X-Git-Url: http://source.jalview.org/gitweb/?a=blobdiff_plain;f=src%2Fjalview%2Fanalysis%2FAAFrequency.java;h=11e7a3d8514be3d0e14e3c249278a64dce353c1e;hb=153dd62dc91da13ae732600e6ea55ddbe15eab39;hp=edcc50fa5cc9bb06890a3a5afa403161546a9330;hpb=dbba2779042727476e41d2cdfa7346780ca00cbd;p=jalview.git
diff --git a/src/jalview/analysis/AAFrequency.java b/src/jalview/analysis/AAFrequency.java
index edcc50f..11e7a3d 100755
--- a/src/jalview/analysis/AAFrequency.java
+++ b/src/jalview/analysis/AAFrequency.java
@@ -1,393 +1,327 @@
-package jalview.analysis;
-
-import jalview.jbgui.*;
-import jalview.datamodel.*;
-import jalview.io.*;
-import jalview.analysis.*;
-
-import java.awt.*;
-import java.applet.Applet;
-import java.util.*;
-import java.net.*;
-import java.io.*;
-
-public class AAFrequency {
-
- // Takes in a vector of sequences and column start and column end
- // and returns a vector of size (end-start+1). Each element of the
- // vector contains a hashtable with the keys being residues and
- // the values being the count of each residue in that column.
- // This class is used extensively in calculating alignment colourschemes
- // that depend on the amount of conservation in each alignment column.
-
-
- public static Vector calculate(Vector sequences,int start,int end) {
-
- Vector result = new Vector();
-
- for (int i = start;i <= end; i++)
- {
-
- Hashtable residueHash = new Hashtable();
- int maxCount = 0;
- String maxResidue = "-";
- int nongap = 0;
- for (int j=0; j < sequences.size(); j++)
- {
-
- if (sequences.elementAt(j) instanceof Sequence)
- {
- Sequence s = (Sequence)sequences.elementAt(j);
-
- if (s.getSequence().length() > i)
- {
-
- String res = s.getSequence().charAt(i)+"";
-
- if (!jalview.util.Comparison.isGap(res.charAt(0)))
- nongap++;
- else
- res = "-"; // we always use this for gaps in the property vectors
-
- if (residueHash.containsKey(res))
- {
-
- int count = ((Integer)residueHash.get(res)).intValue() ;
- count++;
-
- if (!jalview.util.Comparison.isGap(res.charAt(0)) && count >= maxCount)
- {
-
- if(count>maxCount)
- maxResidue = res;
- else if(maxResidue.indexOf(res)==-1)
- maxResidue += res;
-
- maxCount = count;
- }
-
- residueHash.put(res,new Integer(count));
- }
- else
- residueHash.put(res,new Integer(1));
-
-
- }
- else
- {
- if (residueHash.containsKey("-"))
- {
- int count = ((Integer)residueHash.get("-")).intValue() ;
- count++;
- residueHash.put("-",new Integer(count));
- }
- else
- residueHash.put("-",new Integer(1));
-
- }
- }
- }
-
- residueHash.put("maxCount",new Integer(maxCount));
- if(maxCount<0)
- System.out.println("asasa "+maxCount);
- residueHash.put("maxResidue", maxResidue);
- residueHash.put("size", new Integer(sequences.size()));
- residueHash.put("nongap", new Integer(nongap));
- result.addElement(residueHash);
- }
-
- return result;
- }
-
- public static Vector calculatePID(SequenceI refseq,Vector sequences,int window,int start,int end) {
-
- Vector result = new Vector();
-
- boolean init = true;
-
-
- Vector prev = null;
-
- for (int i = start;i <= end; i++) {
- Vector values = new Vector();
-
- result.addElement(values);
-
- // If start < window/2 then set value to zero.
