2 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3 * Copyright (C) $$Year-Rel$$ The Jalview Authors
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
10 * of the License, or (at your option) any later version.
12 * Jalview is distributed in the hope that it will be useful, but
13 * WITHOUT ANY WARRANTY; without even the implied warranty
14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
15 * PURPOSE. See the GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
19 * The Jalview Authors are detailed in the 'AUTHORS' file.
21 package jalview.analysis;
23 import jalview.analysis.ResidueCount.SymbolCounts;
24 import jalview.datamodel.AlignedCodonFrame;
25 import jalview.datamodel.AlignmentAnnotation;
26 import jalview.datamodel.AlignmentI;
27 import jalview.datamodel.Annotation;
28 import jalview.datamodel.SequenceI;
29 import jalview.ext.android.SparseIntArray;
30 import jalview.util.Comparison;
31 import jalview.util.Format;
32 import jalview.util.MappingUtils;
33 import jalview.util.QuickSort;
35 import java.util.Arrays;
36 import java.util.Hashtable;
37 import java.util.List;
40 * Takes in a vector or array of sequences and column start and column end and
41 * returns a new Hashtable[] of size maxSeqLength, if Hashtable not supplied.
42 * This class is used extensively in calculating alignment colourschemes that
43 * depend on the amount of conservation in each alignment column.
48 public class AAFrequency
50 public static final String MAXCOUNT = "C";
52 public static final String MAXRESIDUE = "R";
54 public static final String PID_GAPS = "G";
56 public static final String PID_NOGAPS = "N";
58 public static final String PROFILE = "P";
60 public static final String ENCODED_CHARS = "E";
63 * Quick look-up of String value of char 'A' to 'Z'
65 private static final String[] CHARS = new String['Z' - 'A' + 1];
69 for (char c = 'A'; c <= 'Z'; c++)
71 CHARS[c - 'A'] = String.valueOf(c);
75 public static final Profile[] calculate(List<SequenceI> list,
78 return calculate(list, start, end, false);
81 public static final Profile[] calculate(List<SequenceI> sequences,
82 int start, int end, boolean profile)
84 SequenceI[] seqs = new SequenceI[sequences.size()];
86 synchronized (sequences)
88 for (int i = 0; i < sequences.size(); i++)
90 seqs[i] = sequences.get(i);
91 if (seqs[i].getLength() > width)
93 width = seqs[i].getLength();
97 Profile[] reply = new Profile[width];
104 calculate(seqs, start, end, reply, profile);
110 * Calculate the consensus symbol(s) for each column in the given range.
116 * array in which to store profile per column
117 * @param saveFullProfile
118 * if true, store all symbol counts
120 public static final void calculate(final SequenceI[] sequences,
121 int start, int end, Profile[] result, boolean saveFullProfile)
123 // long now = System.currentTimeMillis();
124 int seqCount = sequences.length;
125 boolean nucleotide = false;
126 int nucleotideCount = 0;
127 int peptideCount = 0;
129 for (int column = start; column < end; column++)
132 * Apply a heuristic to detect nucleotide data (which can
133 * be counted in more compact arrays); here we test for
134 * more than 90% nucleotide; recheck every 10 columns in case
135 * of misleading data e.g. highly conserved Alanine in peptide!
136 * Mistakenly guessing nucleotide has a small performance cost,
137 * as it will result in counting in sparse arrays.
138 * Mistakenly guessing peptide has a small space cost,
139 * as it will use a larger than necessary array to hold counts.
141 if (nucleotideCount > 100 && column % 10 == 0)
143 nucleotide = (9 * peptideCount < nucleotideCount);
145 ResidueCount residueCounts = new ResidueCount(nucleotide);
147 for (int row = 0; row < seqCount; row++)
149 if (sequences[row] == null)
152 .println("WARNING: Consensus skipping null sequence - possible race condition.");
155 char[] seq = sequences[row].getSequence();
156 if (seq.length > column)
158 char c = seq[column];
159 residueCounts.add(c);
160 if (Comparison.isNucleotide(c))
164 else if (!Comparison.isGap(c))
172 * here we count a gap if the sequence doesn't
173 * reach this column (is that correct?)
