X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fio%2FDBRefFetcher.java;h=bd3c73323de37914f913ad59e54ff34fb550b361;hb=dd74fc4938723fe5ec48d4e5fdcfbe58ac42a48d;hp=c11910b2c69e010a2e7492875af681fbe57a38d1;hpb=8e728f4ac9c2dc718f97db2e34fc3571a9622f94;p=jalview.git

diff --git a/src/jalview/io/DBRefFetcher.java b/src/jalview/io/DBRefFetcher.java
index c11910b..bd3c733 100644
--- a/src/jalview/io/DBRefFetcher.java
+++ b/src/jalview/io/DBRefFetcher.java
@@ -1,6 +1,6 @@
 /*
 * Jalview - A Sequence Alignment Editor and Viewer
-* Copyright (C) 2005 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
+* Copyright (C) 2006 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
@@ -41,13 +41,15 @@ import jalview.analysis.AlignSeq;
  */
 public class DBRefFetcher implements Runnable
 {
-  AlignmentI align;
   AlignmentI dataset;
   AlignFrame af;
-  ArrayList unknownSequences;
   CutAndPasteTransfer output = new CutAndPasteTransfer();
   StringBuffer sbuffer = new StringBuffer();
-  boolean uniprotFlag = false;
+  boolean running = false;
+
+  ///This will be a collection of Vectors of sequenceI refs.
+  //The key will be the seq name or accession id of the seq
+  Hashtable seqRefs;
 
   public DBRefFetcher()
   {}
@@ -66,7 +68,6 @@ public class DBRefFetcher implements Runnable
       Unmarshaller unmar = new Unmarshaller(uni);
       unmar.setIgnoreExtraElements(true);
       unmar.setMapping(map);
-   //   unmar.setDebug(true);
 
       uni = (UniprotFile) unmar.unmarshal(new FileReader(file));
     }
@@ -88,14 +89,64 @@ public class DBRefFetcher implements Runnable
   public DBRefFetcher(AlignmentI align, AlignFrame af)
   {
     this.af = af;
-    unknownSequences = new ArrayList();
-    this.align = align;
     this.dataset = align.getDataset();
+  }
 
+  public boolean fetchDBRefs(boolean waitTillFinished)
+  {
     Thread thread = new Thread(this);
     thread.start();
+    running = true;
+
+    if(waitTillFinished)
+    {
+      while(running)
+      {
+        try{
+          Thread.sleep(500);
+        }catch(Exception ex){}
+      }
+    }
+
+    return true;
+  }
+
+  /**
+   * 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 seqs;
+    if(seqRefs.containsKey(key))
+    {
+      seqs = (Vector)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!
    */
@@ -103,6 +154,9 @@ public class DBRefFetcher implements Runnable
   {
     long startTime = System.currentTimeMillis();
     af.setProgressBar("Fetching db refs", startTime);
+    running = true;
+
+    seqRefs = new Hashtable();
 
     try
     {
@@ -111,58 +165,52 @@ public class DBRefFetcher implements Runnable
 
       while (seqIndex < sequences.size())
       {
-        Vector ids = new Vector();
+        StringBuffer queryString = new StringBuffer("uniprot:");
 
