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.assertFalse;
25 import static org.testng.AssertJUnit.assertNotNull;
26 import static org.testng.AssertJUnit.assertNotSame;
27 import static org.testng.AssertJUnit.assertNull;
28 import static org.testng.AssertJUnit.assertSame;
29 import static org.testng.AssertJUnit.assertTrue;
31 import jalview.datamodel.AlignedCodonFrame;
32 import jalview.datamodel.AlignedCodonFrame.SequenceToSequenceMapping;
33 import jalview.datamodel.Alignment;
34 import jalview.datamodel.AlignmentI;
35 import jalview.datamodel.DBRefEntry;
36 import jalview.datamodel.Mapping;
37 import jalview.datamodel.Sequence;
38 import jalview.datamodel.SequenceFeature;
39 import jalview.datamodel.SequenceI;
40 import jalview.gui.JvOptionPane;
41 import jalview.util.DBRefUtils;
42 import jalview.util.MapList;
43 import jalview.ws.SequenceFetcher;
44 import jalview.ws.SequenceFetcherFactory;
46 import java.util.ArrayList;
47 import java.util.Arrays;
48 import java.util.List;
50 import org.testng.annotations.AfterClass;
51 import org.testng.annotations.BeforeClass;
52 import org.testng.annotations.Test;
54 public class CrossRefTest
57 @BeforeClass(alwaysRun = true)
58 public void setUpJvOptionPane()
60 JvOptionPane.setInteractiveMode(false);
61 JvOptionPane.setMockResponse(JvOptionPane.CANCEL_OPTION);
64 @Test(groups = { "Functional" })
65 public void testFindXDbRefs()
67 DBRefEntry ref1 = new DBRefEntry("UNIPROT", "1", "A123");
68 DBRefEntry ref2 = new DBRefEntry("UNIPROTKB/TREMBL", "1", "A123");
69 DBRefEntry ref3 = new DBRefEntry("pdb", "1", "A123");
70 DBRefEntry ref4 = new DBRefEntry("EMBLCDSPROTEIN", "1", "A123");
71 DBRefEntry ref5 = new DBRefEntry("embl", "1", "A123");
72 DBRefEntry ref6 = new DBRefEntry("emblCDS", "1", "A123");
73 DBRefEntry ref7 = new DBRefEntry("GeneDB", "1", "A123");
74 DBRefEntry ref8 = new DBRefEntry("PFAM", "1", "A123");
75 // ENSEMBL is a source of either dna or protein sequence data
76 DBRefEntry ref9 = new DBRefEntry("ENSEMBL", "1", "A123");
77 List<DBRefEntry> refs = Arrays.asList(new DBRefEntry[] { ref1, ref2, ref3, ref4, ref5,
78 ref6, ref7, ref8, ref9 });
83 List<DBRefEntry> found = DBRefUtils.selectDbRefs(true, refs);
84 assertEquals(4, found.size());
85 assertSame(ref5, found.get(0));
86 assertSame(ref6, found.get(1));
87 assertSame(ref7, found.get(2));
88 assertSame(ref9, found.get(3));
91 * Just the protein refs:
93 found = DBRefUtils.selectDbRefs(false, refs);
94 assertEquals(4, found.size());
95 assertSame(ref1, found.get(0));
96 assertSame(ref2, found.get(1));
97 assertSame(ref4, found.get(2));
98 assertSame(ref9, found.get(3));
102 * Test the method that finds a sequence's "product" xref source databases,
103 * which may be direct (dbrefs on the sequence), or indirect (dbrefs on
104 * sequences which share a dbref with the sequence
106 @Test(groups = { "Functional" }, enabled = true)
107 public void testFindXrefSourcesForSequence_proteinToDna()
109 SequenceI seq = new Sequence("Seq1", "MGKYQARLSS");
110 List<String> sources = new ArrayList<>();
111 AlignmentI al = new Alignment(new SequenceI[] {});
114 * first with no dbrefs to search
116 sources = new CrossRef(new SequenceI[] { seq }, al)
117 .findXrefSourcesForSequences(false);
118 assertTrue(sources.isEmpty());
121 * add some dbrefs to sequence
123 // protein db is not a candidate for findXrefSources
124 seq.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
125 // dna coding databatases are
126 seq.addDBRef(new DBRefEntry("EMBL", "0", "E2345"));
127 // a second EMBL xref should not result in a duplicate
128 seq.addDBRef(new DBRefEntry("EMBL", "0", "E2346"));
129 seq.addDBRef(new DBRefEntry("EMBLCDS", "0", "E2347"));
130 seq.addDBRef(new DBRefEntry("GENEDB", "0", "E2348"));
131 seq.addDBRef(new DBRefEntry("ENSEMBL", "0", "E2349"));
132 seq.addDBRef(new DBRefEntry("ENSEMBLGENOMES", "0", "E2350"));
