- */\r
-package jalview.io;\r
-\r
-import jalview.datamodel.*;\r
-\r
-import jalview.util.*;\r
-\r
-import java.io.*;\r
-\r
-import java.util.*;\r
-\r
-\r
-/**\r
- * DOCUMENT ME!\r
- *\r
- * @author $author$\r
- * @version $Revision$\r
- */\r
-public class JPredFile extends AlignFile\r
-{\r
- Vector ids;\r
- Vector conf;\r
- Hashtable Scores; // Hash of names and score vectors\r
- Hashtable Symscores; // indexes of symbol annotation properties in sequenceI vector\r
- private int QuerySeqPosition;\r
-\r
- /**\r
- * Creates a new JPredFile object.\r
- *\r
- * @param inStr DOCUMENT ME!\r
- */\r
- public JPredFile(String inStr)\r
- {\r
- super(inStr);\r
- }\r
-\r
- /**\r
- * Creates a new JPredFile object.\r
- *\r
- * @param inFile DOCUMENT ME!\r
- * @param type DOCUMENT ME!\r
- *\r
- * @throws IOException DOCUMENT ME!\r
- */\r
- public JPredFile(String inFile, String type) throws IOException\r
- {\r
- super(inFile, type);\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param QuerySeqPosition DOCUMENT ME!\r
- */\r
- public void setQuerySeqPosition(int QuerySeqPosition)\r
- {\r
- this.QuerySeqPosition = QuerySeqPosition;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int getQuerySeqPosition()\r
- {\r
- return QuerySeqPosition;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public Hashtable getScores()\r
- {\r
- return Scores;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public Hashtable getSymscores()\r
- {\r
- return Symscores;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- */\r
- public void initData()\r
- {\r
- super.initData();\r
- Scores = new Hashtable();\r
- ids = null;\r
- conf = null;\r
- }\r
-\r
- /**
- * parse a JPred concise file into a sequence-alignment like object.
- */\r
- public void parse() throws IOException\r
- {\r
- // JBPNote log.System.out.println("all read in ");\r
- String line;\r
- QuerySeqPosition = -1;\r
- noSeqs = 0;\r
-\r
- Vector seq_entries = new Vector();\r
- Vector ids = new Vector();\r
- Hashtable Symscores = new Hashtable();\r
-\r
- while ((line = nextLine()) != null)\r
- {\r
- // Concise format allows no comments or non comma-formatted data\r
- StringTokenizer str = new StringTokenizer(line, ":");\r
- String id = "";\r
-\r
- if (!str.hasMoreTokens())\r
- {\r
- continue;\r
- }\r
-\r
- id = str.nextToken();\r
-\r
- String seqsym = str.nextToken();\r
- StringTokenizer symbols = new StringTokenizer(seqsym, ",");\r
-\r
- // decide if we have more than just alphanumeric symbols\r
- int numSymbols = symbols.countTokens();\r
-\r
- if (numSymbols == 0)\r
- {\r
- continue;\r
- }\r
-\r
- if (seqsym.length() != (2 * numSymbols))\r
- {\r
- // Set of scalars for some property\r
- if (Scores.containsKey(id))\r
- {\r
- int i = 1;\r
-\r
- while (Scores.containsKey(id + "_" + i))\r
- {\r
- i++;\r
- }\r
-\r
- id = id + "_" + i;\r
- }\r
-\r
- Vector scores = new Vector();\r
-\r
- // Typecheck from first entry\r
- int i = 0;\r
- String ascore = "dead";\r
-\r
- try\r
- {\r
- // store elements as floats...\r
- while (symbols.hasMoreTokens())\r
- {\r
- ascore = symbols.nextToken();\r
-\r
- Float score = new Float(ascore);\r
- scores.addElement((Object) score);\r
- }\r
-\r
- Scores.put(id, scores);\r
- }\r
- catch (Exception e)\r
- {\r
- // or just keep them as strings\r
- i = scores.size();\r
-\r
- for (int j = 0; j < i; j++)\r
- {\r
- scores.set(j,\r
- (Object) ((Float) scores.get(j)).toString());\r
- }\r
-\r
- scores.addElement((Object) ascore);\r
-\r
- while (symbols.hasMoreTokens())\r
- {\r
- ascore = symbols.nextToken();\r
- scores.addElement((Object) ascore);\r
- }\r
-\r
- Scores.put(id, scores);\r
- }\r
- }\r
- else if (id.equals("jnetconf"))\r
- {\r
- // log.debug System.out.println("here");\r
- id = "Prediction Confidence";\r
- this.conf = new Vector(numSymbols);\r
-\r
- for (int i = 0; i < numSymbols; i++)\r
- {\r
- conf.set(i, (Object) symbols.nextToken());\r
- }\r
- }\r
- else\r
- {\r
- // Sequence or a prediction string (rendered as sequence)\r
- StringBuffer newseq = new StringBuffer();\r
-\r
- for (int i = 0; i < numSymbols; i++)\r
- {\r
- newseq.append(symbols.nextToken());\r
- }\r