-
- if (i< window/2 || i >= refseq.getSequence().length()-window/2) {
- for (int j = 0; j < sequences.size(); j++) {
- values.addElement(new Integer(0));
- }
- } else if (init == true) {
- init = false;
-
- int winstart = i-window/2;
- int winend = i+window/2;
-
- if (window%2 != 0) {
- winend++;
- }
-
- for (int j = 0; j < sequences.size(); j++) {
- values.addElement(new Integer(0));
- }
-
- for (int k = winstart; k <= winend; k++) {
- String refchar = refseq.getSequence().substring(k,k+1);
- if (jalview.util.Comparison.isGap(refchar.charAt(0)))
- refchar="-";
- else {
- for (int j = 0; j < sequences.size(); j++) {
-
- Sequence s = (Sequence)sequences.elementAt(j);
-
- if (s.getSequence().length() > k) {
-
- String res = s.getSequence().substring(k,k+1); // no gapchar test needed
-
- if (res.equals(refchar)) {
- int val = ((Integer)values.elementAt(j)).intValue();
- val++;
- values.setElementAt(new Integer(val),j);
- }
- }
- }
- }
-
- }
-
- prev = values;
- } else {
- int winstart = i-window/2;
- int winend = i+window/2;
-
- if (window%2 != 0) {
- winend++;
- }
- // We need to take the previous set of values
- // subtract the pid at winstart-1
- // and add the pid at winend;
-
- String pre_refchar = refseq.getSequence().substring(winstart-1,winstart);
- String pos_refchar = "-";
-
- if (refseq.getSequence().length() > winend) {
- pos_refchar = refseq.getSequence().substring(winend,winend+1);
- }
-
- for (int j = 0; j < sequences.size(); j++) {
- // First copy the pid value from i-1
-
- int val = ((Integer)prev.elementAt(j)).intValue();
-
- Sequence s = (Sequence)sequences.elementAt(j);
-
- String pre_char = s.getSequence().substring(winstart-1,winstart);
-
- String pos_char = "-";
-
- if (s.getSequence().length() > winend) {
- pos_char = s.getSequence().substring(winend,winend+1);
- }
-
- // Now substract 1 if the chars at winstart-1 match
-
- if (jalview.util.Comparison.isGap(pre_refchar.charAt(0)) == false
- && pre_char.equals(pre_refchar)) {
- val--;
- }
-
- if (jalview.util.Comparison.isGap(pos_refchar.charAt(0)) == false
- && pos_char.equals(pos_refchar)) {
- val++;
- }
-
- values.addElement(new Integer(val));
-
-
- }
- prev = values;
- }
- }
-
- return result;
- }
-
- public static Hashtable findBlocks(Vector seqs, int start, int end,Vector exc) {
-
- // start and end are in real (not relative coords);
-
- // The coords in the hashtable that is returned are in relative coords
- // i.e. start from 0
-
- Hashtable blocks = new Hashtable();
-
- boolean prev = false;
- int bstart = -1;
-
- for (int i = start; i <= end ; i++) {
- SequenceI seq = (SequenceI)seqs.elementAt(0);
-
- char c = seq.getCharAt(i);
-
- boolean found = true;
-
- int j = 1;
-
- while (j < seqs.size() && found == true) {
-
- SequenceI jseq = (SequenceI)seqs.elementAt(j);
-
- if (!exc.contains(jseq)) {
-
- char cc = jseq.getCharAt(i);
-
- if ( cc != c) {
- found = false;
- }
- }
- j++;
- }
-
-
- if (prev == false && found == true) {
- bstart = i;
- } else if (prev == true && found == false && bstart != -1) {
-
- int blockstart = bstart-start;
- int blocklen = i-bstart;
-
- //System.out.println("Start len " + blockstart + " " + blocklen);
-
- for (int jj = blockstart; jj < blockstart + blocklen;jj++) {
