - retrieved sequences will be trimmed to cover longest derived
+ * alignment sequence
+ */
+ private boolean trimDsSeqs = true;
+
+ /**
+ * Creates a new DBRefFetcher object and fetches from the currently selected
+ * set of databases, if this is null then it fetches based on feature settings
+ *
+ * @param seqs
+ * fetch references for these SequenceI array
+ * @param progressIndicatorFrame
+ * the frame for progress bar monitoring
+ * @param sources
+ * array of DbSourceProxy to query references form
+ * @param featureSettings
+ * FeatureSettings to get alternative DbSourceProxy from
+ * @param isNucleotide
+ * indicates if the array of SequenceI are Nucleotides or not
+ */
+ public DBRefFetcher(SequenceI[] seqs,
+ IProgressIndicator progressIndicatorFrame,
+ DbSourceProxy[] sources, FeatureSettings featureSettings,
+ boolean isNucleotide)
+ {
+ listeners = new ArrayList<>();
+ this.progressWindow = progressIndicatorFrame;
+ alseqs = new SequenceI[seqs.length];
+ SequenceI[] ds = new SequenceI[seqs.length];
+ for (int i = 0; i < seqs.length; i++)
+ {
+ alseqs[i] = seqs[i];
+ if (seqs[i].getDatasetSequence() != null)
+ {
+ ds[i] = seqs[i].getDatasetSequence();
+ }
+ else
+ {
+ ds[i] = seqs[i];
+ }
+ }
+ this.dataset = ds;
+ // TODO Jalview 2.5 lots of this code should be in the gui package!
+ sfetcher = jalview.gui.SequenceFetcher.getSequenceFetcherSingleton();
+ // set default behaviour for transferring excess sequence data to the
+ // dataset
+ trimDsSeqs = Cache.getDefault(TRIM_RETRIEVED_SEQUENCES, true);
+ if (sources == null)
+ {
+ setDatabaseSources(featureSettings, isNucleotide);
+ }
+ else
+ {
+ // we assume the caller knows what they're doing and ensured that all the
+ // db source names are valid
+ dbSources = sources;
+ }
+ }
+
+ /**
+ * Helper method to configure the list of database sources to query
+ *
+ * @param featureSettings
+ * @param forNucleotide
+ */
+ void setDatabaseSources(FeatureSettings featureSettings,
+ boolean forNucleotide)
+ {
+ // af.featureSettings_actionPerformed(null);
+ String[] defdb = null;
+ List<DbSourceProxy> selsources = new ArrayList<>();
+ // select appropriate databases based on alignFrame context.
+ if (forNucleotide)
+ {
+ defdb = DBRefSource.DNACODINGDBS;
+ }
+ else
+ {
+ defdb = DBRefSource.PROTEINDBS;
+ }
+ List<DbSourceProxy> srces = new ArrayList<>();
+ for (String ddb : defdb)
+ {
+ List<DbSourceProxy> srcesfordb = sfetcher.getSourceProxy(ddb);
+ if (srcesfordb != null)
+ {
+ for (DbSourceProxy src : srcesfordb)
+ {
+ if (!srces.contains(src))
+ {
+ srces.addAll(srcesfordb);
+ }
+ }
+ }
+ }
+ // append the PDB data source, since it is 'special', catering for both
+ // nucleotide and protein
+ // srces.addAll(sfetcher.getSourceProxy(DBRefSource.PDB));
+
+ srces.addAll(selsources);
+ dbSources = srces.toArray(new DbSourceProxy[srces.size()]);
+ }
+
+ /**
+ * Constructor with only sequences provided
+ *
+ * @param sequences
+ */
+ public DBRefFetcher(SequenceI[] sequences)
+ {
+ this(sequences, null, null, null, false);
+ }
+
+ /**
+ * Add a listener to be notified when sequence fetching is complete
+ *
+ * @param l
+ */
+ public void addListener(FetchFinishedListenerI l)
+ {
+ listeners.add(l);
+ }
+
+ /**
+ * start the fetcher thread
+ *
+ * @param waitTillFinished
+ * true to block until the fetcher has finished
+ */
+ public void fetchDBRefs(boolean waitTillFinished)
+ {
+ if (waitTillFinished)
+ {
+ run();
+ }
+ else
+ {
+ new Thread(this).start();
+ }
+ }
+
+ /**
+ * The sequence will be added to a vector of sequences belonging to key which
+ * could be either seq name or dbref id
+ *
+ * @param seq
+ * SequenceI
+ * @param key
+ * String
+ */
+ void addSeqId(SequenceI seq, String key)
+ {
+ key = key.toUpperCase();
+
+ Vector<SequenceI> seqs;
+ if (seqRefs.containsKey(key))
+ {
+ seqs = seqRefs.get(key);
+
+ if (seqs != null && !seqs.contains(seq))
+ {
+ seqs.addElement(seq);
+ }
+ else if (seqs == null)
+ {
+ seqs = new Vector<>();
+ seqs.addElement(seq);
+ }
+
+ }
+ else
+ {
+ seqs = new Vector<>();
+ seqs.addElement(seq);
+ }
+
+ seqRefs.put(key, seqs);
+ }
+
+ /**
+ * DOCUMENT ME!