175 residueCounts.addGap();
179 int maxCount = residueCounts.getModalCount();
180 String maxResidue = residueCounts.getResiduesForCount(maxCount);
181 int gapCount = residueCounts.getGapCount();
182 Profile profile = new Profile(seqCount, gapCount, maxCount,
187 profile.setCounts(residueCounts);
190 result[column] = profile;
192 // long elapsed = System.currentTimeMillis() - now;
193 // System.out.println(elapsed);
197 * Make an estimate of the profile size we are going to compute i.e. how many
198 * different characters may be present in it. Overestimating has a cost of
199 * using more memory than necessary. Underestimating has a cost of needing to
200 * extend the SparseIntArray holding the profile counts.
202 * @param profileSizes
203 * counts of sizes of profiles so far encountered
206 static int estimateProfileSize(SparseIntArray profileSizes)
208 if (profileSizes.size() == 0)
214 * could do a statistical heuristic here e.g. 75%ile
215 * for now just return the largest value
217 return profileSizes.keyAt(profileSizes.size() - 1);
221 * Derive the consensus annotations to be added to the alignment for display.
222 * This does not recompute the raw data, but may be called on a change in
223 * display options, such as 'show logo', which may in turn result in a change
224 * in the derived values.
227 * the annotation row to add annotations to
229 * the source consensus data
235 * if true, normalise residue percentages ignoring gaps
236 * @param showSequenceLogo
237 * if true include all consensus symbols, else just show modal
240 * number of sequences
242 public static void completeConsensus(AlignmentAnnotation consensus,
243 Profile[] profiles, int iStart, int width, boolean ignoreGaps,
244 boolean showSequenceLogo, long nseq)
246 // long now = System.currentTimeMillis();
247 if (consensus == null || consensus.annotations == null
248 || consensus.annotations.length < width)
251 * called with a bad alignment annotation row
252 * wait for it to be initialised properly
257 final int dp = getPercentageDp(nseq);
259 for (int i = iStart; i < width; i++)
262 if (i >= profiles.length || ((profile = profiles[i]) == null))
265 * happens if sequences calculated over were
266 * shorter than alignment width
268 consensus.annotations[i] = null;
272 float value = profile.getPercentageIdentity(ignoreGaps);
274 String description = getTooltip(profile, value, showSequenceLogo,
277 consensus.annotations[i] = new Annotation(profile.getModalResidue(),
278 description, ' ', value);
280 // long elapsed = System.currentTimeMillis() - now;
281 // System.out.println(-elapsed);
285 * Returns a tooltip showing either
287 * <li>the full profile (percentages of all residues present), if
288 * showSequenceLogo is true, or</li>
289 * <li>just the modal (most common) residue(s), if showSequenceLogo is false</li>
291 * Percentages are as a fraction of all sequence, or only ungapped sequences
292 * if ignoreGaps is true.
296 * @param showSequenceLogo
299 * the number of decimal places to format percentages to
302 static String getTooltip(Profile profile, float pid,
303 boolean showSequenceLogo, boolean ignoreGaps, int dp)
305 ResidueCount counts = profile.getCounts();
307 String description = null;
308 if (counts != null && showSequenceLogo)
310 int normaliseBy = ignoreGaps ? profile.getNonGapped() : profile
312 description = counts.getTooltip(normaliseBy, dp);
316 StringBuilder sb = new StringBuilder(64);
317 String maxRes = profile.getModalResidue();
318 if (maxRes.length() > 1)
320 sb.append("[").append(maxRes).append("] ");
325 sb.append(maxRes).append(" ");
327 Format.appendPercentage(sb, pid, dp);
329 description = sb.toString();
335 * Returns the sorted profile for the given consensus data. The returned array
339 * [profileType, numberOfValues, nonGapCount, charValue1, percentage1, charValue2, percentage2, ...]
