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 static org.testng.AssertJUnit.assertEquals;
24 import static org.testng.AssertJUnit.assertNull;
26 import jalview.datamodel.AlignmentAnnotation;
27 import jalview.datamodel.Annotation;
28 import jalview.datamodel.Profile;
29 import jalview.datamodel.ProfileI;
30 import jalview.datamodel.ProfilesI;
31 import jalview.datamodel.ResidueCount;
32 import jalview.datamodel.Sequence;
33 import jalview.datamodel.SequenceI;
34 import jalview.gui.JvOptionPane;
36 import java.util.Hashtable;
38 import org.testng.annotations.BeforeClass;
39 import org.testng.annotations.Test;
41 public class AAFrequencyTest
44 @BeforeClass(alwaysRun = true)
45 public void setUpJvOptionPane()
47 JvOptionPane.setInteractiveMode(false);
48 JvOptionPane.setMockResponse(JvOptionPane.CANCEL_OPTION);
51 @Test(groups = { "Functional" })
52 public void testCalculate_noProfile()
54 SequenceI seq1 = new Sequence("Seq1", "CAG-T");
55 SequenceI seq2 = new Sequence("Seq2", "CAC-T");
56 SequenceI seq3 = new Sequence("Seq3", "C---G");
57 SequenceI seq4 = new Sequence("Seq4", "CA--t");
58 SequenceI[] seqs = new SequenceI[] { seq1, seq2, seq3, seq4 };
59 int width = seq1.getLength();
60 ProfilesI result = AAFrequency.calculate(seqs, width, 0, width,
64 ProfileI col = result.get(0);
65 assertEquals(100f, col.getPercentageIdentity(false));
66 assertEquals(100f, col.getPercentageIdentity(true));
67 assertEquals(4, col.getMaxCount());
68 assertEquals("C", col.getModalResidue());
69 assertNull(col.getCounts());
73 assertEquals(75f, col.getPercentageIdentity(false));
74 assertEquals(100f, col.getPercentageIdentity(true));
75 assertEquals(3, col.getMaxCount());
76 assertEquals("A", col.getModalResidue());
78 // col 2 is 50% G 50% C or 25/25 counting gaps
80 assertEquals(25f, col.getPercentageIdentity(false));
81 assertEquals(50f, col.getPercentageIdentity(true));
82 assertEquals(1, col.getMaxCount());
83 assertEquals("CG", col.getModalResidue());
87 assertEquals(0f, col.getPercentageIdentity(false));
88 assertEquals(0f, col.getPercentageIdentity(true));
89 assertEquals(0, col.getMaxCount());
90 assertEquals("", col.getModalResidue());
92 // col 4 is 75% T 25% G
94 assertEquals(75f, col.getPercentageIdentity(false));
95 assertEquals(75f, col.getPercentageIdentity(true));
96 assertEquals(3, col.getMaxCount());
97 assertEquals("T", col.getModalResidue());
100 @Test(groups = { "Functional" })
101 public void testCalculate_withProfile()
103 SequenceI seq1 = new Sequence("Seq1", "CAGT");
104 SequenceI seq2 = new Sequence("Seq2", "CACT");
105 SequenceI seq3 = new Sequence("Seq3", "C--G");
106 SequenceI seq4 = new Sequence("Seq4", "CA-t");
107 SequenceI[] seqs = new SequenceI[] { seq1, seq2, seq3, seq4 };
108 int width = seq1.getLength();
109 ProfilesI result = AAFrequency.calculate(seqs, width, 0, width,
112 ProfileI profile = result.get(0);
113 assertEquals(4, profile.getCounts().getCount('C'));
114 assertEquals(4, profile.getHeight());
115 assertEquals(4, profile.getNonGapped());