133 sources = new CrossRef(new SequenceI[] { seq }, al)
134 .findXrefSourcesForSequences(false);
135 // method is patched to remove EMBL from the sources to match
136 assertEquals(4, sources.size());
137 assertEquals("[EMBLCDS, GENEDB, ENSEMBL, ENSEMBLGENOMES]",
141 * add a sequence to the alignment which has a dbref to UNIPROT|A1234
142 * and others to dna coding databases
146 seq.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
147 seq.addDBRef(new DBRefEntry("EMBLCDS", "0", "E2347"));
148 SequenceI seq2 = new Sequence("Seq2", "MGKYQARLSS");
149 seq2.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
150 seq2.addDBRef(new DBRefEntry("EMBL", "0", "E2345"));
151 seq2.addDBRef(new DBRefEntry("GENEDB", "0", "E2348"));
152 // TODO include ENSEMBLGENOMES in DBRefSource.DNACODINGDBS ?
153 al.addSequence(seq2);
154 sources = new CrossRef(new SequenceI[] { seq, seq2 }, al)
155 .findXrefSourcesForSequences(false);
156 // method removed EMBL from sources to match
157 assertEquals(2, sources.size());
158 assertEquals("[EMBLCDS, GENEDB]", sources.toString());
162 * Test for finding 'product' sequences for the case where only an indirect
163 * xref is found - not on the nucleotide sequence but on a peptide sequence in
164 * the alignment which which it shares a nucleotide dbref
166 @Test(groups = { "Functional" }, enabled = true)
167 public void testFindXrefSequences_indirectDbrefToProtein()
171 * - nucleotide dbref EMBL|AF039662
172 * - peptide dbrefs EMBL|AF039662, UNIPROT|Q9ZTS2
174 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
175 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
176 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
177 uniprotSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
178 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
181 * Find UNIPROT xrefs for nucleotide
182 * - it has no UNIPROT dbref of its own
183 * - but peptide with matching nucleotide dbref does, so is returned
185 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
186 Alignment xrefs = new CrossRef(new SequenceI[] { emblSeq }, al)
187 .findXrefSequences("UNIPROT", true);
188 assertEquals(1, xrefs.getHeight());
189 assertSame(uniprotSeq, xrefs.getSequenceAt(0));
193 * Test for finding 'product' sequences for the case where only an indirect
194 * xref is found - not on the peptide sequence but on a nucleotide sequence in
195 * the alignment which which it shares a protein dbref
197 @Test(groups = { "Functional" }, enabled = true)
198 public void testFindXrefSequences_indirectDbrefToNucleotide()
202 * - peptide dbref UNIPROT|Q9ZTS2
203 * - nucleotide dbref EMBL|AF039662, UNIPROT|Q9ZTS2
205 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
206 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
207 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
208 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
209 emblSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
212 * find EMBL xrefs for peptide sequence - it has no direct
213 * dbrefs, but the 'corresponding' nucleotide sequence does, so is returned
216 * Find EMBL xrefs for peptide
217 * - it has no EMBL dbref of its own
218 * - but nucleotide with matching peptide dbref does, so is returned
220 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
221 Alignment xrefs = new CrossRef(new SequenceI[] { uniprotSeq }, al)
222 .findXrefSequences("EMBL", false);
223 assertEquals(1, xrefs.getHeight());
224 assertSame(emblSeq, xrefs.getSequenceAt(0));
228 * Test for finding 'product' sequences for the case where the selected
229 * sequence has no dbref to the desired source, and there are no indirect
230 * references via another sequence in the alignment
232 @Test(groups = { "Functional" })
233 public void testFindXrefSequences_noDbrefs()
236 * two nucleotide sequences, one with UNIPROT dbref
