+/*\r
+* Jalview - A Sequence Alignment Editor and Viewer\r
+* Copyright (C) 2006 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.analysis;\r
\r
-import jalview.jbgui.*;\r
import jalview.datamodel.*;\r
-import jalview.io.*;\r
-import jalview.analysis.*;\r
\r
-import java.awt.*;\r
-import java.applet.Applet;\r
import java.util.*;\r
-import java.net.*;\r
-import java.io.*;\r
\r
-public class AAFrequency {\r
\r
- // Takes in a vector of sequences and column start and column end\r
- // and returns a vector of size (end-start+1). Each element of the\r
- // vector contains a hashtable with the keys being residues and\r
- // the values being the count of each residue in that column.\r
- // This class is used extensively in calculating alignment colourschemes\r
- // that depend on the amount of conservation in each alignment column.\r
-\r
-\r
- public static Vector calculate(Vector sequences,int start,int end) {\r
-\r
- Vector result = new Vector();\r
-\r
- for (int i = start;i <= end; i++)\r
+/**\r
+ * Takes in a vector or array of sequences and column start and column end\r
+ * and returns a new Hashtable[] of size maxSeqLength, if Hashtable not supplied.\r
+ * This class is used extensively in calculating alignment colourschemes\r
+ * that depend on the amount of conservation in each alignment column.\r
+ * @author $author$\r
+ * @version $Revision$\r
+ */\r
+public class AAFrequency\r
+{\r
+ //No need to store 1000s of strings which are not\r
+ //visible to the user.\r
+ public static final String MAXCOUNT = "C";\r
+ public static final String MAXRESIDUE="R";\r
+ public static final String PID_GAPS = "G";\r
+ public static final String PID_NOGAPS="N";\r
+\r
+ public static final Hashtable [] calculate(Vector sequences, int start, int end)\r
+ {\r
+ SequenceI [] seqs = new SequenceI[sequences.size()];\r
+ int width = 0;\r
+ for(int i=0; i<sequences.size(); i++)\r
+ {\r
+ seqs[i] = (SequenceI) sequences.elementAt(i);\r
+ if(seqs[i].getLength()>width)\r
+ width = seqs[i].getLength();\r
+ }\r
+\r
+ Hashtable [] reply = new Hashtable[width];\r
+\r
+ if(end>=width)\r
+ {\r
+ end = width-1;\r
+ }\r
+\r
+ calculate(seqs, start, end, reply);\r
+\r
+ return reply;\r
+ }\r
+\r
+public static final void calculate(SequenceI[] sequences,\r
+ int start, int end,\r
+ Hashtable [] result)\r
+{\r
+ Hashtable residueHash;\r
+ int maxCount, nongap, i, j, v, jSize = sequences.length;\r
+ String maxResidue;\r
+ char c;\r
+ float percentage;\r
+\r
+ int[] values = new int[132];\r
+\r
+ String seq;\r
+\r
+ for (i = start; i < end; i++)\r
+ {\r
+ residueHash = new Hashtable();\r
+ maxCount = 0;\r
+ maxResidue = "";\r
+ nongap = 0;\r
+ values = new int[132];\r
+\r
+ for (j = 0; j < jSize; j++)\r
{\r
-\r
- Hashtable residueHash = new Hashtable();\r
- int maxCount = 0;\r
- String maxResidue = "-";\r
- int nongap = 0;\r
- for (int j=0; j < sequences.size(); j++)\r
+ seq = sequences[j].getSequence();\r
+ if (seq.length() > i)\r
{\r
+ c = seq.charAt(i);\r
\r
- if (sequences.elementAt(j) instanceof Sequence)\r