-\r
- if (id.indexOf(";") > -1)\r
- {\r
- seq_entries.addElement(newseq);\r
-\r
- int i = 1;\r
- String name = id.substring(id.indexOf(";") + 1);\r
-\r
- while (ids.lastIndexOf(name) > -1)\r
- {\r
- name = id.substring(id.indexOf(";") + 1) + "_" + 1;\r
- }\r
-\r
- ids.addElement(name);\r
-\r
- noSeqs++;\r
- }\r
- else\r
- {\r
- if (id.equals("JNETPRED"))\r
- {\r
- id = "Predicted Secondary Structure";\r
- }\r
-\r
- seq_entries.addElement(newseq.toString());\r
- ids.addElement(id);\r
- Symscores.put((Object) id,\r
- (Object) new Integer(ids.size() - 1));\r
- }\r
- }\r
- }\r
-\r
- if (noSeqs < 1)\r
- {\r
- throw new IOException(\r
- "JpredFile Parser: No sequence in the prediction!");\r
- }\r
-\r
- maxLength = seq_entries.elementAt(0).toString().length();\r
-\r
- for (int i = 0; i < ids.size(); i++)\r
- {\r
- // Add all sequence like objects\r
- Sequence newSeq = new Sequence(ids.elementAt(i).toString(),\r
- seq_entries.elementAt(i).toString(), 1,\r
- seq_entries.elementAt(i).toString().length());\r
-\r
- if (!Symscores.containsKey(ids.elementAt(i)) &&\r
- !isValidProteinSequence(newSeq.getSequence()))\r
- {\r
- throw new IOException(\r
- "JPredConcise: Not a valid protein sequence - (" +\r
- ids.elementAt(i).toString() + ")");\r
- }\r
-\r
- if (maxLength != seq_entries.elementAt(i).toString().length())\r
- {\r
- throw new IOException("JPredConcise: Entry (" +\r
- ids.elementAt(i).toString() +\r
- ") has an unexpected number of columns");\r
- }\r
-\r
- if (newSeq.getName().startsWith("QUERY") &&\r
- (QuerySeqPosition == -1))\r
- {\r
- QuerySeqPosition = seqs.size();\r
- }\r
-\r
- seqs.addElement(newSeq);\r
- }\r
- }\r
-\r
- /**
- * print
- *
- * @return String
- */\r
- public String print()\r
- {\r
- return "Not Supported";\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param args DOCUMENT ME!\r
- */\r
- public static void main(String[] args)\r
- {\r
- try\r
- {\r
- JPredFile blc = new JPredFile(args[0], "File");\r
-\r
- for (int i = 0; i < blc.seqs.size(); i++)\r
- {\r
- System.out.println(((Sequence) blc.seqs.elementAt(i)).getName() +\r
- "\n" + ((Sequence) blc.seqs.elementAt(i)).getSequence() +\r
- "\n");\r
- }\r
- }\r
- catch (java.io.IOException e)\r
- {\r
- System.err.println("Exception " + e);\r
- e.printStackTrace();\r
- }\r
- }\r
-}\r
-\r
-\r
-/*
- 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
+ */
+package jalview.io;
+
+import java.util.Locale;
+
+import jalview.datamodel.Alignment;
+import jalview.datamodel.AlignmentAnnotation;
+import jalview.datamodel.Sequence;
+import jalview.datamodel.SequenceI;
+import jalview.util.MessageManager;
+
+import java.io.IOException;
+import java.util.Hashtable;
+import java.util.StringTokenizer;
+import java.util.Vector;
+
+/**
+ * Parser for the JPred/JNet concise format. This is a series of CSV lines, each
+ * line is either a sequence (QUERY), a sequence profile (align;), or jnet
+ * prediction annotation (anything else). Automagic translation happens for
+ * annotation called 'JNETPRED' (translated to Secondary Structure Prediction),
+ * or 'JNETCONF' (translates to 'Prediction Confidence'). Numeric scores are
+ * differentiated from symbolic by being parseable into a float vector. They are
+ * put in Scores. Symscores gets the others. JNetAnnotationMaker translates the
+ * data parsed by this object into annotation on an alignment. It is
+ * automatically called but can be used to transfer the annotation onto a
+ * sequence in another alignment (and insert gaps where necessary)
+ *
+ * @author jprocter
+ * @version $Revision$
+ */
+public class JPredFile extends AlignFile
+{
+ Vector ids;
+
+ Vector conf;
+
+ Hashtable Scores; // Hash of names and score vectors
+
+ Hashtable Symscores; // indexes of symbol annotation properties in sequenceI
+
+ // vector
+
+ private int QuerySeqPosition;
+
+ /**
+ * Creates a new JPredFile object.
+ *
+ * BH allows File or String
+ *
+ * @param inFile
+ * DOCUMENT ME!
+ * @param sourceType
+ * DOCUMENT ME!
+ *
+ * @throws IOException
+ * DOCUMENT ME!
+ */
+ public JPredFile(Object inFile, DataSourceType sourceType)
+ throws IOException