340 * in descending order of percentage value
344 * the data object from which to extract and sort values
346 * if true, only non-gapped values are included in percentage
350 public static int[] extractProfile(Profile profile,
353 int[] rtnval = new int[64];
354 ResidueCount counts = profile.getCounts();
360 SymbolCounts symbolCounts = counts.getSymbolCounts();
361 char[] symbols = symbolCounts.symbols;
362 int[] values = symbolCounts.values;
363 QuickSort.sort(values, symbols);
364 int nextArrayPos = 2;
365 int totalPercentage = 0;
366 final int divisor = ignoreGaps ? profile.getNonGapped() : profile
370 * traverse the arrays in reverse order (highest counts first)
372 for (int i = symbols.length - 1; i >= 0; i--)
374 int theChar = symbols[i];
375 int charCount = values[i];
377 rtnval[nextArrayPos++] = theChar;
378 final int percentage = (charCount * 100) / divisor;
379 rtnval[nextArrayPos++] = percentage;
380 totalPercentage += percentage;
382 rtnval[0] = symbols.length;
383 rtnval[1] = totalPercentage;
384 int[] result = new int[rtnval.length + 1];
385 result[0] = AlignmentAnnotation.SEQUENCE_PROFILE;
386 System.arraycopy(rtnval, 0, result, 1, rtnval.length);
392 * Extract a sorted extract of cDNA codon profile data. The returned array
396 * [profileType, numberOfValues, totalCount, charValue1, percentage1, charValue2, percentage2, ...]
397 * in descending order of percentage value, where the character values encode codon triplets
403 public static int[] extractCdnaProfile(Hashtable hashtable,
406 // this holds #seqs, #ungapped, and then codon count, indexed by encoded
408 int[] codonCounts = (int[]) hashtable.get(PROFILE);
409 int[] sortedCounts = new int[codonCounts.length - 2];
410 System.arraycopy(codonCounts, 2, sortedCounts, 0,
411 codonCounts.length - 2);
413 int[] result = new int[3 + 2 * sortedCounts.length];
414 // first value is just the type of profile data
415 result[0] = AlignmentAnnotation.CDNA_PROFILE;
417 char[] codons = new char[sortedCounts.length];
418 for (int i = 0; i < codons.length; i++)
420 codons[i] = (char) i;
422 QuickSort.sort(sortedCounts, codons);
423 int totalPercentage = 0;
424 int distinctValuesCount = 0;
426 int divisor = ignoreGaps ? codonCounts[1] : codonCounts[0];
427 for (int i = codons.length - 1; i >= 0; i--)
429 final int codonCount = sortedCounts[i];
432 break; // nothing else of interest here
434 distinctValuesCount++;
435 result[j++] = codons[i];
436 final int percentage = codonCount * 100 / divisor;
437 result[j++] = percentage;
438 totalPercentage += percentage;
440 result[2] = totalPercentage;
443 * Just return the non-zero values
445 // todo next value is redundant if we limit the array to non-zero counts
446 result[1] = distinctValuesCount;
447 return Arrays.copyOfRange(result, 0, j);
451 * Compute a consensus for the cDNA coding for a protein alignment.
454 * the protein alignment (which should hold mappings to cDNA
457 * the consensus data stores to be populated (one per column)
459 public static void calculateCdna(AlignmentI alignment,
460 Hashtable[] hconsensus)
462 final char gapCharacter = alignment.getGapCharacter();
463 List<AlignedCodonFrame> mappings = alignment.getCodonFrames();
464 if (mappings == null || mappings.isEmpty())
469 int cols = alignment.getWidth();
470 for (int col = 0; col < cols; col++)
472 // todo would prefer a Java bean for consensus data
473 Hashtable<String, int[]> columnHash = new Hashtable<String, int[]>();
474 // #seqs, #ungapped seqs, counts indexed by (codon encoded + 1)
475 int[] codonCounts = new int[66];
476 codonCounts[0] = alignment.getSequences().size();
477 int ungappedCount = 0;
478 for (SequenceI seq : alignment.getSequences())
480 if (seq.getCharAt(col) == gapCharacter)
484 List<char[]> codons = MappingUtils
485 .findCodonsFor(seq, col, mappings);
486 for (char[] codon : codons)
488 int codonEncoded = CodingUtils.encodeCodon(codon);
489 if (codonEncoded >= 0)
491 codonCounts[codonEncoded + 2]++;
496 codonCounts[1] = ungappedCount;
497 // todo: sort values here, save counts and codons?
498 columnHash.put(PROFILE, codonCounts);
499 hconsensus[col] = columnHash;
504 * Derive displayable cDNA consensus annotation from computed consensus data.