- {\r
- Sequence s = (Sequence)sequences.elementAt(j);\r
-\r
- if (s.getSequence().length() > i)\r
- {\r
-\r
- String res = s.getSequence().charAt(i)+"";\r
-\r
- if (!jalview.util.Comparison.isGap(res.charAt(0)))\r
- nongap++;\r
- else\r
- res = "-"; // we always use this for gaps in the property vectors\r
-\r
- if (residueHash.containsKey(res))\r
- {\r
-\r
- int count = ((Integer)residueHash.get(res)).intValue() ;\r
- count++;\r
-\r
- if (!jalview.util.Comparison.isGap(res.charAt(0)) && count >= maxCount)\r
- {\r
-\r
- if(count>maxCount)\r
- maxResidue = res;\r
- else if(maxResidue.indexOf(res)==-1)\r
- maxResidue += res;\r
-\r
- maxCount = count;\r
- }\r
-\r
- residueHash.put(res,new Integer(count));\r
- }\r
- else\r
- residueHash.put(res,new Integer(1));\r
-\r
-\r
- }\r
- else\r
- {\r
- if (residueHash.containsKey("-"))\r
- {\r
- int count = ((Integer)residueHash.get("-")).intValue() ;\r
- count++;\r
- residueHash.put("-",new Integer(count));\r
- }\r
- else\r
- residueHash.put("-",new Integer(1));\r
+ if(c == '.' || c==' ')\r
+ c = '-';\r
\r
- }\r
+ if(c=='-')\r
+ {\r
+ values['-']++;\r
+ continue;\r
+ }\r
+ else if ('a' <= c && c <= 'z')\r
+ {\r
+ c -= 32 ;//('a' - 'A');\r
}\r
- }\r
-\r
- residueHash.put("maxCount",new Integer(maxCount));\r
- if(maxCount<0)\r
- System.out.println("asasa "+maxCount);\r
- residueHash.put("maxResidue", maxResidue);\r
- residueHash.put("size", new Integer(sequences.size()));\r
- residueHash.put("nongap", new Integer(nongap));\r
- result.addElement(residueHash);\r
- }\r
-\r
- return result;\r
- }\r
-\r
- public static Vector calculatePID(SequenceI refseq,Vector sequences,int window,int start,int end) {\r
-\r
- Vector result = new Vector();\r
-\r
- boolean init = true;\r
-\r
-\r
- Vector prev = null;\r
-\r
- for (int i = start;i <= end; i++) {\r
- Vector values = new Vector();\r
-\r
- result.addElement(values);\r
-\r
- // If start < window/2 then set value to zero.\r
-\r
- if (i< window/2 || i >= refseq.getSequence().length()-window/2) {\r
- for (int j = 0; j < sequences.size(); j++) {\r
- values.addElement(new Integer(0));\r
- }\r
- } else if (init == true) {\r
- init = false;\r
-\r
- int winstart = i-window/2;\r
- int winend = i+window/2;\r
-\r
- if (window%2 != 0) {\r
- winend++;\r
- }\r
-\r
- for (int j = 0; j < sequences.size(); j++) {\r
- values.addElement(new Integer(0));\r
- }\r
-\r
- for (int k = winstart; k <= winend; k++) {\r
- String refchar = refseq.getSequence().substring(k,k+1);\r
- if (jalview.util.Comparison.isGap(refchar.charAt(0)))\r
- refchar="-";\r
- else {\r
- for (int j = 0; j < sequences.size(); j++) {\r
-\r
- Sequence s = (Sequence)sequences.elementAt(j);\r
-\r
- if (s.getSequence().length() > k) {\r
-\r
- String res = s.getSequence().substring(k,k+1); // no gapchar test needed\r
-\r
- if (res.equals(refchar)) {\r
- int val = ((Integer)values.elementAt(j)).intValue();\r
- val++;\r
- values.setElementAt(new Integer(val),j);\r
- }\r
- }\r
- }\r
- }\r
-\r
- }\r
-\r
- prev = values;\r
- } else {\r
- int winstart = i-window/2;\r
