-/*\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.datamodel.SequenceI;\r
-\r
-import jalview.schemes.*;\r
-\r
-import jalview.util.*;\r
-\r
-import java.awt.*;\r
-\r
-import java.util.*;\r
-\r
-\r
-/**\r
- *\r
- *\r
- * @author $author$\r
- * @version $Revision$\r
- */\r
-public class AlignSeq\r
-{\r
- /** DOCUMENT ME!! */\r
- public static java.util.Hashtable dnaHash = new java.util.Hashtable();\r
-\r
- static\r
- {\r
- dnaHash.put("C", new Integer(0));\r
- dnaHash.put("T", new Integer(1));\r
- dnaHash.put("A", new Integer(2));\r
- dnaHash.put("G", new Integer(3));\r
- dnaHash.put("-", new Integer(4));\r
- }\r
-\r
- static String[] dna = { "C", "T", "A", "G", "-" };\r
- static String[] pep =\r
- {\r
- "A", "R", "N", "D", "C", "Q", "E", "G", "H", "I", "L", "K", "M", "F",\r
- "P", "S", "T", "W", "Y", "V", "B", "Z", "X", "-"\r
- };\r
- int[][] score;\r
- int[][] E;\r
- int[][] F;\r
- int[][] traceback;\r
- int[] seq1;\r
- int[] seq2;\r
- SequenceI s1;\r
- SequenceI s2;\r
- public String s1str;\r
- public String s2str;\r
- int maxi;\r
- int maxj;\r
- int[] aseq1;\r
- int[] aseq2;\r
- public String astr1="";\r
- public String astr2="";\r
-\r
- /** DOCUMENT ME!! */\r
- public int seq1start;\r
-\r
- /** DOCUMENT ME!! */\r
- public int seq1end;\r
-\r
- /** DOCUMENT ME!! */\r
- public int seq2start;\r
-\r
- /** DOCUMENT ME!! */\r
- public int seq2end;\r
- int count;\r
-\r
- /** DOCUMENT ME!! */\r
- public int maxscore;\r
- float pid;\r
- int prev = 0;\r
- int gapOpen = 120;\r
- int gapExtend = 20;\r
- int[][] lookup = ResidueProperties.getBLOSUM62();\r
- String[] intToStr = pep;\r
- int defInt = 23;\r
- StringBuffer output = new StringBuffer();\r
- String type;\r
-\r
- /**\r
- * Creates a new AlignSeq object.\r
- *\r
- * @param s1 DOCUMENT ME!\r
- * @param s2 DOCUMENT ME!\r
- * @param type DOCUMENT ME!\r
- */\r
- public AlignSeq(SequenceI s1, SequenceI s2, String type)\r
- {\r
- SeqInit(s1, s1.getSequenceAsString(), s2, s2.getSequenceAsString(), type);\r
- }\r
-\r
- /**\r
- * Creates a new AlignSeq object.\r
- *\r
- * @param s1 DOCUMENT ME!\r
- * @param s2 DOCUMENT ME!\r
- * @param type DOCUMENT ME!\r
- */\r
- public AlignSeq(SequenceI s1,\r
- String string1,\r
- SequenceI s2,\r
- String string2,\r
- String type)\r
- {\r
- SeqInit(s1, string1, s2, string2, type);\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int getMaxScore()\r
- {\r
- return maxscore;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int getSeq2Start()\r
- {\r
- return seq2start;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int getSeq2End()\r
- {\r
- return seq2end;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int getSeq1Start()\r
- {\r
- return seq1start;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int getSeq1End()\r
- {\r
- return seq1end;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public String getOutput()\r
- {\r
- return output.toString();\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public String getAStr1()\r
- {\r
- return astr1;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public String getAStr2()\r
- {\r
- return astr2;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int[] getASeq1()\r
- {\r
- return aseq1;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int[] getASeq2()\r
- {\r
- return aseq2;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public SequenceI getS1()\r
- {\r
- return s1;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public SequenceI getS2()\r
- {\r
- return s2;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param s1 DOCUMENT ME!\r
- * @param s2 DOCUMENT ME!\r
- * @param type DOCUMENT ME!\r
- */\r
- public void SeqInit(SequenceI s1,\r
- String string1,\r
- SequenceI s2,\r
- String string2,\r
- String type)\r
- {\r
-\r
- s1str = extractGaps(jalview.util.Comparison.GapChars, string1);\r
- s2str = extractGaps(jalview.util.Comparison.GapChars, string2);\r
-\r
- if(s1str.length()==0 || s2str.length()==0)\r