117 profile = result.get(1);
118 assertEquals(3, profile.getCounts().getCount('A'));
119 assertEquals(4, profile.getHeight());
120 assertEquals(3, profile.getNonGapped());
122 profile = result.get(2);
123 assertEquals(1, profile.getCounts().getCount('C'));
124 assertEquals(1, profile.getCounts().getCount('G'));
125 assertEquals(4, profile.getHeight());
126 assertEquals(2, profile.getNonGapped());
128 profile = result.get(3);
129 assertEquals(3, profile.getCounts().getCount('T'));
130 assertEquals(1, profile.getCounts().getCount('G'));
131 assertEquals(4, profile.getHeight());
132 assertEquals(4, profile.getNonGapped());
135 @Test(groups = { "Functional" }, enabled = false)
136 public void testCalculate_withProfileTiming()
138 SequenceI seq1 = new Sequence("Seq1", "CAGT");
139 SequenceI seq2 = new Sequence("Seq2", "CACT");
140 SequenceI seq3 = new Sequence("Seq3", "C--G");
141 SequenceI seq4 = new Sequence("Seq4", "CA-t");
142 SequenceI[] seqs = new SequenceI[] { seq1, seq2, seq3, seq4 };
144 // ensure class loaded and initialised
145 int width = seq1.getLength();
146 AAFrequency.calculate(seqs, width, 0, width, true);
149 long start = System.currentTimeMillis();
150 for (int i = 0; i < reps; i++)
152 AAFrequency.calculate(seqs, width, 0, width, true);
154 System.out.println(System.currentTimeMillis() - start);
158 * Test generation of consensus annotation with options 'include gaps'
159 * (profile percentages are of all sequences, whether gapped or not), and
160 * 'show logo' (the full profile with all residue percentages is reported in
161 * the description for the tooltip)
163 @Test(groups = { "Functional" })
164 public void testCompleteConsensus_includeGaps_showLogo()
167 * first compute the profiles
169 SequenceI seq1 = new Sequence("Seq1", "CAG-T");
170 SequenceI seq2 = new Sequence("Seq2", "CAC-T");
171 SequenceI seq3 = new Sequence("Seq3", "C---G");
172 SequenceI seq4 = new Sequence("Seq4", "CA--t");
173 SequenceI[] seqs = new SequenceI[] { seq1, seq2, seq3, seq4 };
174 int width = seq1.getLength();
175 ProfilesI profiles = AAFrequency.calculate(seqs, width, 0, width, true);
177 AlignmentAnnotation consensus = new AlignmentAnnotation("Consensus",
178 "PID", new Annotation[width]);
180 .completeConsensus(consensus, profiles, 0, 5, false, true, 4);
182 Annotation ann = consensus.annotations[0];
183 assertEquals("C 100%", ann.description);
184 assertEquals("C", ann.displayCharacter);
185 ann = consensus.annotations[1];
186 assertEquals("A 75%", ann.description);
187 assertEquals("A", ann.displayCharacter);
188 ann = consensus.annotations[2];
189 assertEquals("C 25%; G 25%", ann.description);
190 assertEquals("+", ann.displayCharacter);
191 ann = consensus.annotations[3];
192 assertEquals("", ann.description);
193 assertEquals("-", ann.displayCharacter);
194 ann = consensus.annotations[4];
195 assertEquals("T 75%; G 25%", ann.description);
196 assertEquals("T", ann.displayCharacter);
200 * Test generation of consensus annotation with options 'ignore gaps' (profile