- blocks.put(new Integer(jj),new Integer(blocklen));
- }
-
- bstart = -1;
- }
- prev = found;
- }
-
- if (bstart != -1) {
-
- int blockstart = bstart-start;
- int blocklen = end-bstart;
-
- // System.out.println("Start len " + blockstart + " " + blocklen);
-
- for (int jj = blockstart; jj < blockstart + blocklen;jj++) {
- blocks.put(new Integer(blockstart),new Integer(blocklen));
- }
-
- }
- return blocks;
- }
-
-
-
- public static Hashtable findKmerCount(SequenceI seq, int start, int end,int window, int step,Vector kmers) {
-
- int tmpstart = start;
- Hashtable vals = new Hashtable();
-
- while (tmpstart <= end) {
-
- String tmpstr = seq.getSequence().substring(tmpstart-window/2,tmpstart+window/2);
-
- int count = 0;
-
- //System.out.println("Str " + tmpstr);
-
- for (int ii = 0; ii < kmers.size(); ii++) {
- String kmer = ((SequenceI)kmers.elementAt(ii)).getSequence();
-
- int i = -1;
-
- while (tmpstr.indexOf(kmer,i) != -1) {
- i = tmpstr.indexOf(kmer,i);
-
- i++;
- count++;
- }
- ii++;
- }
- vals.put(new Integer(tmpstart),new Integer(count));
- tmpstart += step;
- }
- return vals;
- }
-
- public static Hashtable findBlockStarts(Vector seqs, int start, int end,Vector exc) {
-
- // start and end are in real (not relative coords);
-
- // The coords in the hashtable that is returned are in relative coords
- // i.e. start from 0
-
- Hashtable blocks = new Hashtable();
-
- boolean prev = false;
- int bstart = -1;
-
- for (int i = start; i <= end ; i++) {
- SequenceI seq = (SequenceI)seqs.elementAt(0);
-
- char c = seq.getCharAt(i);
-
- boolean found = true;
-
- int j = 1;
-
- while (j < seqs.size() && found == true) {
-
- SequenceI jseq = (SequenceI)seqs.elementAt(j);
-
- if (!exc.contains(jseq)) {
-
- char cc = jseq.getCharAt(i);
-
- if ( cc != c) {
- found = false;
- }
- }
- j++;
- }
-
-
- if (prev == false && found == true) {
- bstart = i;
- } else if (prev == true && found == false && bstart != -1) {
-
- int blockstart = bstart-start;
- int blocklen = i-bstart;
-
- // System.out.println("Start len " + blockstart + " " + blocklen);
-
- //for (int jj = blockstart; jj < blockstart + blocklen;jj++) {
- blocks.put(new Integer(blockstart),new Integer(blocklen));
- // }
-
- bstart = -1;
- }
- prev = found;
- }
-
- if (bstart != -1) {
-
- int blockstart = bstart-start;
- int blocklen = end-bstart;
-
- // System.out.println("Start len " + blockstart + " " + blocklen);
-
- //for (int jj = blockstart; jj < blockstart + blocklen;jj++) {
- blocks.put(new Integer(blockstart),new Integer(blocklen));
- // }
-
- }
- return blocks;
- }
-
-}
-
+/*
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.6)
+ * Copyright (C) 2010 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 .
+ */
+package jalview.analysis;
+
+import java.util.*;
+
+import jalview.datamodel.*;
+
+/**
+ * Takes in a vector or array of sequences and column start and column end and
+ * returns a new Hashtable[] of size maxSeqLength, if Hashtable not supplied.
+ * This class is used extensively in calculating alignment colourschemes that
+ * depend on the amount of conservation in each alignment column.
+ *
+ * @author $author$
+ * @version $Revision$
+ */
+public class AAFrequency
+{
+ // No need to store 1000s of strings which are not
+ // visible to the user.
