Hashtable for previously loaded pdb files
[jalview.git] / src / jalview / gui / SequenceFetcher.java
index cb41e6e..ce366c8 100755 (executable)
@@ -1,6 +1,6 @@
 /*
  * Jalview - A Sequence Alignment Editor and Viewer
- * Copyright (C) 2006 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle
+ * Copyright (C) 2007 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
  */
 package jalview.gui;
 
-import javax.swing.*;
+import java.io.*;
+import java.util.*;
+
 import java.awt.*;
 import java.awt.event.*;
-import jalview.io.EBIFetchClient;
+import javax.swing.*;
+
 import MCview.*;
 import jalview.datamodel.*;
-import jalview.analysis.AlignSeq;
-import java.io.File;
 import jalview.io.*;
-import java.util.*;
 
 public class SequenceFetcher
     extends JPanel implements Runnable
@@ -58,14 +58,19 @@ public class SequenceFetcher
     frame = new JInternalFrame();
     frame.setContentPane(this);
     if (System.getProperty("os.name").startsWith("Mac"))
+    {
       Desktop.addInternalFrame(frame, getFrameTitle(), 400, 140);
+    }
     else
+    {
       Desktop.addInternalFrame(frame, getFrameTitle(), 400, 125);
+    }
   }
 
   private String getFrameTitle()
   {
-    return ( (alignFrame == null) ? "New " : "Additional ") + "Sequence Fetcher";
+    return ( (alignFrame == null) ? "New " : "Additional ") +
+        "Sequence Fetcher";
   }
 
   private void jbInit()
@@ -163,11 +168,15 @@ public class SequenceFetcher
   {
     String error = "";
     if (database.getSelectedItem().equals(noDbSelected))
+    {
       error += "Please select the source database\n";
+    }
     com.stevesoft.pat.Regex empty = new com.stevesoft.pat.Regex("\\s+", "");
     textfield.setText(empty.replaceAll(textfield.getText()));
     if (textfield.getText().length() == 0)
+    {
       error += "Please enter a (semi-colon separated list of) database id(s)";
+    }
     if (error.length() > 0)
     {
       showErrorMessage(error);
@@ -184,7 +193,7 @@ public class SequenceFetcher
              || database.getSelectedItem().equals("EMBLCDS"))
     {
       StringTokenizer st = new StringTokenizer(textfield.getText(), ";");
-      while(st.hasMoreTokens())
+      while (st.hasMoreTokens())
       {
         EBIFetchClient dbFetch = new EBIFetchClient();
 
@@ -196,11 +205,13 @@ public class SequenceFetcher
         if (reply != null)
         {
           for (int i = 0; i < reply.length; i++)
+          {
             result.append(reply[i] + "\n");
+          }
         }
       }
 
-      if(result!=null && result.length()>1) // arbitrary minimum length for a seuqence file
+      if (result != null && result.length() > 1) // arbitrary minimum length for a seuqence file
       {
         System.out.println(result.toString());
 
@@ -211,35 +222,69 @@ public class SequenceFetcher
     {
       StringTokenizer qset = new StringTokenizer(textfield.getText(), ";");
       String query;
-      while (qset.hasMoreTokens() && ((query = qset.nextToken())!=null))
+      SequenceI[] seqs = null;
+      while (qset.hasMoreTokens() && ( (query = qset.nextToken()) != null))
+      {
+        SequenceI[] seqparts = getPDBFile(query.toUpperCase());
+        if (seqparts != null)
+        {
+          if (seqs == null)
+          {
+            seqs = seqparts;
+          }
+          else
+          {
+            SequenceI[] newseqs = new SequenceI[seqs.length + seqparts.length];
+            int i = 0;
+            for (; i < seqs.length; i++)
+            {
+              newseqs[i] = seqs[i];
+              seqs[i] = null;
+            }
+            for (int j = 0; j < seqparts.length; i++, j++)
+            {
+              newseqs[i] = seqparts[j];
+            }
+            seqs = newseqs;
+          }
+          result.append("# Success for " + query.toUpperCase() + "\n");
+        }
+      }
+      if (seqs != null && seqs.length > 0)
       {
-        StringBuffer respart = getPDBFile(query.toUpperCase());
-        if(respart!=null)
-          result.append(respart);
+        if (parseResult(new Alignment(seqs), null, null) != null)
+        {
+          result.append(
+              "# Successfully parsed the PDB File Queries into an Alignment");
+        }
       }
-
-
-      if (result.length()>0)
-        parseResult(result.toString(), null);
     }
-    else if( database.getSelectedItem().equals("PFAM"))
+    else if (database.getSelectedItem().equals("PFAM"))
     {
-      try{
+      try
+      {
         result.append(new FastaFile(
-           "http://www.sanger.ac.uk/cgi-bin/Pfam/getalignment.pl?format=fal&acc="
-           +  textfield.getText().toUpperCase(), "URL").print()
-           );
+            "http://www.sanger.ac.uk/cgi-bin/Pfam/getalignment.pl?format=fal&acc="
+            + textfield.getText().toUpperCase(), "URL").print()
+            );
 