238 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
239 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
240 SequenceI dna2 = new Sequence("AJ307031", "AAACCCTTT");
243 * find UNIPROT xrefs for peptide sequence - it has no direct
244 * dbrefs, and the other sequence (which has a UNIPROT dbref) is not
245 * equatable to it, so no results found
247 AlignmentI al = new Alignment(new SequenceI[] { dna1, dna2 });
248 Alignment xrefs = new CrossRef(new SequenceI[] { dna2 }, al)
249 .findXrefSequences("UNIPROT", true);
254 * Tests for the method that searches an alignment (with one sequence
255 * excluded) for protein/nucleotide sequences with a given cross-reference
257 @Test(groups = { "Functional" }, enabled = true)
258 public void testSearchDataset()
261 * nucleotide sequence with UNIPROT AND EMBL dbref
262 * peptide sequence with UNIPROT dbref
264 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
265 Mapping map = new Mapping(new Sequence("pep2", "MLAVSRG"), new MapList(
266 new int[] { 1, 21 }, new int[] { 1, 7 }, 3, 1));
267 DBRefEntry dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
268 dna1.addDBRef(dbref);
269 dna1.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
270 SequenceI pep1 = new Sequence("Q9ZTS2", "MLAVSRGQ");
271 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
272 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
273 AlignmentI al = new Alignment(new SequenceI[] { dna1, pep1 });
275 List<SequenceI> result = new ArrayList<>();
278 * first search for a dbref nowhere on the alignment:
280 dbref = new DBRefEntry("UNIPROT", "0", "P30419");
281 CrossRef testee = new CrossRef(al.getSequencesArray(), al);
282 AlignedCodonFrame acf = new AlignedCodonFrame();
283 boolean found = testee.searchDataset(true, dna1, dbref, result, acf,
284 true, DBRefUtils.SEARCH_MODE_FULL);
286 assertTrue(result.isEmpty());
287 assertTrue(acf.isEmpty());
290 * search for a protein sequence with dbref UNIPROT:Q9ZTS2
292 acf = new AlignedCodonFrame();
293 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
294 found = testee.searchDataset(!dna1.isProtein(), dna1, dbref, result,
295 acf, false, DBRefUtils.SEARCH_MODE_FULL); // search dataset with a protein xref from a dna
296 // sequence to locate the protein product
298 assertEquals(1, result.size());
299 assertSame(pep1, result.get(0));
300 assertTrue(acf.isEmpty());
303 * search for a nucleotide sequence with dbref UNIPROT:Q9ZTS2
306 acf = new AlignedCodonFrame();
307 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
308 found = testee.searchDataset(!pep1.isProtein(), pep1, dbref, result,
309 acf, false, DBRefUtils.SEARCH_MODE_FULL); // search dataset with a protein's direct dbref to
310 // locate dna sequences with matching xref
312 assertEquals(1, result.size());
313 assertSame(dna1, result.get(0));
314 // should now have a mapping from dna to pep1
315 List<SequenceToSequenceMapping> mappings = acf.getMappings();
316 assertEquals(1, mappings.size());
317 SequenceToSequenceMapping mapping = mappings.get(0);
318 assertSame(dna1, mapping.getFromSeq());
319 assertSame(pep1, mapping.getMapping().getTo());
320 MapList mapList = mapping.getMapping().getMap();
321 assertEquals(1, mapList.getToRatio());
322 assertEquals(3, mapList.getFromRatio());
323 assertEquals(1, mapList.getFromRanges().size());
324 assertEquals(1, mapList.getFromRanges().get(0)[0]);
325 assertEquals(21, mapList.getFromRanges().get(0)[1]);
326 assertEquals(1, mapList.getToRanges().size());
327 assertEquals(1, mapList.getToRanges().get(0)[0]);
328 assertEquals(7, mapList.getToRanges().get(0)[1]);
332 * Test for finding 'product' sequences for the case where the selected
333 * sequence has a dbref with a mapping to a sequence. This represents the case
336 * <li>a fetched sequence is already decorated with its cross-reference (e.g.