         for (int i = 0; (seqIndex < sequences.size()) && (i < 50);
              seqIndex++, i++)
         {
           Sequence sequence = (Sequence) sequences.get(seqIndex);
-          Vector uprefs = jalview.util.DBRefUtils.selectRefs(sequence.getDBRef(), new String[] {
+          DBRefEntry [] uprefs = jalview.util.DBRefUtils.selectRefs(sequence.getDBRef(), new String[] {
               jalview.datamodel.DBRefSource.UNIPROT});
           if (uprefs!=null)
           {
-            // we know the id for this entry, so don't note its ID in the unknownSequences list
-            for (int j=0,k=uprefs.size(); j<k; j++)
-              ids.add(((DBRefEntry) uprefs.get(j)).getAccessionId());
-            unknownSequences.add(sequence);
-          } else {
-            if (!ids.contains(sequence.getName()))
+            if(uprefs.length+i>50)
+              break;
+
+            for(int j=0; j<uprefs.length; j++)
             {
-              ids.add(sequence.getName());
-              unknownSequences.add(sequence);
+              addSeqId(sequence, uprefs[j].getAccessionId());
+              queryString.append(uprefs[j].getAccessionId()+";");
+            }
+          }
+          else
+          {
+            StringTokenizer st = new StringTokenizer(sequence.getName(), "|");
+            if(st.countTokens()+i>50)
+            {
+              //Dont send more than 50 id strings to dbFetch!!
+              seqIndex --;
+            }
+            else
+            {
+                while(st.hasMoreTokens())
+                {
+                  String token = st.nextToken();
+                  addSeqId(sequence, token);
+                  queryString.append(token+";");
+                }
             }
           }
         }
 
         ///////////////////////////////////
         ///READ FROM EBI
-        if (ids.size() > 0)
+        EBIFetchClient ebi = new EBIFetchClient();
+        File file = ebi.fetchDataAsFile(queryString.toString(), "xml", "raw");
+        if (file != null)
         {
-          StringBuffer remainingIds = new StringBuffer("uniprot:");
-          for (int i = 0; i < ids.size(); i++)
-           {
-             if(ids.get(i).toString().indexOf("|")>-1)
-             {
-               remainingIds.append(ids.get(i).toString().substring(
-                   ids.get(i).toString().lastIndexOf("|") + 1));
-               uniprotFlag = true;
-             }
-             else
-               remainingIds.append(ids.get(i));
-
-             remainingIds.append(";");
-           }
-
-          EBIFetchClient ebi = new EBIFetchClient();
-          File file = ebi.fetchDataAsFile(remainingIds.toString(),
-                                          "xml", "raw");
-
-
-
-          if (file != null)
-          {
-            ReadUniprotFile(file, ids);
-          }
+          ReadUniprotFile(file);
         }
       }
     }
@@ -187,31 +235,11 @@ public class DBRefFetcher implements Runnable
     af.setProgressBar("DBRef search completed", startTime);
    // promptBeforeBlast();
 
-  }
-
+   running = false;
 
-  void promptBeforeBlast()
-   {
-     // This must be outside the run() body as java 1.5
-     // will not return any value from the OptionPane to the expired thread.
-      if (unknownSequences.size() > 0)
-      {
-       // int reply = javax.swing.JOptionPane.showConfirmDialog(
-       //     Desktop.desktop, "Couldn't find a match for "+unknownSequences.size()+" sequences."
-        //        +"\nPerform blast for unknown sequences?",
-        //            "Blast for Unidentified Sequences",
-        //             javax.swing.JOptionPane.YES_NO_OPTION, javax.swing.JOptionPane.QUESTION_MESSAGE);
-     javax.swing.JOptionPane.showMessageDialog(
-    Desktop.desktop, "Couldn't find a match for "+unknownSequences.size()+" sequences.",
-            "Unidentified Sequences",
-             javax.swing.JOptionPane.WARNING_MESSAGE);
-
-
-      //  if(reply == javax.swing.JOptionPane.YES_OPTION)
-     //    new WSWUBlastClient(ap, align, unknownSequences);
-      }
   }
 
+
   /**
    * DOCUMENT ME!
    *
@@ -219,77 +247,83 @@ public class DBRefFetcher implements Runnable
    * @param out DOCUMENT ME!
    * @param align DOCUMENT ME!
    */
-  void ReadUniprotFile(File file, Vector ids)
+  void ReadUniprotFile(File file)
   {
-    if(!file.exists())
+    if (!file.exists())
       return;
 
     SequenceI sequence = null;
 
     Vector entries = getUniprotEntries(file);
 