506 * @param consensusAnnotation
507 * the annotation row to be populated for display
508 * @param consensusData
509 * the computed consensus data
510 * @param showProfileLogo
511 * if true show all symbols present at each position, else only the
514 * the number of sequences in the alignment
516 public static void completeCdnaConsensus(
517 AlignmentAnnotation consensusAnnotation,
518 Hashtable[] consensusData, boolean showProfileLogo, int nseqs)
520 if (consensusAnnotation == null
521 || consensusAnnotation.annotations == null
522 || consensusAnnotation.annotations.length < consensusData.length)
524 // called with a bad alignment annotation row - wait for it to be
525 // initialised properly
529 // ensure codon triplet scales with font size
530 consensusAnnotation.scaleColLabel = true;
531 for (int col = 0; col < consensusData.length; col++)
533 Hashtable hci = consensusData[col];
536 // gapped protein column?
539 // array holds #seqs, #ungapped, then codon counts indexed by codon
540 final int[] codonCounts = (int[]) hci.get(PROFILE);
544 * First pass - get total count and find the highest
546 final char[] codons = new char[codonCounts.length - 2];
547 for (int j = 2; j < codonCounts.length; j++)
549 final int codonCount = codonCounts[j];
550 codons[j - 2] = (char) (j - 2);
551 totalCount += codonCount;
555 * Sort array of encoded codons by count ascending - so the modal value
556 * goes to the end; start by copying the count (dropping the first value)
558 int[] sortedCodonCounts = new int[codonCounts.length - 2];
559 System.arraycopy(codonCounts, 2, sortedCodonCounts, 0,
560 codonCounts.length - 2);
561 QuickSort.sort(sortedCodonCounts, codons);
563 int modalCodonEncoded = codons[codons.length - 1];
564 int modalCodonCount = sortedCodonCounts[codons.length - 1];
565 String modalCodon = String.valueOf(CodingUtils
566 .decodeCodon(modalCodonEncoded));
567 if (sortedCodonCounts.length > 1
568 && sortedCodonCounts[codons.length - 2] == sortedCodonCounts[codons.length - 1])
571 * two or more codons share the modal count
575 float pid = sortedCodonCounts[sortedCodonCounts.length - 1] * 100
576 / (float) totalCount;
579 * todo ? Replace consensus hashtable with sorted arrays of codons and
580 * counts (non-zero only). Include total count in count array [0].
584 * Scan sorted array backwards for most frequent values first. Show
585 * repeated values compactly.
587 StringBuilder mouseOver = new StringBuilder(32);
588 StringBuilder samePercent = new StringBuilder();
589 String percent = null;
590 String lastPercent = null;
591 int percentDecPl = getPercentageDp(nseqs);
593 for (int j = codons.length - 1; j >= 0; j--)
595 int codonCount = sortedCodonCounts[j];
599 * remaining codons are 0% - ignore, but finish off the last one if
602 if (samePercent.length() > 0)
604 mouseOver.append(samePercent).append(": ").append(percent)
609 int codonEncoded = codons[j];
610 final int pct = codonCount * 100 / totalCount;
611 String codon = String
612 .valueOf(CodingUtils.decodeCodon(codonEncoded));
613 StringBuilder sb = new StringBuilder();
614 Format.appendPercentage(sb, pct, percentDecPl);
615 percent = sb.toString();
616 if (showProfileLogo || codonCount == modalCodonCount)
618 if (percent.equals(lastPercent) && j > 0)
620 samePercent.append(samePercent.length() == 0 ? "" : ", ");
621 samePercent.append(codon);
625 if (samePercent.length() > 0)
627 mouseOver.append(samePercent).append(": ")
628 .append(lastPercent).append("% ");
630 samePercent.setLength(0);
631 samePercent.append(codon);
633 lastPercent = percent;
637 consensusAnnotation.annotations[col] = new Annotation(modalCodon,
638 mouseOver.toString(), ' ', pid);
643 * Returns the number of decimal places to show for profile percentages. For
644 * less than 100 sequences, returns zero (the integer percentage value will be
645 * displayed). For 100-999 sequences, returns 1, for 1000-9999 returns 2, etc.
650 protected static int getPercentageDp(long nseq)