- int winend = i+window/2;\r
-\r
- if (window%2 != 0) {\r
- winend++;\r
- }\r
- // We need to take the previous set of values\r
- // subtract the pid at winstart-1\r
- // and add the pid at winend;\r
-\r
- String pre_refchar = refseq.getSequence().substring(winstart-1,winstart);\r
- String pos_refchar = "-";\r
-\r
- if (refseq.getSequence().length() > winend) {\r
- pos_refchar = refseq.getSequence().substring(winend,winend+1);\r
- }\r
-\r
- for (int j = 0; j < sequences.size(); j++) {\r
- // First copy the pid value from i-1\r
-\r
- int val = ((Integer)prev.elementAt(j)).intValue();\r
-\r
- Sequence s = (Sequence)sequences.elementAt(j);\r
-\r
- String pre_char = s.getSequence().substring(winstart-1,winstart);\r
-\r
- String pos_char = "-";\r
-\r
- if (s.getSequence().length() > winend) {\r
- pos_char = s.getSequence().substring(winend,winend+1);\r
- }\r
-\r
- // Now substract 1 if the chars at winstart-1 match\r
-\r
- if (jalview.util.Comparison.isGap(pre_refchar.charAt(0)) == false\r
- && pre_char.equals(pre_refchar)) {\r
- val--;\r
- }\r
-\r
- if (jalview.util.Comparison.isGap(pos_refchar.charAt(0)) == false\r
- && pos_char.equals(pos_refchar)) {\r
- val++;\r
- }\r
-\r
- values.addElement(new Integer(val));\r
-\r
-\r
- }\r
- prev = values;\r
- }\r
- }\r
-\r
- return result;\r
- }\r
-\r
- public static Hashtable findBlocks(Vector seqs, int start, int end,Vector exc) {\r
-\r
- // start and end are in real (not relative coords);\r
-\r
- // The coords in the hashtable that is returned are in relative coords\r
- // i.e. start from 0\r
-\r
- Hashtable blocks = new Hashtable();\r
-\r
- boolean prev = false;\r
- int bstart = -1;\r
-\r
- for (int i = start; i <= end ; i++) {\r
- SequenceI seq = (SequenceI)seqs.elementAt(0);\r
-\r
- char c = seq.getCharAt(i);\r
-\r
- boolean found = true;\r
-\r
- int j = 1;\r
-\r
- while (j < seqs.size() && found == true) {\r
-\r
- SequenceI jseq = (SequenceI)seqs.elementAt(j);\r
-\r
- if (!exc.contains(jseq)) {\r
-\r
- char cc = jseq.getCharAt(i);\r
-\r
- if ( cc != c) {\r
- found = false;\r
- }\r
- }\r
- j++;\r
- }\r
-\r
-\r
- if (prev == false && found == true) {\r
- bstart = i;\r
- } else if (prev == true && found == false && bstart != -1) {\r
-\r
- int blockstart = bstart-start;\r
- int blocklen = i-bstart;\r
-\r
- //System.out.println("Start len " + blockstart + " " + blocklen);\r
-\r
- for (int jj = blockstart; jj < blockstart + blocklen;jj++) {\r
- blocks.put(new Integer(jj),new Integer(blocklen));\r
- }\r
-\r
- bstart = -1;\r
- }\r
- prev = found;\r
- }\r
-\r
- if (bstart != -1) {\r
-\r
- int blockstart = bstart-start;\r
- int blocklen = end-bstart;\r
-\r
- // System.out.println("Start len " + blockstart + " " + blocklen);\r
\r
- for (int jj = blockstart; jj < blockstart + blocklen;jj++) {\r
- blocks.put(new Integer(blockstart),new Integer(blocklen));\r
- }\r
+ nongap++;\r
+ values[c]++;\r
\r
- }\r
- return blocks;\r
+ }\r
+ else\r
+ {\r
+ values['-']++;\r
+ }\r
}\r
\r
+ for (v = 'A'; v < 'Z'; v++)\r
+ {\r