201 * percentages are of the non-gapped sequences) and 'no logo' (only the modal
202 * residue[s] percentage is reported in the description for the tooltip)
204 @Test(groups = { "Functional" })
205 public void testCompleteConsensus_ignoreGaps_noLogo()
208 * first compute the profiles
210 SequenceI seq1 = new Sequence("Seq1", "CAG-T");
211 SequenceI seq2 = new Sequence("Seq2", "CAC-T");
212 SequenceI seq3 = new Sequence("Seq3", "C---G");
213 SequenceI seq4 = new Sequence("Seq4", "CA--t");
214 SequenceI[] seqs = new SequenceI[] { seq1, seq2, seq3, seq4 };
215 int width = seq1.getLength();
216 ProfilesI profiles = AAFrequency.calculate(seqs, width, 0, width, true);
218 AlignmentAnnotation consensus = new AlignmentAnnotation("Consensus",
219 "PID", new Annotation[width]);
221 .completeConsensus(consensus, profiles, 0, 5, true, false, 4);
223 Annotation ann = consensus.annotations[0];
224 assertEquals("C 100%", ann.description);
225 assertEquals("C", ann.displayCharacter);
226 ann = consensus.annotations[1];
227 assertEquals("A 100%", ann.description);
228 assertEquals("A", ann.displayCharacter);
229 ann = consensus.annotations[2];
230 assertEquals("[CG] 50%", ann.description);
231 assertEquals("+", ann.displayCharacter);
232 ann = consensus.annotations[3];
233 assertEquals("", ann.description);
234 assertEquals("-", ann.displayCharacter);
235 ann = consensus.annotations[4];
236 assertEquals("T 75%", ann.description);
237 assertEquals("T", ann.displayCharacter);
241 * Test to include rounding down of a non-zero count to 0% (JAL-3202)
243 @Test(groups = { "Functional" })
244 public void testExtractProfile()
247 * 200 sequences of which 30 gapped (170 ungapped)
248 * max count 70 for modal residue 'G'
250 ProfileI profile = new Profile(200, 30, 70, "G");
251 ResidueCount counts = new ResidueCount();
256 profile.setCounts(counts);
259 * [0, noOfValues, totalPercent, char1, count1, ...]
260 * G: 70/170 = 41.2 = 41
261 * R: 60/170 = 35.3 = 35
262 * L: 38/170 = 22.3 = 22
264 * total (rounded) percentages = 99
266 int[] extracted = AAFrequency.extractProfile(profile, true);
267 int[] expected = new int[] { 0, 4, 99, 'G', 41, 'R', 35, 'L', 22, 'H',
269 org.testng.Assert.assertEquals(extracted, expected);
272 * add some counts of 1; these round down to 0% and should be discarded
274 counts.put('G', 68); // 68/170 = 40% exactly (percentages now total 98)
277 extracted = AAFrequency.extractProfile(profile, true);
278 expected = new int[] { 0, 4, 98, 'G', 40, 'R', 35, 'L', 22, 'H', 1 };
279 org.testng.Assert.assertEquals(extracted, expected);
284 * Tests for the profile calculation where gaps are included i.e. the
285 * denominator is the total number of sequences in the column
287 @Test(groups = { "Functional" })
288 public void testExtractProfile_countGaps()
291 * 200 sequences of which 30 gapped (170 ungapped)
292 * max count 70 for modal residue 'G'
294 ProfileI profile = new Profile(200, 30, 70, "G");
295 ResidueCount counts = new ResidueCount();
300 profile.setCounts(counts);
303 * [0, noOfValues, totalPercent, char1, count1, ...]