-         if(result.length()>0)
-           parseResult( result.toString(), textfield.getText().toUpperCase() );
+        if (result.length() > 0)
+        {
+          parseResult(result.toString(), textfield.getText().toUpperCase());
+        }
 
-      }catch(java.io.IOException ex)
-      {   result = null;    }
+      }
+      catch (java.io.IOException ex)
+      {
+        result = null;
+      }
     }
 
     if (result == null || result.length() == 0)
+    {
       showErrorMessage("Error retrieving " + textfield.getText()
                        + " from " + database.getSelectedItem());
+    }
 
     resetDialog();
     return;
@@ -277,7 +322,7 @@ public class SequenceFetcher
 
         if (entry.getProtein() != null)
         {
-           name.append(" " + entry.getProtein().getName().elementAt(0));
+          name.append(" " + entry.getProtein().getName().elementAt(0));
         }
 
         result.append(name + "\n" + entry.getUniprotSequence().getContent() +
@@ -296,7 +341,9 @@ public class SequenceFetcher
         {
           PDBEntry pdb = (PDBEntry) e.nextElement();
           if (!pdb.getType().equals("PDB"))
+          {
             continue;
+          }
 
           onlyPdbEntries.addElement(pdb);
         }
@@ -304,14 +351,12 @@ public class SequenceFetcher
         Enumeration en2 = entry.getAccession().elements();
         while (en2.hasMoreElements())
         {
-          al.getSequenceAt(i).getDatasetSequence().addDBRef(new DBRefEntry(DBRefSource.UNIPROT,
-                    "0",
-                    en2.nextElement().toString()));
+          al.getSequenceAt(i).getDatasetSequence().addDBRef(new DBRefEntry(
+              DBRefSource.UNIPROT,
+              "0",
+              en2.nextElement().toString()));
         }
 
-
-
-
         al.getSequenceAt(i).getDatasetSequence().setPDBId(onlyPdbEntries);
         if (entry.getFeature() != null)
         {
@@ -320,16 +365,16 @@ public class SequenceFetcher
           {
             SequenceFeature sf = (SequenceFeature) e.nextElement();
             sf.setFeatureGroup("Uniprot");
-            al.getSequenceAt(i).getDatasetSequence().addSequenceFeature( sf );
+            al.getSequenceAt(i).getDatasetSequence().addSequenceFeature(sf);
           }
         }
       }
     }
   }
 
-  StringBuffer getPDBFile(String id)
+  SequenceI[] getPDBFile(String id)
   {
-    StringBuffer result = new StringBuffer();
+    Vector result = new Vector();
     String chain = null;
     if (id.indexOf(":") > -1)
     {
@@ -338,9 +383,12 @@ public class SequenceFetcher
     }
 