+ if (values[v] < 2 || values[v] < maxCount)\r
+ continue;\r
\r
-\r
- public static Hashtable findKmerCount(SequenceI seq, int start, int end,int window, int step,Vector kmers) {\r
-\r
- int tmpstart = start;\r
- Hashtable vals = new Hashtable();\r
-\r
- while (tmpstart <= end) {\r
-\r
- String tmpstr = seq.getSequence().substring(tmpstart-window/2,tmpstart+window/2);\r
-\r
- int count = 0;\r
-\r
- //System.out.println("Str " + tmpstr);\r
-\r
- for (int ii = 0; ii < kmers.size(); ii++) {\r
- String kmer = ((SequenceI)kmers.elementAt(ii)).getSequence();\r
-\r
- int i = -1;\r
-\r
- while (tmpstr.indexOf(kmer,i) != -1) {\r
- i = tmpstr.indexOf(kmer,i);\r
-\r
- i++;\r
- count++;\r
- }\r
- ii++;\r
- }\r
- vals.put(new Integer(tmpstart),new Integer(count));\r
- tmpstart += step;\r
- }\r
- return vals;\r
+ if (values[v] > maxCount)\r
+ {\r
+ maxResidue = String.valueOf( (char) v);\r
+ }\r
+ else if (values[v] == maxCount)\r
+ {\r
+ maxResidue += String.valueOf( (char) v);\r
+ }\r
+ maxCount = values[v];\r
}\r
\r
- public static Hashtable findBlockStarts(Vector seqs, int start, int end,Vector exc) {\r
-\r
- // start and end are in real (not relative coords);\r
-\r
- // The coords in the hashtable that is returned are in relative coords\r
- // i.e. start from 0\r
-\r
- Hashtable blocks = new Hashtable();\r
-\r
- boolean prev = false;\r
- int bstart = -1;\r
-\r
- for (int i = start; i <= end ; i++) {\r
- SequenceI seq = (SequenceI)seqs.elementAt(0);\r
-\r
- char c = seq.getCharAt(i);\r
-\r
- boolean found = true;\r
-\r
- int j = 1;\r
-\r
- while (j < seqs.size() && found == true) {\r
-\r
- SequenceI jseq = (SequenceI)seqs.elementAt(j);\r
+ if(maxResidue.length()==0)\r
+ maxResidue = "-";\r
\r
- if (!exc.contains(jseq)) {\r
+ residueHash.put(MAXCOUNT, new Integer(maxCount));\r
+ residueHash.put(MAXRESIDUE, maxResidue);\r
\r
- char cc = jseq.getCharAt(i);\r
-\r
- if ( cc != c) {\r
- found = false;\r
- }\r
- }\r
- j++;\r
- }\r
-\r
-\r
- if (prev == false && found == true) {\r
- bstart = i;\r
- } else if (prev == true && found == false && bstart != -1) {\r
-\r
- int blockstart = bstart-start;\r
- int blocklen = i-bstart;\r
-\r
- // System.out.println("Start len " + blockstart + " " + blocklen);\r
-\r
- //for (int jj = blockstart; jj < blockstart + blocklen;jj++) {\r
- blocks.put(new Integer(blockstart),new Integer(blocklen));\r
- // }\r
-\r
- bstart = -1;\r
- }\r
- prev = found;\r
- }\r
-\r
- if (bstart != -1) {\r
-\r
- int blockstart = bstart-start;\r
- int blocklen = end-bstart;\r
-\r
- // System.out.println("Start len " + blockstart + " " + blocklen);\r
-\r
- //for (int jj = blockstart; jj < blockstart + blocklen;jj++) {\r
- blocks.put(new Integer(blockstart),new Integer(blocklen));\r
- // }\r
-\r
- }\r
- return blocks;\r
- }\r
+ percentage = ( (float) maxCount * 100) / (float) jSize;\r
+ residueHash.put(PID_GAPS, new Float(percentage));\r
\r
+ percentage = ( (float) maxCount * 100) / (float) nongap;\r
+ residueHash.put(PID_NOGAPS, new Float(percentage));\r
+ result[i] = residueHash;\r
+ }\r
}\r
+}\r
+\r
\r