+ public static final String MAXCOUNT = "C";
+
+ public static final String MAXRESIDUE = "R";
+
+ public static final String PID_GAPS = "G";
+
+ public static final String PID_NOGAPS = "N";
+
+ public static final String PROFILE = "P";
+
+ public static final Hashtable[] calculate(Vector sequences, int start,
+ int end)
+ {
+ return calculate(sequences, start, end, false);
+ }
+
+ public static final Hashtable[] calculate(Vector sequences, int start,
+ int end, boolean profile)
+ {
+ SequenceI[] seqs = new SequenceI[sequences.size()];
+ int width = 0;
+ for (int i = 0; i < sequences.size(); i++)
+ {
+ seqs[i] = (SequenceI) sequences.elementAt(i);
+ if (seqs[i].getLength() > width)
+ {
+ width = seqs[i].getLength();
+ }
+ }
+
+ Hashtable[] reply = new Hashtable[width];
+
+ if (end >= width)
+ {
+ end = width;
+ }
+
+ calculate(seqs, start, end, reply, profile);
+
+ return reply;
+ }
+
+ public static final void calculate(SequenceI[] sequences, int start,
+ int end, Hashtable[] result)
+ {
+ calculate(sequences, start, end, result, false);
+ }
+
+ public static final void calculate(SequenceI[] sequences, int start,
+ int end, Hashtable[] result, boolean profile)
+ {
+ Hashtable residueHash;
+ int maxCount, nongap, i, j, v, jSize = sequences.length;
+ String maxResidue;
+ char c;
+ float percentage;
+
+ int[] values = new int[255];
+
+ char[] seq;
+
+ for (i = start; i < end; i++)
+ {
+ residueHash = new Hashtable();
+ maxCount = 0;
+ maxResidue = "";
+ nongap = 0;
+ values = new int[255];
+
+ for (j = 0; j < jSize; j++)
+ {
+ seq = sequences[j].getSequence();
+ if (seq.length > i)
+ {
+ c = seq[i];
+
+ if (c == '.' || c == ' ')
+ {
+ c = '-';
+ }
+
+ if (c == '-')
+ {
+ values['-']++;
+ continue;
+ }
+ else if ('a' <= c && c <= 'z')
+ {
+ c -= 32; // ('a' - 'A');
+ }
+
+ nongap++;
+ values[c]++;
+
+ }
+ else
+ {
+ values['-']++;
+ }
+ }
+
+ for (v = 'A'; v < 'Z'; v++)
+ {
+ if (values[v] < 2 || values[v] < maxCount)
+ {
+ continue;
+ }
+
+ if (values[v] > maxCount)
+ {
+ maxResidue = String.valueOf((char) v);
+ }
+ else if (values[v] == maxCount)
+ {
+ maxResidue += String.valueOf((char) v);
+ }
+ maxCount = values[v];
+ }
+
+ if (maxResidue.length() == 0)
+ {
+ maxResidue = "-";
+ }
+ if (profile)
+ {
+ residueHash.put(PROFILE, new int[][]
+ { values, new int[]
+ { jSize, nongap } });
+ }
+ residueHash.put(MAXCOUNT, new Integer(maxCount));
+ residueHash.put(MAXRESIDUE, maxResidue);
+
+ percentage = ((float) maxCount * 100) / (float) jSize;
+ residueHash.put(PID_GAPS, new Float(percentage));
+
+ percentage = ((float) maxCount * 100) / (float) nongap;
+ residueHash.put(PID_NOGAPS, new Float(percentage));
+ result[i] = residueHash;
+ }
+ }
+
+ /**
+ * Compute all or part of the annotation row from the given consensus
+ * hashtable
+ *
+ * @param consensus
+ * - pre-allocated annotation row
+ * @param hconsensus
+ * @param iStart
+ * @param width
+ * @param ignoreGapsInConsensusCalculation
+ * @param includeAllConsSymbols
+ */
+ public static void completeConsensus(AlignmentAnnotation consensus,
+ Hashtable[] hconsensus, int iStart, int width,
+ boolean ignoreGapsInConsensusCalculation,
+ boolean includeAllConsSymbols)
+ {
+ completeConsensus(consensus, hconsensus, iStart, width,
