-/*\r
-* Jalview - A Sequence Alignment Editor and Viewer\r
-* Copyright (C) 2005 AM Waterhouse, J Procter, G Barton, M Clamp, S Searle\r
-*\r
-* This program 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 2\r
-* of the License, or (at your option) any later version.\r
-*\r
-* This program is distributed in the hope that it will be useful,\r
-* but WITHOUT ANY WARRANTY; without even the implied warranty of\r
-* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the\r
-* GNU General Public License for more details.\r
-*\r
-* You should have received a copy of the GNU General Public License\r
-* along with this program; if not, write to the Free Software\r
-* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA\r
-*/\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
-public class PfamFile extends AlignFile {\r
- Vector ids;\r
-\r
- public PfamFile() {\r
- }\r
-\r
- public PfamFile(String inStr) {\r
- super(inStr);\r
- }\r
-\r
- public PfamFile(String inFile, String type) throws IOException {\r
- super(inFile, type);\r
- }\r
-\r
- public void initData() {\r
- super.initData();\r
- ids = new Vector();\r
- }\r
-\r
- public void parse() throws IOException {\r
- int i = 0;\r
- String line;\r
-\r
- Hashtable seqhash = new Hashtable();\r
- Vector headers = new Vector();\r
-\r
- while ((line = nextLine()) != null) {\r
- if (line.indexOf(" ") != 0) {\r
- if (line.indexOf("#") != 0) {\r
- StringTokenizer str = new StringTokenizer(line, " ");\r
- String id = "";\r
-\r
- if (str.hasMoreTokens()) {\r
- id = str.nextToken();\r
-\r
- StringBuffer tempseq;\r
-\r
- if (seqhash.containsKey(id)) {\r
- tempseq = (StringBuffer) seqhash.get(id);\r
- } else {\r
- tempseq = new StringBuffer();\r
- seqhash.put(id, tempseq);\r
- }\r
-\r
- if (!(headers.contains(id))) {\r
- headers.addElement(id);\r
- }\r
-\r
- tempseq.append(str.nextToken());\r
- }\r
- }\r
- }\r
- }\r
-\r
- this.noSeqs = headers.size();\r
-\r
- if (noSeqs < 1) {\r
- throw new IOException("No sequences found (PFAM input)");\r
- }\r
-\r
- for (i = 0; i < headers.size(); i++) {\r
- if (seqhash.get(headers.elementAt(i)) != null) {\r
- if (maxLength < seqhash.get(headers.elementAt(i)).toString()\r
- .length()) {\r
- maxLength = seqhash.get(headers.elementAt(i)).toString()\r
- .length();\r
- }\r
-\r
- String head = headers.elementAt(i).toString();\r
- int start = 1;\r
- int end = seqhash.get(headers.elementAt(i)).toString().length();\r
-\r
- if (head.indexOf("/") > 0) {\r
- StringTokenizer st = new StringTokenizer(head, "/");\r
-\r
- if (st.countTokens() == 2) {\r
- ids.addElement(st.nextToken());\r
-\r
- String tmp = st.nextToken();\r
- st = new StringTokenizer(tmp, "-");\r
-\r
- if (st.countTokens() == 2) {\r
- start = Integer.valueOf(st.nextToken()).intValue();\r
- end = Integer.valueOf(st.nextToken()).intValue();\r
- } else {\r
- start = -1;\r
- end = -1;\r
- }\r
- } else {\r
- ids.addElement(headers.elementAt(i));\r
- }\r
- } else {\r
- ids.addElement(headers.elementAt(i));\r
- }\r
-\r
- Sequence newSeq = null;\r
-\r
- if ((start != -1) && (end != -1)) {\r
- newSeq = new Sequence(ids.elementAt(i).toString(),\r
- seqhash.get(headers.elementAt(i).toString())\r
- .toString(), start, end);\r
- seqs.addElement(newSeq);\r
- } else {\r
- newSeq = new Sequence(ids.elementAt(i).toString(),\r
- seqhash.get(headers.elementAt(i).toString())\r
- .toString(), 1,\r
- seqhash.get(headers.elementAt(i).toString())\r
- .toString().length());\r
- seqs.addElement(newSeq);\r
- }\r
-\r
- if (!isValidProteinSequence(newSeq.getSequence())) {\r
- throw new IOException(\r
- "Not a valid protein sequence - (PFAM input)");\r
- }\r
- } else {\r
- System.err.println("PFAM File reader: Can't find sequence for " +\r
- headers.elementAt(i));\r
- }\r
- }\r
- }\r
-\r
- public static String print(SequenceI[] s) {\r
- StringBuffer out = new StringBuffer("");\r
-\r
- int max = 0;\r
- int maxid = 0;\r
-\r
- int i = 0;\r
-\r
- while ((i < s.length) && (s[i] != null)) {\r
- String tmp = s[i].getName() + "/" + s[i].getStart() + "-" +\r
- s[i].getEnd();\r
-\r
- if (s[i].getSequence().length() > max) {\r
- max = s[i].getSequence().length();\r
- }\r
-\r
- if (tmp.length() > maxid) {\r
- maxid = tmp.length();\r
- }\r
-\r
- i++;\r
- }\r
-\r
- if (maxid < 15) {\r
- maxid = 15;\r
- }\r
-\r
- int j = 0;\r
-\r
- while ((j < s.length) && (s[j] != null)) {\r
- out.append(new Format("%-" + maxid + "s").form(s[j].getName() +\r
- "/" + s[j].getStart() + "-" + s[j].getEnd()) + " ");\r
-\r
- out.append(s[j].getSequence() + "\n");\r
- j++;\r
- }\r
-\r
- out.append("\n");\r
-\r
- return out.toString();\r
- }\r
-\r
- public String print() {\r
- return print(getSeqsAsArray());\r
- }\r
-}\r
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)
+ * Copyright (C) 2011 J Procter, AM Waterhouse, J Engelhardt, LM Lui, G Barton, M Clamp, S Searle
+ *
+ * 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/>.