+ */
+ @Override
+ public void run()
+ {
+ if (dbSources == null)
+ {
+ throw new Error(MessageManager
+ .getString("error.implementation_error_must_init_dbsources"));
+ }
+ long startTime = System.currentTimeMillis();
+ if (progressWindow != null)
+ {
+ progressWindow.setProgressBar(
+ MessageManager.getString("status.fetching_db_refs"),
+ startTime);
+ }
+ try
+ {
+ if (Cache.getDefault("DBREFFETCH_USEPICR", false))
+ {
+ picrClient = new AccessionMapperServiceLocator()
+ .getAccessionMapperPort();
+ }
+ } catch (Exception e)
+ {
+ System.err.println("Couldn't locate PICR service instance.\n");
+ e.printStackTrace();
+ }
+
+ Vector<SequenceI> sdataset = new Vector<>(
+ Arrays.asList(dataset));
+ List<String> warningMessages = new ArrayList<>();
+
+ // clear any old feature display settings recorded from past sessions
+ featureDisplaySettings = null;
+
+ int db = 0;
+ while (sdataset.size() > 0 && db < dbSources.length)
+ {
+ int maxqlen = 1; // default number of queries made at one time
+ System.out.println("Verifying against " + dbSources[db].getDbName());
+
+ // iterate through db for each remaining un-verified sequence
+ SequenceI[] currSeqs = new SequenceI[sdataset.size()];
+ sdataset.copyInto(currSeqs);// seqs that are to be validated against
+ // dbSources[db]
+ Vector<String> queries = new Vector<>(); // generated queries curSeq
+ seqRefs = new Hashtable<>();
+
+ int seqIndex = 0;
+
+ DbSourceProxy dbsource = dbSources[db];
+ // for moment, we dumbly iterate over all retrieval sources for a
+ // particular database
+ // TODO: introduce multithread multisource queries and logic to remove a
+ // query from other sources if any source for a database returns a
+ // record
+ maxqlen = dbsource.getMaximumQueryCount();
+
+ while (queries.size() > 0 || seqIndex < currSeqs.length)
+ {
+ if (queries.size() > 0)
+ {
+ // Still queries to make for current seqIndex
+ StringBuffer queryString = new StringBuffer("");
+ int numq = 0;
+ int nqSize = (maxqlen > queries.size()) ? queries.size()
+ : maxqlen;
+
+ while (queries.size() > 0 && numq < nqSize)
+ {
+ String query = queries.elementAt(0);
+ if (dbsource.isValidReference(query))
+ {
+ queryString.append(
+ (numq == 0) ? "" : dbsource.getAccessionSeparator());
+ queryString.append(query);
+ numq++;
+ }
+ // remove the extracted query string
+ queries.removeElementAt(0);
+ }
+ // make the queries and process the response
+ AlignmentI retrieved = null;
+ try
+ {
+ if (Cache.log.isDebugEnabled())
+ {
+ Cache.log.debug("Querying " + dbsource.getDbName()
+ + " with : '" + queryString.toString() + "'");
+ }
+ retrieved = dbsource.getSequenceRecords(queryString.toString());
+ } catch (Exception ex)
+ {
+ ex.printStackTrace();
+ } catch (OutOfMemoryError err)
+ {
+ new OOMWarning("retrieving database references ("
+ + queryString.toString() + ")", err);
+ }
+ if (retrieved != null)
+ {
+ transferReferences(sdataset, dbsource, retrieved,