- {\r
- System.out.println("ALL GAPS: " +\r
- (s1str.length()==0?s1.getName():" ")\r
- +(s2str.length()==0?s2.getName():""));\r
- return;\r
- }\r
-\r
- this.s1 = s1;\r
- this.s2 = s2;\r
-\r
- this.type = type;\r
-\r
- if (type.equals("pep"))\r
- {\r
- lookup = ResidueProperties.getBLOSUM62();\r
- intToStr = pep;\r
- defInt = 23;\r
- }\r
- else if (type.equals("dna"))\r
- {\r
- lookup = ResidueProperties.getDNA();\r
- intToStr = dna;\r
- defInt = 4;\r
- }\r
- else\r
- {\r
- output.append("Wrong type = dna or pep only");\r
- System.exit(0);\r
- }\r
-\r
- //System.out.println("lookuip " + rt.freeMemory() + " "+ rt.totalMemory());\r
- seq1 = new int[s1str.length()];\r
-\r
- //System.out.println("seq1 " + rt.freeMemory() +" " + rt.totalMemory());\r
- seq2 = new int[s2str.length()];\r
-\r
- //System.out.println("seq2 " + rt.freeMemory() + " " + rt.totalMemory());\r
- score = new int[s1str.length()][s2str.length()];\r
-\r
- //System.out.println("score " + rt.freeMemory() + " " + rt.totalMemory());\r
- E = new int[s1str.length()][s2str.length()];\r
-\r
- //System.out.println("E " + rt.freeMemory() + " " + rt.totalMemory());\r
- F = new int[s1str.length()][s2str.length()];\r
- traceback = new int[s1str.length()][s2str.length()];\r
-\r
- //System.out.println("F " + rt.freeMemory() + " " + rt.totalMemory());\r
- seq1 = stringToInt(s1str, type);\r
-\r
- //System.out.println("seq1 " + rt.freeMemory() + " " + rt.totalMemory());\r
- seq2 = stringToInt(s2str, type);\r
-\r
- //System.out.println("Seq2 " + rt.freeMemory() + " " + rt.totalMemory());\r
- // long tstart = System.currentTimeMillis();\r
- // calcScoreMatrix();\r
- //long tend = System.currentTimeMillis();\r
- //System.out.println("Time take to calculate score matrix = " + (tend-tstart) + " ms");\r
- // printScoreMatrix(score);\r
- //System.out.println();\r
- //printScoreMatrix(traceback);\r
- //System.out.println();\r
- // printScoreMatrix(E);\r
- //System.out.println();\r
- ///printScoreMatrix(F);\r
- //System.out.println();\r
- // tstart = System.currentTimeMillis();\r
- //traceAlignment();\r
- //tend = System.currentTimeMillis();\r
- //System.out.println("Time take to traceback alignment = " + (tend-tstart) + " ms");\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- */\r
- public void traceAlignment()\r
- {\r
- // Find the maximum score along the rhs or bottom row\r
- int max = -9999;\r
-\r
- for (int i = 0; i < seq1.length; i++)\r
- {\r
- if (score[i][seq2.length - 1] > max)\r
- {\r
- max = score[i][seq2.length - 1];\r
- maxi = i;\r
- maxj = seq2.length - 1;\r
- }\r
- }\r
-\r
- for (int j = 0; j < seq2.length; j++)\r
- {\r
- if (score[seq1.length - 1][j] > max)\r
- {\r
- max = score[seq1.length - 1][j];\r
- maxi = seq1.length - 1;\r
- maxj = j;\r
- }\r
- }\r
-\r
- // System.out.println(maxi + " " + maxj + " " + score[maxi][maxj]);\r
- int i = maxi;\r
- int j = maxj;\r
- int trace;\r
- maxscore = score[i][j] / 10;\r
-\r
- seq1end = maxi + 1;\r
- seq2end = maxj + 1;\r
-\r
- aseq1 = new int[seq1.length + seq2.length];\r
- aseq2 = new int[seq1.length + seq2.length];\r
-\r
- count = (seq1.length + seq2.length) - 1;\r
-\r
- while ((i > 0) && (j > 0))\r
- {\r
- if ((aseq1[count] != defInt) && (i >= 0))\r
- {\r
- aseq1[count] = seq1[i];\r
- astr1 = intToStr[seq1[i]] + astr1;\r
- }\r
-\r
- if ((aseq2[count] != defInt) && (j > 0))\r
- {\r
- aseq2[count] = seq2[j];\r
- astr2 = intToStr[seq2[j]] + astr2;\r
- }\r
-\r
- trace = findTrace(i, j);\r
-\r
- if (trace == 0)\r
- {\r
- i--;\r
- j--;\r
- }\r
- else if (trace == 1)\r
- {\r
- j--;\r
- aseq1[count] = defInt;\r
- astr1 = "-" + astr1.substring(1);\r
- }\r
- else if (trace == -1)\r
- {\r
- i--;\r
- aseq2[count] = defInt;\r
- astr2 = "-" + astr2.substring(1);\r
- }\r
-\r
- count--;\r
- }\r
-\r
- seq1start = i + 1;\r
- seq2start = j + 1;\r
-\r
- if (aseq1[count] != defInt)\r
- {\r
- aseq1[count] = seq1[i];\r
- astr1 = intToStr[seq1[i]] + astr1;\r
- }\r
-\r
- if (aseq2[count] != defInt)\r