-    int i, iSize = entries==null?0:entries.size();
+    int i, iSize = entries == null ? 0 : entries.size();
     UniprotEntry entry;
     for (i = 0; i < iSize; i++)
     {
       entry = (UniprotEntry) entries.elementAt(i);
-      String idmatch = entry.getAccession().elementAt(0).toString();
-      sequence = dataset.findName(idmatch);
 
-      if (sequence == null)
+      //Work out which sequences this Uniprot file has matches to,
+      //taking into account all accessionIds and names in the file
+      Vector sequenceMatches = new Vector();
+      for (int j = 0; j < entry.getAccession().size(); j++)
       {
-        //Sequence maybe Name, not Accession
-        idmatch = entry.getName().elementAt(0).toString();
-        sequence = dataset.findName(idmatch);
+        String accessionId = entry.getAccession().elementAt(j).toString();
+        if (seqRefs.containsKey(accessionId))
+        {
+          Vector seqs = (Vector) seqRefs.get(accessionId);
+          for (int jj = 0; jj < seqs.size(); jj++)
+          {
+            sequence = (SequenceI) seqs.elementAt(jj);
+            if (!sequenceMatches.contains(sequence))
+              sequenceMatches.addElement(sequence);
+          }
+        }
       }
-
-      if(sequence!=null)
-        ids.remove(sequence.getName());
-
-      else  if (sequence == null && uniprotFlag)
+      for (int j = 0; j < entry.getName().size(); j++)
       {
-          StringBuffer upid = new StringBuffer("UniProt/Swiss-Prot|");
-          for(int u=0; u<entry.getAccession().size(); u++)
-            upid.append(entry.getAccession().elementAt(u)+"|");
-
-          sequence = dataset.findName(upid+idmatch);
-          ids.remove(idmatch);
+        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(sequence ==null)
+      for (int m = 0; m < sequenceMatches.size(); m++)
       {
-        System.out.println(idmatch+" not found");
-        continue;
-      }
-
-
-      sequence.addDBRef(new DBRefEntry(DBRefSource.UNIPROT,
-                                       "1.0",
-                                       entry.getAccession().elementAt(0).toString()));
+        sequence = (SequenceI) sequenceMatches.elementAt(m);
+        sequence.addDBRef(new DBRefEntry(DBRefSource.UNIPROT,
+                                         "0",
+                                         entry.getAccession().elementAt(0).
+                                         toString()));
 
-      System.out.println("Adding dbref to "+sequence.getName()+" : "+
-                         entry.getAccession().elementAt(0).toString());
+        System.out.println("Adding dbref to " + sequence.getName() + " : " +
+                           entry.getAccession().elementAt(0).toString());
 
-      String nonGapped = AlignSeq.extractGaps("-. ", sequence.getSequence());
+        String nonGapped = AlignSeq.extractGaps("-. ",
+                                                sequence.getSequenceAsString()).
+            toUpperCase();
 
-      int absStart = entry.getUniprotSequence().getContent().indexOf(
-          nonGapped.toString());
-
-      if (absStart == -1)
-      {
-        // Is UniprotSequence contained in dataset sequence?
-        absStart = nonGapped.toString().indexOf(entry.getUniprotSequence().getContent());
-        if(absStart == -1)
-        {
-          sbuffer.append(sequence.getName() +
-                         " SEQUENCE NOT %100 MATCH \n");
+        int absStart = entry.getUniprotSequence().getContent().indexOf(
+            nonGapped.toString());
 
-          continue;
-        }
-        else
+        if (absStart == -1)
         {
+          // Is UniprotSequence contained in dataset sequence?
+          absStart = nonGapped.toString().indexOf(entry.getUniprotSequence().
+                                                  getContent());
+          if (absStart == -1)
+          {
+            sbuffer.append(sequence.getName() + " SEQUENCE NOT %100 MATCH \n");
+            continue;
+          }
 
