basic jpred concise prediction reader.
authorjprocter <Jim Procter>
Wed, 23 Feb 2005 15:57:04 +0000 (15:57 +0000)
committerjprocter <Jim Procter>
Wed, 23 Feb 2005 15:57:04 +0000 (15:57 +0000)
src/jalview/io/JPredFile.java [new file with mode: 0755]

diff --git a/src/jalview/io/JPredFile.java b/src/jalview/io/JPredFile.java
new file mode 100755 (executable)
index 0000000..cc8fb56
--- /dev/null
@@ -0,0 +1,236 @@
+/**
+ * PredFile.java
+ * JalviewX / Vamsas Project
+ * JPred.seq.concise reader
+ */
+package jalview.io;
+
+import jalview.datamodel.*;
+import jalview.util.*;
+
+import java.io.*;
+import java.util.*;
+
+public class JPredFile
+    extends AlignFile
+{
+  Vector ids;
+  Vector conf;
+  Hashtable Scores; // Hash of names and score vectors
+
+  public JPredFile(String inStr)
+  {
+    super(inStr);
+  }
+
+  public void initData()
+  {
+
+    super.initData();
+    Scores = new Hashtable();
+  }
+
+  public JPredFile(String inFile, String type)
+      throws IOException
+  {
+
+    super(inFile, type);
+  }
+
+  /**
+   * parse a JPred concise file into a sequence-alignment like object.
+   */
+
+  public void parse()
+      throws IOException
+  {
+
+    String line;
+
+    noSeqs = 0;
+    Vector seq_entries = new Vector();
+    Vector ids = new Vector();
+
+    while ( (line = nextLine()) != null)
+    {
+      // Concise format allows no comments or non comma-formatted data
+      StringTokenizer str = new StringTokenizer(line, ":");
+      String id = "";
+      String seq = "";
+      if (str.hasMoreTokens())
+      {
+        id = str.nextToken();
+        String seqsym = str.nextToken();
+        StringTokenizer symbols = new StringTokenizer(seqsym, ",");
+        // decide if we have more than just alphanumeric symbols
+        int numSymbols = symbols.countTokens();
+        if (seq.length() != (2 * numSymbols))
+        {
+          // Set of scalars for some property
+          if (Scores.containsKey(id))
+          {
+            int i = 1;
+            while (Scores.containsKey(id + "_" + i))
+            {
+              i++;
+            }
+            id = id + "_" + i;
+          }
+          Vector scores = new Vector(numSymbols);
+          // Typecheck from first entry
+          int i = 0;
+          String ascore = symbols.nextToken();
+          try
+          {
+            // store elements as floats...
+            do
+            {
+              Float score = new Float(ascore);
+              scores.set(i, (Object) score);
+              ascore = symbols.nextToken();
+            }
+            while (++i < numSymbols);
+          }
+          catch (Exception e)
+          {
+            // or just keep them as strings
+            for (int j = 0; j < i; j++)
+            {
+              scores.set(j,
+                         (Object)
+                         ( (Float) scores.get(j)).toString());
+            }
+            do
+            {
+              scores.set(i, ascore);
+              ascore = symbols.nextToken();
+            }
+            while (++i < numSymbols);
+          }
+          Scores.put(id, scores);
+        }
+        else
+        {
+          if (id.equals("jnetconf"))
+          {
+            id = "Prediction Confidence";
+            this.conf = new Vector(numSymbols);
+            for (int i = 0; i < numSymbols; i++)
+            {
+              conf.set(i, (Object) symbols.nextToken());
+            }
+          }
+          else
+          {
+            // Sequence or a prediction string (rendered as sequence)
+
+            StringBuffer newseq = new StringBuffer();
+            for (int i = 0; i < numSymbols; i++)
+            {
+              newseq.append(symbols.nextToken());
+            }
+            if (id.indexOf(";") > -1)
+            {
+              seq_entries.addElement(newseq);
+              ids.addElement(id.substring(id.indexOf(";")));
+              noSeqs++;
+            }
+            else
+            {
+              if (id.equals("JNETPRED"))
+              {
+                id = "Predicted Secondary Structure";
+              }
+              seq_entries.addElement(newseq);
+              ids.addElement(id);
+            }
+          }
+        }
+
+      }
+    }
+
+    if (noSeqs < 1)
+    {
+      throw new IOException("JpredFile Parser: No sequence in the prediction!");
+    }
+    maxLength = seq_entries.elementAt(0).toString().length();
+    for (int i = 0; i < ids.size(); i++)
+    {
+      // Add all sequence like objects
+
+      Sequence newSeq = new Sequence(ids.elementAt(i).toString(),
+                                     seq_entries.elementAt(i).toString(), 1,
+                                     seq_entries.elementAt(i).toString().length());
+      if (!isValidProteinSequence(newSeq.getSequence()))
+      {
+        throw new IOException("JPredConcise: Not a valid protein sequence - ("
+                              + ids.elementAt(i).toString() + ")");
+      }
+
+      if (maxLength != seq_entries.elementAt(i).toString().length())
+      {
+        throw new IOException("JPredConcise: Entry (" +
+                              ids.elementAt(i).toString()
+                              + ") has an unexpected number of columns");
+      }
+      seqs.addElement(newSeq);
+
+    }
+  }
+
+  /**
+   * print
+   *
+   * @return String
+   */
+
+  public String print()
+  {
+    return "Not Supported";
+  }
+}
+/*
+StringBuffer out = new StringBuffer();
+
+out.append("START PRED\n");
+for (int i = 0; i < s[0].sequence.length(); i++)
+{
+  out.append(s[0].sequence.substring(i, i + 1) + " ");
+  out.append(s[1].sequence.substring(i, i + 1) + " ");
+  out.append(s[1].score[0].elementAt(i) + " ");
+  out.append(s[1].score[1].elementAt(i) + " ");
+  out.append(s[1].score[2].elementAt(i) + " ");
+  out.append(s[1].score[3].elementAt(i) + " ");
+
+  out.append("\n");
+}
+out.append("END PRED\n");
+return out.toString();
+}
+
+    public static void main(String[] args)
+{
+  try
+  {
+    BLCFile blc = new BLCFile(args[0], "File");
+    DrawableSequence[] s = new DrawableSequence[blc.seqs.size()];
+    for (int i = 0; i < blc.seqs.size(); i++)
+    {
+      s[i] = new DrawableSequence( (Sequence) blc.seqs.elementAt(i));
+    }
+    String out = BLCFile.print(s);
+
+    AlignFrame af = new AlignFrame(null, s);
+    af.resize(700, 500);
+    af.show();
+    System.out.println(out);
+  }
+  catch (java.io.IOException e)
+  {
+    System.out.println("Exception " + e);
+  }
+}
+
+}
+*/