+ trimDsSeqs, warningMessages);
+ }
+ }
+ else
+ {
+ // make some more strings for use as queries
+ for (int i = 0; (seqIndex < dataset.length)
+ && (i < 50); seqIndex++, i++)
+ {
+ SequenceI sequence = dataset[seqIndex];
+ List<DBRefEntry> uprefs = DBRefUtils
+ .selectRefs(sequence.getDBRefs(), new String[]
+ { dbsource.getDbSource() }); // jalview.datamodel.DBRefSource.UNIPROT
+ // });
+ // check for existing dbrefs to use
+ if (uprefs != null && uprefs.size() > 0)
+ {
+ for (int j = 0, n = uprefs.size(); j < n; j++)
+ {
+ DBRefEntry upref = uprefs.get(j);
+ addSeqId(sequence, upref.getAccessionId());
+ queries.addElement(
+ upref.getAccessionId().toUpperCase());
+ }
+ }
+ else
+ {
+ // generate queries from sequence ID string
+ StringTokenizer st = new StringTokenizer(sequence.getName(),
+ "|");
+ while (st.hasMoreTokens())
+ {
+ String token = st.nextToken();
+ UPEntry[] presp = null;
+ if (picrClient != null)
+ {
+ // resolve the string against PICR to recover valid IDs
+ try
+ {
+ presp = picrClient.getUPIForAccession(token, null,
+ picrClient.getMappedDatabaseNames(), null,
+ true);
+ } catch (Exception e)
+ {
+ System.err.println(
+ "Exception with Picr for '" + token + "'\n");
+ e.printStackTrace();
+ }
+ }
+ if (presp != null && presp.length > 0)
+ {
+ for (int id = 0; id < presp.length; id++)
+ {
+ // construct sequences from response if sequences are
+ // present, and do a transferReferences
+ // otherwise transfer non sequence x-references directly.
+ }
+ System.out.println(
+ "Validated ID against PICR... (for what its worth):"
+ + token);
+ addSeqId(sequence, token);
+ queries.addElement(token.toUpperCase());
+ }
+ else
+ {
+ // if ()
+ // System.out.println("Not querying source with
+ // token="+token+"\n");
+ addSeqId(sequence, token);
+ queries.addElement(token.toUpperCase());
+ }
+ }
+ }
+ }
+ }
+ }
+ // advance to next database
+ db++;
+ } // all databases have been queried
+ if (!warningMessages.isEmpty())
+ {
+ StringBuilder sb = new StringBuilder(warningMessages.size() * 30);
+ sb.append(MessageManager
+ .getString("label.your_sequences_have_been_verified"));
+ for (String msg : warningMessages)
+ {
+ sb.append(msg).append(NEWLINE);
+ }
+ output.setText(sb.toString());
+
+ Desktop.addInternalFrame(output,
+ MessageManager.getString("label.sequences_updated"), 600,
+ 300);
+ // The above is the dataset, we must now find out the index
+ // of the viewed sequence
+
+ }
+ if (progressWindow != null)
+ {
+ progressWindow.setProgressBar(
+ MessageManager.getString("label.dbref_search_completed"),
+ startTime);
+ }
+
+ for (FetchFinishedListenerI listener : listeners)
+ {
+ listener.finished();
+ }
+ }
+
+ /**
+ * Verify local sequences in seqRefs against the retrieved sequence database
+ * records. Returns true if any sequence was modified as a result (start/end
+ * changed and/or sequence enlarged), else false.