- {\r
- aseq2[count] = seq2[j];\r
- astr2 = intToStr[seq2[j]] + astr2;\r
- }\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- */\r
- public void printAlignment(java.io.PrintStream os)\r
- {\r
- // Find the biggest id length for formatting purposes\r
- int maxid = s1.getName().length();\r
-\r
- if (s2.getName().length() > maxid)\r
- {\r
- maxid = s2.getName().length();\r
- }\r
-\r
- int len = 72 - maxid - 1;\r
- int nochunks = ((aseq1.length - count) / len) + 1;\r
- pid = 0;\r
-\r
- output.append("Score = " + score[maxi][maxj] + "\n");\r
- output.append("Length of alignment = " + (aseq1.length - count) + "\n");\r
- output.append("Sequence ");\r
- output.append(new Format("%" + maxid + "s").form(s1.getName()));\r
- output.append(" : " + s1.getStart() + " - " + s1.getEnd() + " (Sequence length = " +\r
- s1str.length() + ")\n");\r
- output .append("Sequence ");\r
- output.append(new Format("%" + maxid + "s").form(s2.getName()));\r
- output.append(" : " + s2.getStart() + " - " + s2.getEnd() + " (Sequence length = " +\r
- s2str.length() + ")\n\n");\r
-\r
- for (int j = 0; j < nochunks; j++)\r
- {\r
- // Print the first aligned sequence\r
- output.append(new Format("%" + (maxid) + "s").form(s1.getName()) + " ");\r
-\r
- for (int i = 0; i < len; i++)\r
- {\r
- if ((count + i + (j * len)) < aseq1.length)\r
- {\r
- output.append(new Format("%s").form(intToStr[aseq1[count + i +\r
- (j * len)]]));\r
- }\r
- }\r
-\r
- output.append("\n");\r
- output.append(new Format("%" + (maxid) + "s").form(" ") + " ");\r
-\r
- // Print out the matching chars\r
- for (int i = 0; i < len; i++)\r
- {\r
- if ((count + i + (j * len)) < aseq1.length)\r
- {\r
- if (intToStr[aseq1[count + i + (j * len)]].equals(\r
- intToStr[aseq2[count + i + (j * len)]]) &&\r
- !intToStr[aseq1[count + i + (j * len)]].equals("-"))\r
- {\r
- pid++;\r
- output.append("|");\r
- }\r
- else if (type.equals("pep"))\r
- {\r
- if (ResidueProperties.getPAM250(\r
- intToStr[aseq1[count + i + (j * len)]],\r
- intToStr[aseq2[count + i + (j * len)]]) > 0)\r
- {\r
- output.append(".");\r
- }\r
- else\r
- {\r
- output.append(" ");\r
- }\r
- }\r
- else\r
- {\r
- output.append(" ");\r
- }\r
- }\r
- }\r
-\r
- // Now print the second aligned sequence\r
- output = output.append("\n");\r
- output = output.append(new Format("%" + (maxid) + "s").form(s2.getName()) + " ");\r
-\r
- for (int i = 0; i < len; i++)\r
- {\r
- if ((count + i + (j * len)) < aseq1.length)\r
- {\r
- output .append(new Format("%s").form(intToStr[aseq2[count + i +\r
- (j * len)]]));\r
- }\r
- }\r
-\r
- output = output .append("\n\n");\r
- }\r
-\r
- pid = pid / (float) (aseq1.length - count) * 100;\r
- output = output.append(new Format("Percentage ID = %2.2f\n\n").form(pid));\r
-\r
- try{\r
- os.print(output.toString());\r
- }catch(Exception ex){}\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param mat DOCUMENT ME!\r
- */\r
- public void printScoreMatrix(int[][] mat)\r
- {\r
- int n = seq1.length;\r
- int m = seq2.length;\r
-\r
- for (int i = 0; i < n; i++)\r
- {\r
- // Print the top sequence\r
- if (i == 0)\r
- {\r
- Format.print(System.out, "%8s", s2str.substring(0, 1));\r
-\r
- for (int jj = 1; jj < m; jj++)\r
- {\r
- Format.print(System.out, "%5s", s2str.substring(jj, jj + 1));\r
- }\r
-\r
- System.out.println();\r
- }\r
-\r
- for (int j = 0; j < m; j++)\r
- {\r
- if (j == 0)\r
- {\r
- Format.print(System.out, "%3s", s1str.substring(i, i + 1));\r
- }\r
-\r
- Format.print(System.out, "%3d ", mat[i][j] / 10);\r
- }\r
-\r
- System.out.println();\r
- }\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param i DOCUMENT ME!\r
- * @param j DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int findTrace(int i, int j)\r
- {\r
- int t = 0;\r
- int max = score[i - 1][j - 1] + (lookup[seq1[i]][seq2[j]] * 10);\r
-\r
- if (F[i][j] > max)\r
- {\r
- max = F[i][j];\r
- t = -1;\r
- }\r
- else if (F[i][j] == max)\r
- {\r
- if (prev == -1)\r
- {\r