     EBIFetchClient ebi = new EBIFetchClient();
-    String file = ebi.fetchDataAsFile("pdb:" + id, "pdb", "raw").getAbsolutePath();
+    String file = ebi.fetchDataAsFile("pdb:" + id, "pdb", "raw").
+        getAbsolutePath();
     if (file == null)
+    {
       return null;
+    }
     try
     {
       PDBfile pdbfile = new PDBfile(file, jalview.io.AppletFormatAdapter.FILE);
@@ -349,14 +397,35 @@ public class SequenceFetcher
         if (chain == null ||
             ( (PDBChain) pdbfile.chains.elementAt(i)).id.
             toUpperCase().equals(chain))
-
-          result.append("\n>PDB|" + id + "|" +
-                        ( (PDBChain) pdbfile.chains.elementAt(i)).sequence.
-                        getName() +
-                        "\n"
-                        +
-                        ( (PDBChain) pdbfile.chains.elementAt(i)).sequence.
-                        getSequenceAsString());
+        {
+          PDBChain pdbchain = (PDBChain) pdbfile.chains.elementAt(i);
+          // Get the Chain's Sequence - who's dataset includes any special features added from the PDB file
+          SequenceI sq = pdbchain.sequence;
+          // Specially formatted name for the PDB chain sequences retrieved from the PDB
+          sq.setName("PDB|" + id + "|" + sq.getName());
+          // Might need to add more metadata to the PDBEntry object
+          // like below
+          /*
+           * PDBEntry entry = new PDBEntry();
+                       // Construct the PDBEntry
+                       entry.setId(id);
+                       if (entry.getProperty() == null)
+              entry.setProperty(new Hashtable());
+                       entry.getProperty().put("chains",
+                      pdbchain.id
+                      + "=" + sq.getStart()
+                      + "-" + sq.getEnd());
+                       sq.getDatasetSequence().addPDBId(entry);
+           */
+          // Add PDB DB Refs
+          // We make a DBRefEtntry because we have obtained the PDB file from a verifiable source
+          // JBPNote - PDB DBRefEntry should also carry the chain and mapping information
+          DBRefEntry dbentry = new DBRefEntry(jalview.datamodel.DBRefSource.PDB,
+                                              "0", id + pdbchain.id);
+          sq.addDBRef(dbentry);
+          // and add seuqence to the retrieved set
+          result.addElement(sq.deriveSequence());
+        }
       }
     }
     catch (Exception ex) // Problem parsing PDB file
@@ -366,124 +435,93 @@ public class SequenceFetcher
                                  database.getSelectedItem(), ex);
       return null;
     }
-
-    return result;
+    SequenceI[] results = new SequenceI[result.size()];
+    for (int i = 0, j = result.size(); i < j; i++)
+    {
+      results[i] = (SequenceI) result.elementAt(i);
+      result.setElementAt(null, i);
+    }
+    return results;
   }
 
   Alignment parseResult(String result, String title)
   {
     String format = new IdentifyFile().Identify(result, "Paste");
-    Alignment al = null;
-
+    Alignment sequences = null;
     if (FormatAdapter.isValidFormat(format))
     {
+      sequences = null;
+      try
+      {
+        sequences = new FormatAdapter().readFile(result.toString(), "Paste",
+                                                 format);
+      }
+      catch (Exception ex)
+      {}
 
-      try{ al = new FormatAdapter().readFile(result.toString(), "Paste",
-                                               format);}
-      catch(Exception ex){}
+      if (sequences != null)
+      {
+        return parseResult(sequences, title, format);
+      }
+    }
+    else
+    {
+      showErrorMessage("Error retrieving " + textfield.getText()
+                       + " from " + database.getSelectedItem());
+    }
+
+    return null;
+  }
+
+  Alignment parseResult(Alignment al, String title, String currentFileFormat)
+  {
 
-      if (al != null && al.getHeight() > 0)
+    if (al != null && al.getHeight() > 0)
+    {
+      if (alignFrame == null)
       {
-        if (alignFrame == null)
+        AlignFrame af = new AlignFrame(al,
+                                       AlignFrame.DEFAULT_WIDTH,
+                                       AlignFrame.DEFAULT_HEIGHT);
+        if (currentFileFormat != null)
         {
-          AlignFrame af = new AlignFrame(al,
-                                           AlignFrame.DEFAULT_WIDTH,
-                                           AlignFrame.DEFAULT_HEIGHT
-);
-          af.currentFileFormat = format;
-          if(title==null)
-            title = "Retrieved from " + database.getSelectedItem();
-          Desktop.addInternalFrame(af,
-                                   title,
-                                   AlignFrame.DEFAULT_WIDTH,
-                                   AlignFrame.DEFAULT_HEIGHT);
-          af.statusBar.setText("Successfully pasted alignment file");
-
-          try
-          {
-            af.setMaximum(jalview.bin.Cache.getDefault("SHOW_FULLSCREEN", false));
-          }
-          catch (Exception ex)
-          {}
+          af.currentFileFormat = currentFileFormat; // WHAT IS THE DEFAULT FORMAT FOR NON-FormatAdapter Sourced Alignments?
         }
-        else
-        {
-          for (int i = 0; i < al.getHeight(); i++)
-          {
-            alignFrame.viewport.alignment.addSequence(al.getSequenceAt(i));
-
-            ////////////////////////////
-            //Dataset needs extension;
-            /////////////////////////////
-            Sequence ds = new Sequence(al.getSequenceAt(i).getName(),
-                                       AlignSeq.extractGaps("-. ",
-                al.getSequenceAt(i).getSequenceAsString()),
-                                       al.getSequenceAt(i).getStart(),
-                                       al.getSequenceAt(i).getEnd());
-            al.getSequenceAt(i).setDatasetSequence(ds);
-            alignFrame.viewport.alignment.getDataset().addSequence(ds);
-          }
-          alignFrame.viewport.setEndSeq(alignFrame.viewport.alignment.
-                                        getHeight());
-          alignFrame.viewport.alignment.getWidth();
-          alignFrame.viewport.firePropertyChange("alignment", null,
-                                                 alignFrame.viewport.
-                                                 getAlignment().getSequences());
 