+ ignoreGapsInConsensusCalculation, includeAllConsSymbols, null); // new
+ // char[]
+ // { 'A', 'C', 'G', 'T', 'U' });
+ }
+
+ public static void completeConsensus(AlignmentAnnotation consensus,
+ Hashtable[] hconsensus, int iStart, int width,
+ boolean ignoreGapsInConsensusCalculation,
+ boolean includeAllConsSymbols, char[] alphabet)
+ {
+ float tval, value;
+ if (consensus==null || consensus.annotations==null || consensus.annotations.length=hconsensus.length) {
+ // happens if sequences calculated over were shorter than alignment width
+ consensus.annotations[i]=null;
+ continue;
+ }
+ value = 0;
+ if (ignoreGapsInConsensusCalculation)
+ {
+ value = ((Float) hconsensus[i].get(AAFrequency.PID_NOGAPS))
+ .floatValue();
+ }
+ else
+ {
+ value = ((Float) hconsensus[i].get(AAFrequency.PID_GAPS))
+ .floatValue();
+ }
+
+ String maxRes = hconsensus[i].get(AAFrequency.MAXRESIDUE).toString();
+ String mouseOver = hconsensus[i].get(AAFrequency.MAXRESIDUE) + " ";
+ if (maxRes.length() > 1)
+ {
+ mouseOver = "[" + maxRes + "] ";
+ maxRes = "+";
+ }
+ int[][] profile = (int[][]) hconsensus[i].get(AAFrequency.PROFILE);
+ if (profile != null && includeAllConsSymbols)
+ {
+ mouseOver = "";
+ if (alphabet != null)
+ {
+ for (int c = 0; c < alphabet.length; c++)
+ {
+ tval = ((float) profile[0][alphabet[c]])
+ * 100f
+ / (float) profile[1][ignoreGapsInConsensusCalculation ? 1
+ : 0];
+ mouseOver += ((c == 0) ? "" : "; ") + alphabet[c] + " "
+ + ((int) tval) + "%";
+ }
+ }
+ else
+ {
+ Object[] ca = new Object[profile[0].length];
+ float[] vl = new float[profile[0].length];
+ for (int c = 0; c < ca.length; c++)
+ {
+ ca[c] = new char[]
+ { (char) c };
+ vl[c] = (float) profile[0][c];
+ }
+ ;
+ jalview.util.QuickSort.sort(vl, ca);
+ for (int p = 0, c = ca.length - 1; profile[0][((char[]) ca[c])[0]] > 0; c--)
+ {
+ if (((char[]) ca[c])[0] != '-')
+ {
+ tval = ((float) profile[0][((char[]) ca[c])[0]])
+ * 100f
+ / (float) profile[1][ignoreGapsInConsensusCalculation ? 1
+ : 0];
+ mouseOver += ((p == 0) ? "" : "; ") + ((char[]) ca[c])[0]
+ + " " + ((int) tval) + "%";
+ p++;
+
+ }
+ }
+
+ }
+ }
+ else
+ {
+ mouseOver += ((int) value + "%");
+ }
+ consensus.annotations[i] = new Annotation(maxRes, mouseOver, ' ',
+ value);
+ }
+ }
+
+ /**
+ * get the sorted profile for the given position of the consensus
+ *
+ * @param hconsensus
+ * @return
+ */
+ public static int[] extractProfile(Hashtable hconsensus,
+ boolean ignoreGapsInConsensusCalculation)
+ {
+ int[] rtnval = new int[64];
+ int[][] profile = (int[][]) hconsensus.get(AAFrequency.PROFILE);
+ if (profile == null)
+ return null;
+ Object[] ca = new Object[profile[0].length];
+ float[] vl = new float[profile[0].length];
+ for (int c = 0; c < ca.length; c++)
+ {
+ ca[c] = new char[]
+ { (char) c };
+ vl[c] = (float) profile[0][c];
+ }
+ ;
+ jalview.util.QuickSort.sort(vl, ca);
+ rtnval[0] = 1;
+ for (int c = ca.length - 1; profile[0][((char[]) ca[c])[0]] > 0; c--)
+ {
+ if (((char[]) ca[c])[0] != '-')
+ {
+ rtnval[rtnval[0]++] = ((char[]) ca[c])[0];
+ rtnval[rtnval[0]++] = (int) (((float) profile[0][((char[]) ca[c])[0]]) * 100f / (float) profile[1][ignoreGapsInConsensusCalculation ? 1
+ : 0]);
+ }
+ }
+ return rtnval;
+ }
+}