- max = F[i][j];\r
- t = -1;\r
- }\r
- }\r
-\r
- if (E[i][j] >= max)\r
- {\r
- max = E[i][j];\r
- t = 1;\r
- }\r
- else if (E[i][j] == max)\r
- {\r
- if (prev == 1)\r
- {\r
- max = E[i][j];\r
- t = 1;\r
- }\r
- }\r
-\r
- prev = t;\r
-\r
- return t;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- */\r
- public void calcScoreMatrix()\r
- {\r
- int n = seq1.length;\r
- int m = seq2.length;\r
-\r
- // top left hand element\r
- score[0][0] = lookup[seq1[0]][seq2[0]] * 10;\r
- E[0][0] = -gapExtend;\r
- F[0][0] = 0;\r
-\r
- // Calculate the top row first\r
- for (int j = 1; j < m; j++)\r
- {\r
- // What should these values be? 0 maybe\r
- E[0][j] = max(score[0][j - 1] - gapOpen, E[0][j - 1] - gapExtend);\r
- F[0][j] = -gapExtend;\r
-\r
- score[0][j] = max(lookup[seq1[0]][seq2[j]] * 10, -gapOpen,\r
- -gapExtend);\r
-\r
- traceback[0][j] = 1;\r
- }\r
-\r
- // Now do the left hand column\r
- for (int i = 1; i < n; i++)\r
- {\r
- E[i][0] = -gapOpen;\r
- F[i][0] = max(score[i - 1][0] - gapOpen, F[i - 1][0] - gapExtend);\r
-\r
- score[i][0] = max(lookup[seq1[i]][seq2[0]] * 10, E[i][0], F[i][0]);\r
- traceback[i][0] = -1;\r
- }\r
-\r
- // Now do all the other rows\r
- for (int i = 1; i < n; i++)\r
- {\r
- for (int j = 1; j < m; j++)\r
- {\r
- E[i][j] = max(score[i][j - 1] - gapOpen, E[i][j - 1] -\r
- gapExtend);\r
- F[i][j] = max(score[i - 1][j] - gapOpen, F[i - 1][j] -\r
- gapExtend);\r
-\r
- score[i][j] = max(score[i - 1][j - 1] +\r
- (lookup[seq1[i]][seq2[j]] * 10), E[i][j], F[i][j]);\r
- traceback[i][j] = findTrace(i, j);\r
- }\r
- }\r
- }\r
-\r
-\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param gapChar DOCUMENT ME!\r
- * @param seq DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public static String extractGaps(String gapChar, String seq)\r
- {\r
- StringTokenizer str = new StringTokenizer(seq, gapChar);\r
- StringBuffer newString = new StringBuffer();\r
-\r
- while (str.hasMoreTokens())\r
- {\r
- newString.append( str.nextToken() );\r
- }\r
-\r
- return newString.toString();\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param i1 DOCUMENT ME!\r
- * @param i2 DOCUMENT ME!\r
- * @param i3 DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int max(int i1, int i2, int i3)\r
- {\r
- int max = i1;\r
-\r
- if (i2 > i1)\r
- {\r
- max = i2;\r
- }\r
-\r
- if (i3 > max)\r
- {\r
- max = i3;\r
- }\r
-\r
- return max;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param i1 DOCUMENT ME!\r
- * @param i2 DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int max(int i1, int i2)\r
- {\r
- int max = i1;\r
-\r
- if (i2 > i1)\r
- {\r
- max = i2;\r
- }\r
-\r
- return max;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param s DOCUMENT ME!\r
- * @param type DOCUMENT ME!\r
- *\r
- * @return DOCUMENT ME!\r
- */\r
- public int[] stringToInt(String s, String type)\r
- {\r
- int[] seq1 = new int[s.length()];\r
-\r
- for (int i = 0; i < s.length(); i++)\r
- {\r
- // String ss = s.substring(i, i + 1).toUpperCase();\r
- char c = s.charAt(i);\r
- if ('a' <= c && c <= 'z')\r
- {\r
- // TO UPPERCASE !!!\r
- c -= ('a' - 'A');\r
- }\r
-\r
-\r
- try\r
- {\r
- if (type.equals("pep"))\r
- {\r
- seq1[i] = ResidueProperties.aaIndex[c];\r
- }\r
- else if (type.equals("dna"))\r
- {\r
- seq1[i] = ResidueProperties.nucleotideIndex[c];\r
- }\r
-\r
- if (seq1[i] > 23)\r
- {\r
- seq1[i] = 23;\r
- }\r
- }\r
- catch (Exception e)\r
- {\r
- if (type.equals("dna"))\r
- {\r
- seq1[i] = 4;\r
- }\r
- else\r
- {\r
- seq1[i] = 23;\r
- }\r
- }\r
- }\r
-\r
- return seq1;\r
- }\r
-\r
- /**\r
- * DOCUMENT ME!\r
- *\r
- * @param g DOCUMENT ME!\r
- * @param mat DOCUMENT ME!\r
- * @param n DOCUMENT ME!\r
- * @param m DOCUMENT ME!\r
- * @param psize DOCUMENT ME!\r
- */\r
- public static void displayMatrix(Graphics g, int[][] mat, int n, int m,\r
- int psize)\r
- {\r
- int max = -1000;\r
- int min = 1000;\r
-\r
- for (int i = 0; i < n; i++)\r
- {\r
- for (int j = 0; j < m; j++)\r
- {\r
- if (mat[i][j] >= max)\r
