2 * Jalview - A Sequence Alignment Editor and Viewer (Version 2.8)
3 * Copyright (C) 2012 J Procter, AM Waterhouse, LM Lui, J Engelhardt, G Barton, M Clamp, S Searle
5 * This file is part of Jalview.
7 * Jalview is free software: you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License
9 * as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
11 * Jalview is distributed in the hope that it will be useful, but
12 * WITHOUT ANY WARRANTY; without even the implied warranty
13 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
14 * PURPOSE. See the GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License along with Jalview. If not, see <http://www.gnu.org/licenses/>.
18 package jalview.analysis;
22 import jalview.datamodel.*;
25 * Takes in a vector or array of sequences and column start and column end and
26 * returns a new Hashtable[] of size maxSeqLength, if Hashtable not supplied.
27 * This class is used extensively in calculating alignment colourschemes that
28 * depend on the amount of conservation in each alignment column.
33 public class AAFrequency
35 // No need to store 1000s of strings which are not
36 // visible to the user.
37 public static final String MAXCOUNT = "C";
39 public static final String MAXRESIDUE = "R";
41 public static final String PID_GAPS = "G";
43 public static final String PID_NOGAPS = "N";
45 public static final String PROFILE = "P";
47 public static final Hashtable[] calculate(List<SequenceI> list,
50 return calculate(list, start, end, false);
53 public static final Hashtable[] calculate(List<SequenceI> sequences,
54 int start, int end, boolean profile)
56 SequenceI[] seqs = new SequenceI[sequences.size()];
58 synchronized (sequences)
60 for (int i = 0; i < sequences.size(); i++)
62 seqs[i] = sequences.get(i);
63 if (seqs[i].getLength() > width)
65 width = seqs[i].getLength();
69 Hashtable[] reply = new Hashtable[width];
76 calculate(seqs, start, end, reply, profile);
81 public static final void calculate(SequenceI[] sequences, int start,
82 int end, Hashtable[] result)
84 calculate(sequences, start, end, result, false);
87 public static final void calculate(SequenceI[] sequences, int start,
88 int end, Hashtable[] result, boolean profile)
90 Hashtable residueHash;
91 int maxCount, nongap, i, j, v, jSize = sequences.length;
96 int[] values = new int[255];
100 for (i = start; i < end; i++)
102 residueHash = new Hashtable();
106 values = new int[255];
108 for (j = 0; j < jSize; j++)
110 if (sequences[j] == null)
113 .println("WARNING: Consensus skipping null sequence - possible race condition.");
116 seq = sequences[j].getSequence();
121 if (c == '.' || c == ' ')
131 else if ('a' <= c && c <= 'z')
133 c -= 32; // ('a' - 'A');
146 for (v = 'A'; v < 'Z'; v++)
148 if (values[v] < 2 || values[v] < maxCount)
153 if (values[v] > maxCount)
155 maxResidue = String.valueOf((char) v);
157 else if (values[v] == maxCount)
159 maxResidue += String.valueOf((char) v);
161 maxCount = values[v];
164 if (maxResidue.length() == 0)
170 residueHash.put(PROFILE, new int[][]
172 { jSize, nongap } });
174 residueHash.put(MAXCOUNT, new Integer(maxCount));
175 residueHash.put(MAXRESIDUE, maxResidue);
177 percentage = ((float) maxCount * 100) / jSize;
178 residueHash.put(PID_GAPS, new Float(percentage));
180 percentage = ((float) maxCount * 100) / nongap;
181 residueHash.put(PID_NOGAPS, new Float(percentage));
182 result[i] = residueHash;
