\r
\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
+ * 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
*/\r
public class AAFrequency\r
{\r
- /** Takes in a !!ARRAY!! of sequences and column start and column end\r
- * and fills given 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
+ //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
\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
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
+ maxResidue = "";\r
nongap = 0;\r
values = new int[132];\r
\r
for (j = 0; j < jSize; j++)\r
{\r
- if (sequences[j].getLength() > i)\r
+ seq = sequences[j].getSequence();\r
+ if (seq.length() > i)\r
{\r
- c = sequences[j].getCharAt(i);\r
+ c = seq.charAt(i);\r
\r
- if ('a' <= c && c <= 'z')\r
- {\r
- c -= ('a' - 'A');\r
- }\r
+ if(c == '.' || c==' ')\r
+ c = '-';\r
\r
- if (jalview.util.Comparison.isGap(c))\r
+ if(c=='-')\r
{\r
- c = '-'; // we always use this for gaps in the property vectors\r
+ values['-']++;\r
+ continue;\r
}\r
- else\r
+ else if ('a' <= c && c <= 'z')\r
{\r
- nongap++;\r
+ c -= 32 ;//('a' - 'A');\r
}\r
\r
+ nongap++;\r
values[c]++;\r
\r
}\r
\r
for (v = 'A'; v < 'Z'; v++)\r
{\r
- if (values[v] == 0 || values[v] < maxCount)\r
+ if (values[v] < 2 || values[v] < maxCount)\r
continue;\r
\r
if (values[v] > maxCount)\r
maxCount = values[v];\r
}\r
\r
+ if(maxResidue.length()==0)\r
+ maxResidue = "-";\r
\r
- residueHash.put("maxCount", new Integer(maxCount));\r
- residueHash.put("maxResidue", maxResidue);\r
+ residueHash.put(MAXCOUNT, new Integer(maxCount));\r
+ residueHash.put(MAXRESIDUE, maxResidue);\r
\r
percentage = ( (float) maxCount * 100) / (float) jSize;\r
- residueHash.put("pid_gaps", new Float(percentage));\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
+ residueHash.put(PID_NOGAPS, new Float(percentage));\r
result[i] = residueHash;\r
}\r
}\r