- {\r
- max = mat[i][j];\r
- }\r
-\r
- if (mat[i][j] <= min)\r
- {\r
- min = mat[i][j];\r
- }\r
- }\r
- }\r
-\r
- System.out.println(max + " " + min);\r
-\r
- for (int i = 0; i < n; i++)\r
- {\r
- for (int j = 0; j < m; j++)\r
- {\r
- int x = psize * i;\r
- int y = psize * j;\r
-\r
- // System.out.println(mat[i][j]);\r
- float score = (float) (mat[i][j] - min) / (float) (max - min);\r
- g.setColor(new Color(score, 0, 0));\r
- g.fillRect(x, y, psize, psize);\r
-\r
- // System.out.println(x + " " + y + " " + score);\r
- }\r
- }\r
- }\r
-}\r
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.7)
+ * Copyright (C) 2011 J Procter, AM Waterhouse, 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.analysis;
+
+import java.util.*;
+
+import java.awt.*;
+
+import jalview.datamodel.*;
+import jalview.schemes.*;
+import jalview.util.*;
+
+/**
+ *
+ *
+ * @author $author$
+ * @version $Revision$
+ */
+public class AlignSeq
+{
+ public static final String PEP = "pep";
+
+ public static final String DNA = "dna";
+
+ static String[] dna =
+ { "A", "C", "G", "T", "-" };
+
+ // "C", "T", "A", "G", "-"};
+ static String[] pep =
+ { "A", "R", "N", "D", "C", "Q", "E", "G", "H", "I", "L", "K", "M", "F",
+ "P", "S", "T", "W", "Y", "V", "B", "Z", "X", "-" };
+
+ int[][] score;
+
+ int[][] E;
+
+ int[][] F;
+
+ int[][] traceback;
+
+ int[] seq1;
+
+ int[] seq2;
+
+ SequenceI s1;
+
+ SequenceI s2;
+
+ public String s1str;
+
+ public String s2str;
+
+ int maxi;
+
+ int maxj;
+
+ int[] aseq1;
+
+ int[] aseq2;
+
+ public String astr1 = "";
+
+ public String astr2 = "";
+
+ /** DOCUMENT ME!! */
+ public int seq1start;
+
+ /** DOCUMENT ME!! */
+ public int seq1end;
+
+ /** DOCUMENT ME!! */
+ public int seq2start;
+
+ /** DOCUMENT ME!! */
+ public int seq2end;
+
+ int count;
+
+ /** DOCUMENT ME!! */
+ public int maxscore;
+
+ float pid;
+
+ int prev = 0;
+
+ int gapOpen = 120;
+
+ int gapExtend = 20;
+
+ int[][] lookup = ResidueProperties.getBLOSUM62();
+
+ String[] intToStr = pep;
+
+ int defInt = 23;
+
+ StringBuffer output = new StringBuffer();
+
+ String type;
+
+ private int[] charToInt;
+
+ /**
+ * Creates a new AlignSeq object.
+ *
+ * @param s1
+ * DOCUMENT ME!
+ * @param s2
+ * DOCUMENT ME!
+ * @param type
+ * DOCUMENT ME!
+ */
+ public AlignSeq(SequenceI s1, SequenceI s2, String type)
+ {
+ SeqInit(s1, s1.getSequenceAsString(), s2, s2.getSequenceAsString(),
+ type);
+ }
+
+ /**
+ * Creates a new AlignSeq object.
+ *
+ * @param s1
+ * DOCUMENT ME!
+ * @param s2
+ * DOCUMENT ME!
+ * @param type
+ * DOCUMENT ME!
+ */
+ public AlignSeq(SequenceI s1, String string1, SequenceI s2,
+ String string2, String type)
+ {
+ SeqInit(s1, string1.toUpperCase(), s2, string2.toUpperCase(), type);
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int getMaxScore()
+ {
+ return maxscore;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int getSeq2Start()
+ {
+ return seq2start;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int getSeq2End()
+ {
+ return seq2end;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int getSeq1Start()
+ {
+ return seq1start;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int getSeq1End()
+ {
+ return seq1end;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public String getOutput()
+ {
+ return output.toString();
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public String getAStr1()
+ {
+ return astr1;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public String getAStr2()
+ {
+ return astr2;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int[] getASeq1()
+ {
+ return aseq1;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int[] getASeq2()
+ {
+ return aseq2;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public SequenceI getS1()
+ {
+ return s1;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public SequenceI getS2()
+ {
+ return s2;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param s1
+ * DOCUMENT ME!