-          if(entry.getFeature()!=null)
+          if (entry.getFeature() != null)
           {
             Enumeration e = entry.getFeature().elements();
             while (e.hasMoreElements())
@@ -298,77 +332,41 @@ public class DBRefFetcher implements Runnable
               sf.setBegin(sf.getBegin() + absStart + 1);
               sf.setEnd(sf.getEnd() + absStart + 1);
             }
+
+            sbuffer.append(sequence.getName() +
+                           " HAS " + absStart +
+                           " PREFIXED RESIDUES COMPARED TO UNIPROT - ANY SEQUENCE FEATURES"
+                           + " HAVE BEEN ADJUSTED ACCORDINGLY \n");
+            absStart = 0;
           }
 
-          sbuffer.append(sequence.getName() +
-                         " HAS "+absStart+" PREFIXED RESIDUES COMPARED TO UNIPROT - ANY SEQUENCE FEATURES"
-                        +" HAVE BEEN ADJUSTED ACCORDINGLY \n");
-          absStart = 0;
         }
 
-      }
+        //unknownSequences.remove(sequence);
 
-      unknownSequences.remove(sequence);
-
-      int absEnd = absStart + nonGapped.toString().length();
-      absStart += 1;
-
-      Enumeration e = entry.getDbReference().elements();
-      Vector onlyPdbEntries = new Vector();
-      while(e.hasMoreElements())
-      {
-        PDBEntry pdb = (PDBEntry)e.nextElement();
-        if(!pdb.getType().equals("PDB"))
-          continue;
-
-        onlyPdbEntries.addElement(pdb);
-      }
+        int absEnd = absStart + nonGapped.toString().length();
+        absStart += 1;
 
-      sequence.setPDBId(onlyPdbEntries);
-
-      if (entry.getFeature()!=null) {
-        //e = entry.getFeature().elements();
-      //  while (e.hasMoreElements())
+        Enumeration e = entry.getDbReference().elements();
+        Vector onlyPdbEntries = new Vector();
+        while (e.hasMoreElements())
         {
-        //  SequenceFeature sf = (SequenceFeature) e.nextElement();
-         // sf.setFeatureGroup("Uniprot");
-        //  sequence.addSequenceFeature( sf );
-        }
-      }
-      sequence.setStart(absStart);
-      sequence.setEnd(absEnd);
+          PDBEntry pdb = (PDBEntry) e.nextElement();
+          if (!pdb.getType().equals(DBRefSource.PDB))
+            continue;
 
+          sequence.addDBRef(new DBRefEntry(DBRefSource.PDB,
+                                           "0",
+                                           pdb.getId()));
 
-      int n = 0;
-      SequenceI seq2;
-      while (n < align.getHeight())
-      {
-        //This loop enables multiple sequences with the same
-        //id to have features added and seq limits updated
-        seq2 = align.getSequenceAt(n);
-        if (seq2.getName().equals(idmatch))
-        {
-          nonGapped = AlignSeq.extractGaps("-. ", seq2.getSequence());
-
-          absStart = sequence.getSequence().indexOf(nonGapped);
-          absEnd = absStart + nonGapped.toString().length() - 1;
+          onlyPdbEntries.addElement(pdb);
+        }
 
-          // This is the Viewd alignment sequences
-          // No need to tell the user of the dataset updates
-          if ( (seq2.getStart() != absStart+sequence.getStart())
-             || (seq2.getEnd() != absEnd+sequence.getStart()))
-          {
-            sbuffer.append("Updated: " + seq2.getName() + " " +
-                           seq2.getStart() + "/" + seq2.getEnd() +
-                           "  to  " + (absStart + sequence.getStart()) + "/" +
-                           (absEnd + sequence.getStart()) + "\n");
+        sequence.setPDBId(onlyPdbEntries);
 
-            seq2.setStart(absStart + sequence.getStart());
-            seq2.setEnd(absEnd + sequence.getStart());
-          }
-        }
+        sequence.setStart(absStart);
+        sequence.setEnd(absEnd);
 
-        n++;
       }
     }
   }