187 * Compute all or part of the annotation row from the given consensus
191 * - pre-allocated annotation row
195 * @param ignoreGapsInConsensusCalculation
196 * @param includeAllConsSymbols
198 public static void completeConsensus(AlignmentAnnotation consensus,
199 Hashtable[] hconsensus, int iStart, int width,
200 boolean ignoreGapsInConsensusCalculation,
201 boolean includeAllConsSymbols)
203 completeConsensus(consensus, hconsensus, iStart, width,
204 ignoreGapsInConsensusCalculation, includeAllConsSymbols, null); // new
206 // { 'A', 'C', 'G', 'T', 'U' });
209 public static void completeConsensus(AlignmentAnnotation consensus,
210 Hashtable[] hconsensus, int iStart, int width,
211 boolean ignoreGapsInConsensusCalculation,
212 boolean includeAllConsSymbols, char[] alphabet)
215 if (consensus == null || consensus.annotations == null
216 || consensus.annotations.length < width)
218 // called with a bad alignment annotation row - wait for it to be
219 // initialised properly
222 for (int i = iStart; i < width; i++)
225 if (i >= hconsensus.length || ((hci = hconsensus[i]) == null))
227 // happens if sequences calculated over were shorter than alignment
229 consensus.annotations[i] = null;
235 if (ignoreGapsInConsensusCalculation)
237 fv = (Float) hci.get(AAFrequency.PID_NOGAPS);
241 fv = (Float) hci.get(AAFrequency.PID_GAPS);
245 consensus.annotations[i] = null;
246 // data has changed below us .. give up and
249 value = fv.floatValue();
250 String maxRes = hci.get(AAFrequency.MAXRESIDUE).toString();
251 String mouseOver = hci.get(AAFrequency.MAXRESIDUE) + " ";
252 if (maxRes.length() > 1)
254 mouseOver = "[" + maxRes + "] ";
257 int[][] profile = (int[][]) hci.get(AAFrequency.PROFILE);
258 if (profile != null && includeAllConsSymbols)
261 if (alphabet != null)
263 for (int c = 0; c < alphabet.length; c++)
265 tval = profile[0][alphabet[c]] * 100f
266 / profile[1][ignoreGapsInConsensusCalculation ? 1 : 0];
267 mouseOver += ((c == 0) ? "" : "; ") + alphabet[c] + " "
268 + ((int) tval) + "%";
273 Object[] ca = new Object[profile[0].length];
274 float[] vl = new float[profile[0].length];
275 for (int c = 0; c < ca.length; c++)
279 vl[c] = profile[0][c];
282 jalview.util.QuickSort.sort(vl, ca);
283 for (int p = 0, c = ca.length - 1; profile[0][((char[]) ca[c])[0]] > 0; c--)
285 if (((char[]) ca[c])[0] != '-')
287 tval = profile[0][((char[]) ca[c])[0]]
289 / profile[1][ignoreGapsInConsensusCalculation ? 1 : 0];
290 mouseOver += ((p == 0) ? "" : "; ") + ((char[]) ca[c])[0]
291 + " " + ((int) tval) + "%";
301 mouseOver += ((int) value + "%");
303 consensus.annotations[i] = new Annotation(maxRes, mouseOver, ' ',
309 * get the sorted profile for the given position of the consensus
314 public static int[] extractProfile(Hashtable hconsensus,
315 boolean ignoreGapsInConsensusCalculation)
317 int[] rtnval = new int[64];
318 int[][] profile = (int[][]) hconsensus.get(AAFrequency.PROFILE);
321 Object[] ca = new Object[profile[0].length];
322 float[] vl = new float[profile[0].length];
323 for (int c = 0; c < ca.length; c++)
327 vl[c] = profile[0][c];
330 jalview.util.QuickSort.sort(vl, ca);
333 for (int c = ca.length - 1; profile[0][((char[]) ca[c])[0]] > 0; c--)
335 if (((char[]) ca[c])[0] != '-')
337 rtnval[rtnval[0]++] = ((char[]) ca[c])[0];
338 rtnval[rtnval[0]] = (int) (profile[0][((char[]) ca[c])[0]] * 100f / profile[1][ignoreGapsInConsensusCalculation ? 1
340 rtnval[1] += rtnval[rtnval[0]++];