+ * @param string1
+ * - string to align for sequence1
+ * @param s2
+ * sequence 2
+ * @param string2
+ * - string to align for sequence2
+ * @param type
+ * DNA or PEPTIDE
+ */
+ public void SeqInit(SequenceI s1, String string1, SequenceI s2,
+ String string2, String type)
+ {
+ this.s1 = s1;
+ this.s2 = s2;
+ setDefaultParams(type);
+ SeqInit(string1, string2);
+ }
+
+ public void SeqInit(SequenceI s1, String string1, SequenceI s2,
+ String string2, ScoreMatrix scoreMatrix)
+ {
+ this.s1 = s1;
+ this.s2 = s2;
+ setType(scoreMatrix.isDNA() ? AlignSeq.DNA : AlignSeq.PEP);
+ lookup = scoreMatrix.getMatrix();
+ }
+
+ /**
+ * construct score matrix for string1 and string2 (after removing any existing
+ * gaps
+ *
+ * @param string1
+ * @param string2
+ */
+ private void SeqInit(String string1, String string2)
+ {
+ s1str = extractGaps(jalview.util.Comparison.GapChars, string1);
+ s2str = extractGaps(jalview.util.Comparison.GapChars, string2);
+
+ if (s1str.length() == 0 || s2str.length() == 0)
+ {
+ output.append("ALL GAPS: "
+ + (s1str.length() == 0 ? s1.getName() : " ")
+ + (s2str.length() == 0 ? s2.getName() : ""));
+ return;
+ }
+
+ // System.out.println("lookuip " + rt.freeMemory() + " "+ rt.totalMemory());
+ seq1 = new int[s1str.length()];
+
+ // System.out.println("seq1 " + rt.freeMemory() +" " + rt.totalMemory());
+ seq2 = new int[s2str.length()];
+
+ // System.out.println("seq2 " + rt.freeMemory() + " " + rt.totalMemory());
+ score = new int[s1str.length()][s2str.length()];
+
+ // System.out.println("score " + rt.freeMemory() + " " + rt.totalMemory());
+ E = new int[s1str.length()][s2str.length()];
+
+ // System.out.println("E " + rt.freeMemory() + " " + rt.totalMemory());
+ F = new int[s1str.length()][s2str.length()];
+ traceback = new int[s1str.length()][s2str.length()];
+
+ // System.out.println("F " + rt.freeMemory() + " " + rt.totalMemory());
+ seq1 = stringToInt(s1str, type);
+
+ // System.out.println("seq1 " + rt.freeMemory() + " " + rt.totalMemory());
+ seq2 = stringToInt(s2str, type);
+
+ // System.out.println("Seq2 " + rt.freeMemory() + " " + rt.totalMemory());
+ // long tstart = System.currentTimeMillis();
+ // calcScoreMatrix();
+ // long tend = System.currentTimeMillis();
+ // System.out.println("Time take to calculate score matrix = " +
+ // (tend-tstart) + " ms");
+ // printScoreMatrix(score);
+ // System.out.println();
+ // printScoreMatrix(traceback);
+ // System.out.println();
+ // printScoreMatrix(E);
+ // System.out.println();
+ // /printScoreMatrix(F);
+ // System.out.println();
+ // tstart = System.currentTimeMillis();
+ // traceAlignment();
+ // tend = System.currentTimeMillis();
+ // System.out.println("Time take to traceback alignment = " + (tend-tstart)
+ // + " ms");
+ }
+
+ private void setDefaultParams(String type)
+ {
+ setType(type);
+
+ if (type.equals(AlignSeq.PEP))
+ {
+ lookup = ResidueProperties.getDefaultPeptideMatrix();
+ }
+ else if (type.equals(AlignSeq.DNA))
+ {
+ lookup = ResidueProperties.getDefaultDnaMatrix();
+ }
+ }
+
+ private void setType(String type2)
+ {
+ this.type = type2;
+ if (type.equals(AlignSeq.PEP))
+ {
+ intToStr = pep;
+ charToInt = ResidueProperties.aaIndex;
+ defInt = 23;
+ }
+ else if (type.equals(AlignSeq.DNA))
+ {
+ intToStr = dna;
+ charToInt = ResidueProperties.nucleotideIndex;
+ defInt = 4;
+ }
+ else
+ {
+ output.append("Wrong type = dna or pep only");
+ throw new Error("Unknown Type " + type2
+ + " - dna or pep are the only allowed values.");
+ }
+ }
+
+ /**
+ * DOCUMENT ME!
+ */
+ public void traceAlignment()
+ {
+ // Find the maximum score along the rhs or bottom row
+ int max = -9999;
+
+ for (int i = 0; i < seq1.length; i++)
+ {
+ if (score[i][seq2.length - 1] > max)
+ {
+ max = score[i][seq2.length - 1];
+ maxi = i;
+ maxj = seq2.length - 1;
+ }
+ }
+
+ for (int j = 0; j < seq2.length; j++)
+ {
+ if (score[seq1.length - 1][j] > max)
+ {
+ max = score[seq1.length - 1][j];
+ maxi = seq1.length - 1;
+ maxj = j;
+ }
+ }
+
+ // System.out.println(maxi + " " + maxj + " " + score[maxi][maxj]);
+ int i = maxi;
+ int j = maxj;
+ int trace;
+ maxscore = score[i][j] / 10;
+
+ seq1end = maxi + 1;
+ seq2end = maxj + 1;
+
+ aseq1 = new int[seq1.length + seq2.length];
+ aseq2 = new int[seq1.length + seq2.length];
+
+ count = (seq1.length + seq2.length) - 1;
+
+ while ((i > 0) && (j > 0))
+ {
+ if ((aseq1[count] != defInt) && (i >= 0))
+ {
+ aseq1[count] = seq1[i];
+ astr1 = s1str.charAt(i) + astr1;
+ }
+
+ if ((aseq2[count] != defInt) && (j > 0))
+ {
+ aseq2[count] = seq2[j];
+ astr2 = s2str.charAt(j) + astr2;
+ }
+
+ trace = findTrace(i, j);
+
+ if (trace == 0)
+ {
+ i--;
+ j--;
+ }
+ else if (trace == 1)
+ {
+ j--;
+ aseq1[count] = defInt;
+ astr1 = "-" + astr1.substring(1);
+ }
+ else if (trace == -1)
+ {
+ i--;
+ aseq2[count] = defInt;
+ astr2 = "-" + astr2.substring(1);
+ }
+
+ count--;
+ }
+
+ seq1start = i + 1;
+ seq2start = j + 1;
+
+ if (aseq1[count] != defInt)
+ {
+ aseq1[count] = seq1[i];
+ astr1 = s1str.charAt(i) + astr1;
+ }
+
+ if (aseq2[count] != defInt)
+ {
+ aseq2[count] = seq2[j];
+ astr2 = s2str.charAt(j) + astr2;
+ }
+ }
+
+ /**
+ * DOCUMENT ME!