308 * total (rounded) percentages = 85
310 int[] extracted = AAFrequency.extractProfile(profile, false);
311 int[] expected = new int[] { AlignmentAnnotation.SEQUENCE_PROFILE, 4,
312 85, 'G', 35, 'R', 30, 'L', 19, 'H',
314 org.testng.Assert.assertEquals(extracted, expected);
317 * add some counts of 1; these round down to 0% and should be discarded
319 counts.put('G', 68); // 68/200 = 34%
322 extracted = AAFrequency.extractProfile(profile, false);
323 expected = new int[] { AlignmentAnnotation.SEQUENCE_PROFILE, 4, 84, 'G',
324 34, 'R', 30, 'L', 19, 'H', 1 };
325 org.testng.Assert.assertEquals(extracted, expected);
329 @Test(groups = { "Functional" })
330 public void testExtractCdnaProfile()
333 * 200 sequences of which 30 gapped (170 ungapped)
334 * max count 70 for modal residue 'G'
336 Hashtable profile = new Hashtable();
339 * cdna profile is {seqCount, ungappedCount, codonCount1, ...codonCount64}
340 * where 1..64 positions correspond to encoded codons
341 * see CodingUtils.encodeCodon()
343 int[] codonCounts = new int[66];
344 char[] codon1 = new char[] { 'G', 'C', 'A' };
345 char[] codon2 = new char[] { 'c', 'C', 'A' };
346 char[] codon3 = new char[] { 't', 'g', 'A' };
347 char[] codon4 = new char[] { 'G', 'C', 't' };
348 int encoded1 = CodingUtils.encodeCodon(codon1);
349 int encoded2 = CodingUtils.encodeCodon(codon2);
350 int encoded3 = CodingUtils.encodeCodon(codon3);
351 int encoded4 = CodingUtils.encodeCodon(codon4);
352 codonCounts[2 + encoded1] = 30;
353 codonCounts[2 + encoded2] = 70;
354 codonCounts[2 + encoded3] = 9;
355 codonCounts[2 + encoded4] = 1;
356 codonCounts[0] = 120;
357 codonCounts[1] = 110;
358 profile.put(AAFrequency.PROFILE, codonCounts);
361 * [0, noOfValues, totalPercent, char1, count1, ...]
362 * codon1: 30/110 = 27.2 = 27%
363 * codon2: 70/110 = 63.6% = 63%
364 * codon3: 9/110 = 8.1% = 8%
365 * codon4: 1/110 = 0.9% = 0% should be discarded
366 * total (rounded) percentages = 98
368 int[] extracted = AAFrequency.extractCdnaProfile(profile, true);
369 int[] expected = new int[] { AlignmentAnnotation.CDNA_PROFILE, 3, 98,
370 encoded2, 63, encoded1, 27, encoded3, 8 };
371 org.testng.Assert.assertEquals(extracted, expected);
374 @Test(groups = { "Functional" })
375 public void testExtractCdnaProfile_countGaps()
378 * 200 sequences of which 30 gapped (170 ungapped)
379 * max count 70 for modal residue 'G'
381 Hashtable profile = new Hashtable();
384 * cdna profile is {seqCount, ungappedCount, codonCount1, ...codonCount64}
385 * where 1..64 positions correspond to encoded codons
386 * see CodingUtils.encodeCodon()
388 int[] codonCounts = new int[66];
389 char[] codon1 = new char[] { 'G', 'C', 'A' };
390 char[] codon2 = new char[] { 'c', 'C', 'A' };
391 char[] codon3 = new char[] { 't', 'g', 'A' };
392 char[] codon4 = new char[] { 'G', 'C', 't' };
393 int encoded1 = CodingUtils.encodeCodon(codon1);
394 int encoded2 = CodingUtils.encodeCodon(codon2);
395 int encoded3 = CodingUtils.encodeCodon(codon3);
396 int encoded4 = CodingUtils.encodeCodon(codon4);
397 codonCounts[2 + encoded1] = 30;
398 codonCounts[2 + encoded2] = 70;
399 codonCounts[2 + encoded3] = 9;
400 codonCounts[2 + encoded4] = 1;
401 codonCounts[0] = 120;
402 codonCounts[1] = 110;
403 profile.put(AAFrequency.PROFILE, codonCounts);
406 * [0, noOfValues, totalPercent, char1, count1, ...]
407 * codon1: 30/120 = 25%
408 * codon2: 70/120 = 58.3 = 58%
409 * codon3: 9/120 = 7.5 = 7%
410 * codon4: 1/120 = 0.8 = 0% should be discarded
411 * total (rounded) percentages = 90
413 int[] extracted = AAFrequency.extractCdnaProfile(profile, false);
414 int[] expected = new int[] { AlignmentAnnotation.CDNA_PROFILE, 3, 90,
415 encoded2, 58, encoded1, 25, encoded3, 7 };
416 org.testng.Assert.assertEquals(extracted, expected);