+ */
+package jalview.io;
+
+import java.io.*;
+import java.util.*;
+
+import jalview.datamodel.*;
+import jalview.util.*;
+
+public class PfamFile extends AlignFile
+{
+
+ public PfamFile()
+ {
+ }
+
+ public PfamFile(String inFile, String type) throws IOException
+ {
+ super(inFile, type);
+ }
+
+ public PfamFile(FileParse source) throws IOException
+ {
+ super(source);
+ }
+
+ public void initData()
+ {
+ super.initData();
+ }
+
+ public void parse() throws IOException
+ {
+ int i = 0;
+ String line;
+
+ Hashtable seqhash = new Hashtable();
+ Vector headers = new Vector();
+
+ while ((line = nextLine()) != null)
+ {
+ if (line.indexOf(" ") != 0)
+ {
+ if (line.indexOf("#") != 0)
+ {
+ // TODO: verify pfam format requires spaces and not tab characters -
+ // if not upgrade to use stevesoft regex and look for whitespace.
+ StringTokenizer str = new StringTokenizer(line, " ");
+ String id = "";
+
+ if (str.hasMoreTokens())
+ {
+ id = str.nextToken();
+
+ StringBuffer tempseq;
+
+ if (seqhash.containsKey(id))
+ {
+ tempseq = (StringBuffer) seqhash.get(id);
+ }
+ else
+ {
+ tempseq = new StringBuffer();
+ seqhash.put(id, tempseq);
+ }
+
+ if (!(headers.contains(id)))
+ {
+ headers.addElement(id);
+ }
+ if (str.hasMoreTokens())
+ {
+ tempseq.append(str.nextToken());
+ }
+ }
+ }
+ }
+ }
+
+ this.noSeqs = headers.size();
+
+ if (noSeqs < 1)
+ {
+ throw new IOException("No sequences found (PFAM input)");
+ }
+
+ for (i = 0; i < headers.size(); i++)
+ {
+ if (seqhash.get(headers.elementAt(i)) != null)
+ {
+ if (maxLength < seqhash.get(headers.elementAt(i)).toString()
+ .length())
+ {
+ maxLength = seqhash.get(headers.elementAt(i)).toString().length();
+ }
+
+ Sequence newSeq = parseId(headers.elementAt(i).toString());
+ newSeq.setSequence(seqhash.get(headers.elementAt(i).toString())
+ .toString());
+ seqs.addElement(newSeq);
+ }
+ else
+ {
+ System.err.println("PFAM File reader: Can't find sequence for "
+ + headers.elementAt(i));
+ }
+ }
+ }
+
+ public String print(SequenceI[] s)
+ {
+ StringBuffer out = new StringBuffer("");
+
+ int max = 0;
+ int maxid = 0;
+
+ int i = 0;
+
+ while ((i < s.length) && (s[i] != null))
+ {
+ String tmp = printId(s[i]);
+
+ if (s[i].getSequence().length > max)
+ {
+ max = s[i].getSequence().length;
+ }
+
+ if (tmp.length() > maxid)
+ {
+ maxid = tmp.length();
+ }
+
+ i++;
+ }
+
+ if (maxid < 15)
+ {
+ maxid = 15;
+ }
+
+ int j = 0;
+
+ while ((j < s.length) && (s[j] != null))
+ {
+ out.append(new Format("%-" + maxid + "s").form(printId(s[j]) + " "));
+
+ out.append(s[j].getSequenceAsString());
+ out.append(newline);
+ j++;
+ }
+
+ out.append(newline);
+
+ return out.toString();
+ }
+
+ public String print()
+ {
+ return print(getSeqsAsArray());
+ }
+}
git://source.jalview.org / jalview.git / blobdiff