+ */
+ public void printAlignment(java.io.PrintStream os)
+ {
+ // TODO: Use original sequence characters rather than re-translated
+ // characters in output
+ // Find the biggest id length for formatting purposes
+ String s1id = s1.getName(), s2id = s2.getName();
+ int maxid = s1.getName().length();
+ if (s2.getName().length() > maxid)
+ {
+ maxid = s2.getName().length();
+ }
+ if (maxid > 30)
+ {
+ maxid = 30;
+ // JAL-527 - truncate the sequence ids
+ if (s1.getName().length() > maxid)
+ {
+ s1id = s1.getName().substring(0, 30);
+ }
+ if (s2.getName().length() > maxid)
+ {
+ s2id = s2.getName().substring(0, 30);
+ }
+ }
+ int len = 72 - maxid - 1;
+ int nochunks = ((aseq1.length - count) / len) + 1;
+ pid = 0;
+
+ output.append("Score = " + score[maxi][maxj] + "\n");
+ output.append("Length of alignment = " + (aseq1.length - count) + "\n");
+ output.append("Sequence ");
+ output.append(new Format("%" + maxid + "s").form(s1.getName()));
+ output.append(" : " + s1.getStart() + " - " + s1.getEnd()
+ + " (Sequence length = " + s1str.length() + ")\n");
+ output.append("Sequence ");
+ output.append(new Format("%" + maxid + "s").form(s2.getName()));
+ output.append(" : " + s2.getStart() + " - " + s2.getEnd()
+ + " (Sequence length = " + s2str.length() + ")\n\n");
+
+ for (int j = 0; j < nochunks; j++)
+ {
+ // Print the first aligned sequence
+ output.append(new Format("%" + (maxid) + "s").form(s1id) + " ");
+
+ for (int i = 0; i < len; i++)
+ {
+ if ((i + (j * len)) < astr1.length())
+ {
+ output.append(astr1.charAt(i + (j * len)));
+ }
+ }
+
+ output.append("\n");
+ output.append(new Format("%" + (maxid) + "s").form(" ") + " ");
+
+ // Print out the matching chars
+ for (int i = 0; i < len; i++)
+ {
+ if ((i + (j * len)) < astr1.length())
+ {
+ if (astr1.charAt(i + (j * len)) == astr2.charAt(i + (j * len))
+ && !jalview.util.Comparison.isGap(astr1.charAt(i
+ + (j * len))))
+ {
+ pid++;
+ output.append("|");
+ }
+ else if (type.equals("pep"))
+ {
+ if (ResidueProperties.getPAM250(astr1.charAt(i + (j * len)),
+ astr2.charAt(i + (j * len))) > 0)
+ {
+ output.append(".");
+ }
+ else
+ {
+ output.append(" ");
+ }
+ }
+ else
+ {
+ output.append(" ");
+ }
+ }
+ }
+
+ // Now print the second aligned sequence
+ output = output.append("\n");
+ output = output.append(new Format("%" + (maxid) + "s").form(s2id)
+ + " ");
+
+ for (int i = 0; i < len; i++)
+ {
+ if ((i + (j * len)) < astr2.length())
+ {
+ output.append(astr2.charAt(i + (j * len)));
+ }
+ }
+
+ output = output.append("\n\n");
+ }
+
+ pid = pid / (float) (aseq1.length - count) * 100;
+ output = output.append(new Format("Percentage ID = %2.2f\n\n")
+ .form(pid));
+
+ try
+ {
+ os.print(output.toString());
+ } catch (Exception ex)
+ {
+ }
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param mat
+ * DOCUMENT ME!
+ */
+ public void printScoreMatrix(int[][] mat)
+ {
+ int n = seq1.length;
+ int m = seq2.length;
+
+ for (int i = 0; i < n; i++)
+ {
+ // Print the top sequence
+ if (i == 0)
+ {
+ Format.print(System.out, "%8s", s2str.substring(0, 1));
+
+ for (int jj = 1; jj < m; jj++)
+ {
+ Format.print(System.out, "%5s", s2str.substring(jj, jj + 1));
+ }
+
+ System.out.println();
+ }
+
+ for (int j = 0; j < m; j++)
+ {
+ if (j == 0)
+ {
+ Format.print(System.out, "%3s", s1str.substring(i, i + 1));
+ }
+
+ Format.print(System.out, "%3d ", mat[i][j] / 10);
+ }
+
+ System.out.println();
+ }
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param i
+ * DOCUMENT ME!
+ * @param j
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int findTrace(int i, int j)
+ {
+ int t = 0;
+ int max = score[i - 1][j - 1] + (lookup[seq1[i]][seq2[j]] * 10);
+
+ if (F[i][j] > max)
+ {
+ max = F[i][j];
+ t = -1;
+ }
+ else if (F[i][j] == max)
+ {
+ if (prev == -1)
+ {
+ max = F[i][j];
+ t = -1;
+ }
+ }
+
+ if (E[i][j] >= max)
+ {
+ max = E[i][j];
+ t = 1;
+ }
+ else if (E[i][j] == max)
+ {
+ if (prev == 1)
+ {
+ max = E[i][j];
+ t = 1;
+ }
+ }
+
+ prev = t;
+
+ return t;
+ }
+
+ /**
+ * DOCUMENT ME!