+ *
+ * @param sdataset
+ * dataset sequences we are retrieving for
+ * @param dbSource
+ * database source we are retrieving from
+ * @param retrievedAl
+ * retrieved sequences as alignment
+ * @param trimDatasetSeqs
+ * if true, sequences will not be enlarged to match longer retrieved
+ * sequences, only their start/end adjusted
+ * @param warningMessages
+ * a list of messages to add to
+ */
+ boolean transferReferences(Vector<SequenceI> sdataset,
+ DbSourceProxy dbSourceProxy,
+ AlignmentI retrievedAl, boolean trimDatasetSeqs,
+ List<String> warningMessages)
+ {
+ // System.out.println("trimming ? " + trimDatasetSeqs);
+ if (retrievedAl == null || retrievedAl.getHeight() == 0)
+ {
+ return false;
+ }
+
+ String dbSource = dbSourceProxy.getDbName();
+ boolean modified = false;
+ SequenceI[] retrieved = recoverDbSequences(
+ retrievedAl.getSequencesArray());
+ SequenceI sequence = null;
+
+ for (SequenceI retrievedSeq : retrieved)
+ {
+ // Work out which sequences this sequence matches,
+ // taking into account all accessionIds and names in the file
+ Vector<SequenceI> sequenceMatches = new Vector<>();
+ // look for corresponding accession ids
+ List<DBRefEntry> entryRefs = DBRefUtils
+ .selectRefs(retrievedSeq.getDBRefs(), new String[]
+ { dbSource });
+ if (entryRefs == null)
+ {
+ System.err
+ .println("Dud dbSource string ? no entryrefs selected for "
+ + dbSource + " on " + retrievedSeq.getName());
+ continue;
+ }
+ for (int j = 0, n = entryRefs.size(); j < n; j++)
+ {
+ DBRefEntry ref = entryRefs.get(j);
+ String accessionId = ref.getAccessionId();
+ // match up on accessionId
+ if (seqRefs.containsKey(accessionId.toUpperCase()))
+ {
+ Vector<SequenceI> seqs = seqRefs.get(accessionId);
+ for (int jj = 0; jj < seqs.size(); jj++)
+ {
+ sequence = seqs.elementAt(jj);
+ if (!sequenceMatches.contains(sequence))
+ {
+ sequenceMatches.addElement(sequence);
+ }
+ }
+ }
+ }
+ if (sequenceMatches.isEmpty())
+ {
+ // failed to match directly on accessionId==query so just compare all
+ // sequences to entry
+ Enumeration<String> e = seqRefs.keys();
+ while (e.hasMoreElements())
+ {
+ Vector<SequenceI> sqs = seqRefs.get(e.nextElement());
+ if (sqs != null && sqs.size() > 0)
+ {
+ Enumeration<SequenceI> sqe = sqs.elements();
+ while (sqe.hasMoreElements())
+ {
+ sequenceMatches.addElement(sqe.nextElement());
+ }
+ }
+ }
+ }
+ // look for corresponding names
+ // this is uniprot specific ?
+ // could be useful to extend this so we try to find any 'significant'
+ // information in common between two sequence objects.
