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.
23 import java.util.ArrayList;
24 import java.util.List;
26 import jalview.datamodel.SequenceI;
29 * Assorted methods for analysing or comparing sequences.
31 public class Comparison
33 private static final int EIGHTY_FIVE = 85;
35 private static final int NINETY_NINE = 99;
37 private static final int TO_UPPER_CASE = 'a' - 'A';
39 public static final char GAP_SPACE = ' ';
41 public static final char GAP_DOT = '.';
43 public static final char GAP_DASH = '-';
45 public static final String GapChars = new String(
47 { GAP_SPACE, GAP_DOT, GAP_DASH });
57 * @return DOCUMENT ME!
59 public static final float compare(SequenceI ii, SequenceI jj)
61 return Comparison.compare(ii, jj, 0, ii.getLength() - 1);
65 * this was supposed to be an ungapped pid calculation
77 public static float compare(SequenceI ii, SequenceI jj, int start,
80 String si = ii.getSequenceAsString();
81 String sj = jj.getSequenceAsString();
83 int ilen = si.length() - 1;
84 int jlen = sj.length() - 1;
86 while (Comparison.isGap(si.charAt(start + ilen)))
91 while (Comparison.isGap(sj.charAt(start + jlen)))
102 for (int j = 0; j < jlen; j++)
104 if (si.substring(start + j, start + j + 1)
105 .equals(sj.substring(start + j, start + j + 1)))
113 pid = (float) match / (float) ilen * 100;
117 for (int j = 0; j < jlen; j++)
119 if (si.substring(start + j, start + j + 1)
120 .equals(sj.substring(start + j, start + j + 1)))
128 pid = (float) match / (float) jlen * 100;
135 * this is a gapped PID calculation
142 * @deprecated use PIDModel.computePID()
145 public final static float PID(String seq1, String seq2)
147 return PID(seq1, seq2, 0, seq1.length());
150 static final int caseShift = 'a' - 'A';
152 // Another pid with region specification
154 * @deprecated use PIDModel.computePID()
157 public final static float PID(String seq1, String seq2, int start,
160 return PID(seq1, seq2, start, end, true, false);
164 * Calculate percent identity for a pair of sequences over a particular range,
165 * with different options for ignoring gaps.
174 * - if true - gaps match any character, if false, do not match
176 * @param ungappedOnly
177 * - if true - only count PID over ungapped columns
179 * @deprecated use PIDModel.computePID()
182 public final static float PID(String seq1, String seq2, int start,
183 int end, boolean wcGaps, boolean ungappedOnly)
185 int s1len = seq1.length();
186 int s2len = seq2.length();
188 int len = Math.min(s1len, s2len);
197 start = len - 1; // we just use a single residue for the difference
200 int elen = len - start, bad = 0;
204 for (int i = start; i < len; i++)
206 chr1 = seq1.charAt(i);
208 chr2 = seq2.charAt(i);
209 agap = isGap(chr1) || isGap(chr2);
210 if ('a' <= chr1 && chr1 <= 'z')
213 // Faster than toUpperCase
216 if ('a' <= chr2 && chr2 <= 'z')
219 // Faster than toUpperCase
247 return ((float) 100 * (elen - bad)) / elen;
251 * Answers true if the supplied character is a recognised gap character, else
252 * false. Currently hard-coded to recognise '-', '-' or ' ' (hyphen / dot /
259 public static final boolean isGap(char c)
261 return c == GAP_DASH || c == GAP_DOT || c == GAP_SPACE;
265 * Overloaded method signature to test whether a single sequence is nucleotide
266 * (that is, more than 85% CGTAUNX)
271 public static final boolean isNucleotide(SequenceI seq)
282 int len = seq.getLength();
283 for (int i = 0; i < len; i++)
285 char c = seq.getCharAt(i);
299 if (isNucleotideAmbiguity(c))
307 * Check for nucleotide count > 85% of total count (in a form that evades
308 * int / float conversion or divide by zero).
310 if ((ntCount + nCount) * 100 > EIGHTY_FIVE * (ntCount + aaCount))
312 return ntCount > 0; // all N is considered protein. Could use a threshold
317 // check for very large proportion of nucleotide and all ambiguity codes
318 if ((ntCount + nCount + ntaCount) * 100 >= NINETY_NINE
319 * (ntCount + aaCount))
328 * Answers true if more than 85% of the sequence residues (ignoring gaps) are
329 * A, G, C, T or U, else false. This is just a heuristic guess and may give a
330 * wrong answer (as AGCT are also amino acid codes).
335 public static final boolean isNucleotide(SequenceI[] seqs)
341 // true if we have seen a nucleotide sequence
343 for (SequenceI seq : seqs)
350 // TODO could possibly make an informed guess just from the first sequence
351 // to save a lengthy calculation
354 // if even one looks like protein, the alignment is protein
362 * Answers true if the character is one of aAcCgGtTuU
367 public static boolean isNucleotide(char c)
369 return isNucleotide(c, false);
372 public static boolean isNucleotide(char c, boolean countAmbiguity)
374 char C = Character.toUpperCase(c);
386 boolean ambiguity = isNucleotideAmbiguity(C);
393 public static boolean isNucleotideAmbiguity(char c)
395 switch (Character.toUpperCase(c))
410 case 'N': // not counting N as nucleotide
415 public static boolean isN(char c)
417 return 'n' == Character.toLowerCase(c);
420 public static boolean isX(char c)
422 return 'x' == Character.toLowerCase(c);
426 * Answers true if every character in the string is one of aAcCgGtTuU, or
427 * (optionally) a gap character (dot, dash, space), else false
433 public static boolean isNucleotideSequence(String s, boolean allowGaps)
439 for (int i = 0; i < s.length(); i++)
441 char c = s.charAt(i);
442 if (!isNucleotide(c))
444 if (!allowGaps || !isGap(c))
454 * Convenience overload of isNucleotide
459 public static boolean isNucleotide(SequenceI[][] seqs)
465 List<SequenceI> flattened = new ArrayList<SequenceI>();
466 for (SequenceI[] ss : seqs)
468 for (SequenceI s : ss)
473 final SequenceI[] oneDArray = flattened
474 .toArray(new SequenceI[flattened.size()]);
475 return isNucleotide(oneDArray);
479 * Compares two residues either case sensitively or case insensitively
480 * depending on the caseSensitive flag
485 * second char to compare with
486 * @param caseSensitive
487 * if true comparison will be case sensitive otherwise its not
490 public static boolean isSameResidue(char c1, char c2,
491 boolean caseSensitive)
493 return caseSensitive ? c1 == c2
494 : Character.toUpperCase(c1) == Character.toUpperCase(c2);