+ */
+ public void calcScoreMatrix()
+ {
+ int n = seq1.length;
+ int m = seq2.length;
+
+ // top left hand element
+ score[0][0] = lookup[seq1[0]][seq2[0]] * 10;
+ E[0][0] = -gapExtend;
+ F[0][0] = 0;
+
+ // Calculate the top row first
+ for (int j = 1; j < m; j++)
+ {
+ // What should these values be? 0 maybe
+ E[0][j] = max(score[0][j - 1] - gapOpen, E[0][j - 1] - gapExtend);
+ F[0][j] = -gapExtend;
+
+ score[0][j] = max(lookup[seq1[0]][seq2[j]] * 10, -gapOpen, -gapExtend);
+
+ traceback[0][j] = 1;
+ }
+
+ // Now do the left hand column
+ for (int i = 1; i < n; i++)
+ {
+ E[i][0] = -gapOpen;
+ F[i][0] = max(score[i - 1][0] - gapOpen, F[i - 1][0] - gapExtend);
+
+ score[i][0] = max(lookup[seq1[i]][seq2[0]] * 10, E[i][0], F[i][0]);
+ traceback[i][0] = -1;
+ }
+
+ // Now do all the other rows
+ for (int i = 1; i < n; i++)
+ {
+ for (int j = 1; j < m; j++)
+ {
+ E[i][j] = max(score[i][j - 1] - gapOpen, E[i][j - 1] - gapExtend);
+ F[i][j] = max(score[i - 1][j] - gapOpen, F[i - 1][j] - gapExtend);
+
+ score[i][j] = max(score[i - 1][j - 1]
+ + (lookup[seq1[i]][seq2[j]] * 10), E[i][j], F[i][j]);
+ traceback[i][j] = findTrace(i, j);
+ }
+ }
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param gapChar
+ * DOCUMENT ME!
+ * @param seq
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public static String extractGaps(String gapChar, String seq)
+ {
+ StringTokenizer str = new StringTokenizer(seq, gapChar);
+ StringBuffer newString = new StringBuffer();
+
+ while (str.hasMoreTokens())
+ {
+ newString.append(str.nextToken());
+ }
+
+ return newString.toString();
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param i1
+ * DOCUMENT ME!
+ * @param i2
+ * DOCUMENT ME!
+ * @param i3
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int max(int i1, int i2, int i3)
+ {
+ int max = i1;
+
+ if (i2 > i1)
+ {
+ max = i2;
+ }
+
+ if (i3 > max)
+ {
+ max = i3;
+ }
+
+ return max;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param i1
+ * DOCUMENT ME!
+ * @param i2
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int max(int i1, int i2)
+ {
+ int max = i1;
+
+ if (i2 > i1)
+ {
+ max = i2;
+ }
+
+ return max;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param s
+ * DOCUMENT ME!
+ * @param type
+ * DOCUMENT ME!
+ *
+ * @return DOCUMENT ME!
+ */
+ public int[] stringToInt(String s, String type)
+ {
+ int[] seq1 = new int[s.length()];
+
+ for (int i = 0; i < s.length(); i++)
+ {
+ // String ss = s.substring(i, i + 1).toUpperCase();
+ char c = s.charAt(i);
+ if ('a' <= c && c <= 'z')
+ {
+ // TO UPPERCASE !!!
+ c -= ('a' - 'A');
+ }
+
+ try
+ {
+ seq1[i] = charToInt[c]; // set accordingly from setType
+ if (seq1[i] < 0 || seq1[i] > defInt) // set from setType: 23 for
+ // peptides, or 4 for NA.
+ {
+ seq1[i] = defInt;
+ }
+
+ } catch (Exception e)
+ {
+ seq1[i] = defInt;
+ }
+ }
+
+ return seq1;
+ }
+
+ /**
+ * DOCUMENT ME!
+ *
+ * @param g
+ * DOCUMENT ME!
+ * @param mat
+ * DOCUMENT ME!
+ * @param n
+ * DOCUMENT ME!
+ * @param m
+ * DOCUMENT ME!
+ * @param psize
+ * DOCUMENT ME!
+ */
+ public static void displayMatrix(Graphics g, int[][] mat, int n, int m,
+ int psize)
+ {
+ int max = -1000;
+ int min = 1000;
+
+ for (int i = 0; i < n; i++)
+ {
+ for (int j = 0; j < m; j++)
+ {
+ if (mat[i][j] >= max)
+ {
+ max = mat[i][j];
+ }
+
+ if (mat[i][j] <= min)
+ {
+ min = mat[i][j];
+ }
+ }
+ }
+
+ System.out.println(max + " " + min);
+
+ for (int i = 0; i < n; i++)
+ {
+ for (int j = 0; j < m; j++)
+ {
+ int x = psize * i;
+ int y = psize * j;
+
+ // System.out.println(mat[i][j]);
+ float score = (float) (mat[i][j] - min) / (float) (max - min);
+ g.setColor(new Color(score, 0, 0));
+ g.fillRect(x, y, psize, psize);
+
+ // System.out.println(x + " " + y + " " + score);
+ }
+ }
+ }
+}
git://source.jalview.org / jalview.git / blobdiff