+ /*
+ * List<DBRefEntry> entryRefs =
+ * jalview.util.DBRefUtils.selectRefs(entry.getDBRef(), new String[] {
+ * dbSource }); for (int j = 0; j < entry.getName().size(); j++) { String
+ * name = entry.getName().elementAt(j).toString(); if
+ * (seqRefs.containsKey(name)) { Vector seqs = (Vector) seqRefs.get(name);
+ * for (int jj = 0; jj < seqs.size(); jj++) { sequence = (SequenceI)
+ * seqs.elementAt(jj); if (!sequenceMatches.contains(sequence)) {
+ * sequenceMatches.addElement(sequence); } } } }
+ */
+ if (sequenceMatches.size() > 0)
+ {
+ addFeatureSettings(dbSourceProxy);
+ }
+ // sequenceMatches now contains the set of all sequences associated with
+ // the returned db record
+ final String retrievedSeqString = retrievedSeq.getSequenceAsString();
+ String entrySeq = retrievedSeqString.toUpperCase();
+ for (int m = 0; m < sequenceMatches.size(); m++)
+ {
+ sequence = sequenceMatches.elementAt(m);
+ // only update start and end positions and shift features if there are
+ // no existing references
+ // TODO: test for legacy where uniprot or EMBL refs exist but no
+ // mappings are made (but content matches retrieved set)
+ boolean updateRefFrame = sequence.getDBRefs() == null
+ || sequence.getDBRefs().size() == 0;
+ // TODO:
+ // verify sequence against the entry sequence
+
+ Mapping mp;
+ final int sequenceStart = sequence.getStart();
+
+ boolean remoteEnclosesLocal = false;
+ String nonGapped = AlignSeq
+ .extractGaps("-. ", sequence.getSequenceAsString())
+ .toUpperCase();
+ int absStart = entrySeq.indexOf(nonGapped);
+ if (absStart == -1)
+ {
+ // couldn't find local sequence in sequence from database, so check if
+ // the database sequence is a subsequence of local sequence
+ absStart = nonGapped.indexOf(entrySeq);
+ if (absStart == -1)
+ {
+ // verification failed. couldn't find any relationship between
+ // entrySeq and local sequence
+ // messages suppressed as many-to-many matches are confusing
+ // String msg = sequence.getName()
+ // + " Sequence not 100% match with "
+ // + retrievedSeq.getName();
+ // addWarningMessage(warningMessages, msg);
+ continue;
+ }
+ /*
+ * retrieved sequence is a proper subsequence of local sequence
+ */
+ String msg = sequence.getName() + " has " + absStart
+ + " prefixed residues compared to "
+ + retrievedSeq.getName();
+ addWarningMessage(warningMessages, msg);
+
+ /*
+ * So create a mapping to the external entry from the matching region of
+ * the local sequence, and leave local start/end untouched.
+ */
+ mp = new Mapping(null,
+ new int[]
+ { sequenceStart + absStart,
+ sequenceStart + absStart + entrySeq.length() - 1 },
+ new int[]
+ { retrievedSeq.getStart(),
+ retrievedSeq.getStart() + entrySeq.length() - 1 },
+ 1, 1);
+ updateRefFrame = false;
+ }
+ else
+ {
+ /*
+ * local sequence is a subsequence of (or matches) retrieved sequence
+ */
+ remoteEnclosesLocal = true;
+ mp = null;
+
+ if (updateRefFrame)
+ {
+ /*
+ * relocate existing sequence features by offset
+ */
+ int startShift = absStart - sequenceStart + 1;
+ if (startShift != 0)
+ {
+ modified |= sequence.getFeatures().shiftFeatures(1,
+ startShift);
+ }
+ }
+ }
+
+ System.out.println("Adding dbrefs to " + sequence.getName()
+ + " from " + dbSource + " sequence : "
+ + retrievedSeq.getName());
+ sequence.transferAnnotation(retrievedSeq, mp);
+
+ absStart += retrievedSeq.getStart();
+ int absEnd = absStart + nonGapped.length() - 1;
+ if (!trimDatasetSeqs)
+ {
+ /*
+ * update start position and/or expand to longer retrieved sequence
+ */
+ if (!retrievedSeqString.equals(sequence.getSequenceAsString())
+ && remoteEnclosesLocal)
+ {
+ sequence.setSequence(retrievedSeqString);
+ modified = true;
+ addWarningMessage(warningMessages,