+        if (title == null)
+        {
+          title = "Retrieved from " + database.getSelectedItem();
         }
 
-        if (database.getSelectedItem().equals("PDB"))
+        Desktop.addInternalFrame(af,
+                                 title,
+                                 AlignFrame.DEFAULT_WIDTH,
+                                 AlignFrame.DEFAULT_HEIGHT);
+
+        af.statusBar.setText("Successfully pasted alignment file");
+
+        try
         {
-          // Parse out the ids from the structured names
-          boolean errors = false;
-          for (int i = 0; i < al.getHeight(); i++)
-          {
-            PDBEntry entry = new PDBEntry();
-            com.stevesoft.pat.Regex idbits = new com.stevesoft.pat.Regex(
-                "PDB\\|([0-9A-z]{4})\\|(.)");
-            if (idbits.search(al.getSequenceAt(i).getName()))
-            {
-              String pdbid = idbits.substring(1);
-              String pdbccode = idbits.substring(2);
-              // Construct the PDBEntry
-              entry.setId(pdbid);
-              if (entry.getProperty() == null)
-                entry.setProperty(new Hashtable());
-              entry.getProperty().put("chains",
-                                      pdbccode
-                                      + "=" + al.getSequenceAt(i).getStart()
-                                      + "-" + al.getSequenceAt(i).getEnd());
-              al.getSequenceAt(i).getDatasetSequence().addPDBId(entry);
-
-              // We make a DBRefEtntry because we have obtained the PDB file from a verifiable source
-              // JBPNote - PDB DBRefEntry should also carry the chain and mapping information
-              DBRefEntry dbentry = new DBRefEntry(jalview.datamodel.DBRefSource.PDB,"0",pdbid);
-              al.getSequenceAt(i).getDatasetSequence().addDBRef(dbentry);
-            }
-            else
-            {
-              // don't add an entry for this chain, but this is probably a bug
-              // that the user should know about.
-              jalview.bin.Cache.log.warn(
-                  "No PDBEntry constructed for sequence " + i + " : " +
-                  al.getSequenceAt(i).getName());
-              errors = true;
-            }
-          }
-          if (errors)
-            jalview.bin.Cache.log.warn(
-                "Query string that resulted in PDBEntry construction failure was :\n" +
-                textfield.getText());
+          af.setMaximum(jalview.bin.Cache.getDefault("SHOW_FULLSCREEN", false));
         }
-
+        catch (Exception ex)
+        {}
       }
       else
-        showErrorMessage("Error retrieving " + textfield.getText()
-                         + " from " + database.getSelectedItem());
+      {
+        for (int i = 0; i < al.getHeight(); i++)
+        {
+          alignFrame.viewport.alignment.addSequence(al.getSequenceAt(i)); // this also creates dataset sequence entries
+        }
+        alignFrame.viewport.setEndSeq(alignFrame.viewport.alignment.
+                                      getHeight());
+        alignFrame.viewport.alignment.getWidth();
+        alignFrame.viewport.firePropertyChange("alignment", null,
+                                               alignFrame.viewport.
+                                               getAlignment().getSequences());
+      }
     }
-
     return al;
-
   }
 
   void showErrorMessage(final String error)
@@ -495,9 +533,8 @@ public class SequenceFetcher
       {
         JOptionPane.showInternalMessageDialog(Desktop.desktop,
                                               error, "Error Retrieving Data",
-                                          JOptionPane.WARNING_MESSAGE);
+                                              JOptionPane.WARNING_MESSAGE);
       }
     });
   }
 }
-