+ "Sequence for " + sequence.getName() + " expanded from "
+ + retrievedSeq.getName());
+ }
+ if (sequence.getStart() != retrievedSeq.getStart())
+ {
+ sequence.setStart(retrievedSeq.getStart());
+ modified = true;
+ if (absStart != sequenceStart)
+ {
+ addWarningMessage(warningMessages,
+ "Start/end position for " + sequence.getName()
+ + " updated from " + retrievedSeq.getName());
+ }
+ }
+ }
+ if (updateRefFrame)
+ {
+ // finally, update local sequence reference frame if we're allowed
+ if (trimDatasetSeqs)
+ {
+ // just fix start/end
+ if (sequence.getStart() != absStart
+ || sequence.getEnd() != absEnd)
+ {
+ sequence.setStart(absStart);
+ sequence.setEnd(absEnd);
+ modified = true;
+ addWarningMessage(warningMessages,
+ "Start/end for " + sequence.getName()
+ + " updated from " + retrievedSeq.getName());
+ }
+ }
+ // search for alignment sequences to update coordinate frame for
+ for (int alsq = 0; alsq < alseqs.length; alsq++)
+ {
+ if (alseqs[alsq].getDatasetSequence() == sequence)
+ {
+ String ngAlsq = AlignSeq
+ .extractGaps("-. ",
+ alseqs[alsq].getSequenceAsString())
+ .toUpperCase();
+ int oldstrt = alseqs[alsq].getStart();
+ alseqs[alsq].setStart(sequence.getSequenceAsString()
+ .toUpperCase().indexOf(ngAlsq) + sequence.getStart());
+ if (oldstrt != alseqs[alsq].getStart())
+ {
+ alseqs[alsq].setEnd(
+ ngAlsq.length() + alseqs[alsq].getStart() - 1);
+ modified = true;
+ }
+ }
+ }
+ // TODO: search for all other references to this dataset sequence, and
+ // update start/end
+ // TODO: update all AlCodonMappings which involve this alignment
+ // sequence (e.g. Q30167 cdna translation from exon2 product (vamsas
+ // demo)
+ }
+ // and remove it from the rest
+ // TODO: decide if we should remove annotated sequence from set
+ sdataset.remove(sequence);
+ }
+ }
+ return modified;
+ }
+
+ Map<String, FeatureSettingsModelI> featureDisplaySettings = null;
+
+ private void addFeatureSettings(DbSourceProxy dbSourceProxy)
+ {
+ FeatureSettingsModelI fsettings = dbSourceProxy
+ .getFeatureColourScheme();
+ if (fsettings != null)
+ {
+ if (featureDisplaySettings == null)
+ {
+ featureDisplaySettings = new HashMap<>();
+ }
+ featureDisplaySettings.put(dbSourceProxy.getDbName(), fsettings);
+ }
+ }
+
+ /**
+ *
+ * @return any feature settings associated with sources that have provided sequences
+ */
+ public List<FeatureSettingsModelI>getFeatureSettingsModels()
+ {
+ return featureDisplaySettings == null
+ ? Arrays.asList(new FeatureSettingsModelI[0])
+ : Arrays.asList(featureDisplaySettings.values()
+ .toArray(new FeatureSettingsModelI[1]));
+ }
+ /**
+ * Adds the message to the list unless it already contains it
+ *
+ * @param messageList
+ * @param msg
+ */
+ void addWarningMessage(List<String> messageList, String msg)
+ {
+ if (!messageList.contains(msg))
+ {
+ messageList.add(msg);
+ }
+ }
+
+ /**
+ * loop thru and collect additional sequences in Map.
+ *
+ * @param sequencesArray
+ * @return
+ */
+ private SequenceI[] recoverDbSequences(SequenceI[] sequencesArray)
+ {
+ int n;
+ if (sequencesArray == null || (n = sequencesArray.length) == 0)
+ return sequencesArray;
+ ArrayList<SequenceI> nseq = new ArrayList<>();
+ for (int i = 0;i < n; i++)
+ {
+ nseq.add(sequencesArray[i]);
+ List<DBRefEntry> dbr = sequencesArray[i].getDBRefs();
+ Mapping map = null;
+ if (dbr != null)
+ {
+ for (int r = 0, rn = dbr.size(); r < rn; r++)
+ {
+ if ((map = dbr.get(r).getMap()) != null)
+ {
+ if (map.getTo() != null && !nseq.contains(map.getTo()))
+ {
+ nseq.add(map.getTo());
+ }
+ }
+ }
+ }
+ }
+ // BH 2019.01.25 question here if this is the right logic. Return the original if nothing found?
+ if (nseq.size() > 0)
+ {
+ return nseq.toArray(new SequenceI[nseq.size()]);
+ }
+ return sequencesArray;
+ }
+}