337 * EMBL + translation), or</li>
338 * <li>Get Cross-References has been done once resulting in instantiated
339 * cross-reference mappings</li>
342 @Test(groups = { "Functional" })
343 public void testFindXrefSequences_fromDbRefMap()
346 * scenario: nucleotide sequence AF039662
347 * with dbref + mapping to Q9ZTS2 and P30419
348 * which themselves each have a dbref and feature
350 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
351 SequenceI pep1 = new Sequence("Q9ZTS2", "MALFQRSV");
352 SequenceI pep2 = new Sequence("P30419", "MTRRSQIF");
353 dna1.createDatasetSequence();
354 pep1.createDatasetSequence();
355 pep2.createDatasetSequence();
357 pep1.getDatasetSequence().addDBRef(
358 new DBRefEntry("Pfam", "0", "PF00111"));
359 pep1.addSequenceFeature(new SequenceFeature("type", "desc", 12, 14, 1f,
361 pep2.getDatasetSequence().addDBRef(new DBRefEntry("PDB", "0", "3JTK"));
362 pep2.addSequenceFeature(new SequenceFeature("type2", "desc2", 13, 15,
365 MapList mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 },
367 Mapping map = new Mapping(pep1, mapList);
368 DBRefEntry dbRef1 = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
369 dna1.getDatasetSequence().addDBRef(dbRef1);
370 mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 }, 3, 1);
371 map = new Mapping(pep2, mapList);
372 DBRefEntry dbRef2 = new DBRefEntry("UNIPROT", "0", "P30419", map);
373 dna1.getDatasetSequence().addDBRef(dbRef2);
376 * find UNIPROT xrefs for nucleotide sequence - it should pick up
379 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
380 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
381 .findXrefSequences("UNIPROT", true);
382 assertEquals(2, xrefs.getHeight());
385 * cross-refs alignment holds copies of the mapped sequences
386 * including copies of their dbrefs and features
388 checkCopySequence(pep1, xrefs.getSequenceAt(0));
389 checkCopySequence(pep2, xrefs.getSequenceAt(1));
393 * Helper method that verifies that 'copy' has the same name, start, end,
394 * sequence and dataset sequence object as 'original' (but is not the same
400 private void checkCopySequence(SequenceI copy, SequenceI original)
402 assertNotSame(copy, original);
403 assertSame(copy.getDatasetSequence(), original.getDatasetSequence());
404 assertEquals(copy.getName(), original.getName());
405 assertEquals(copy.getStart(), original.getStart());
406 assertEquals(copy.getEnd(), original.getEnd());
407 assertEquals(copy.getSequenceAsString(), original.getSequenceAsString());
411 * Test for finding 'product' sequences for the case where the selected
412 * sequence has a dbref with no mapping, triggering a fetch from database
414 @Test(groups = { "Functional_Failing" })
415 public void testFindXrefSequences_withFetch()
417 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
418 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "Q9ZTS2"));
419 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "P30419"));
420 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "P00314"));
421 final SequenceI pep1 = new Sequence("Q9ZTS2", "MYQLIRSSW");
422 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
424 final SequenceI pep2 = new Sequence("P00314", "MRKLLAASG");
425 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "P00314"));
428 * argument false suppresses adding DAS sources
429 * todo: define an interface type SequenceFetcherI and mock that
431 SequenceFetcher mockFetcher = new SequenceFetcher()
434 public boolean isFetchable(String source)
440 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
442 return new SequenceI[] { pep1, pep2 };
445 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
448 * find UNIPROT xrefs for nucleotide sequence
450 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
451 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
452 .findXrefSequences("UNIPROT", true);
453 assertEquals(2, xrefs.getHeight());
454 assertSame(pep1, xrefs.getSequenceAt(0));
455 assertSame(pep2, xrefs.getSequenceAt(1));
458 @AfterClass(alwaysRun = true)
459 public void tearDown()
461 SequenceFetcherFactory.setSequenceFetcher(null);
465 * Test for finding 'product' sequences for the case where both gene and
466 * transcript sequences have dbrefs to Uniprot.
468 @Test(groups = { "Functional_Failing" })
469 public void testFindXrefSequences_forGeneAndTranscripts()
474 SequenceI gene = new Sequence("ENSG00000157764", "CGCCTCCCTTCCCC");
475 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
476 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
479 * 'transcript' with CDS feature (supports mapping to protein)
481 SequenceI braf001 = new Sequence("ENST00000288602", "taagATGGCGGCGCTGa");
482 braf001.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
483 braf001.addSequenceFeature(new SequenceFeature("CDS", "", 5, 16, 0f,
487 * 'spliced transcript' with CDS ranges
489 SequenceI braf002 = new Sequence("ENST00000497784", "gCAGGCtaTCTGTTCaa");
490 braf002.addDBRef(new DBRefEntry("UNIPROT", "ENSEMBL|0", "H7C5K3"));
491 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 2, 6, 0f,
493 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 9, 15, 0f,
497 * TODO code is fragile - use of SequenceIdMatcher depends on fetched
498 * sequences having a name starting Source|Accession
499 * which happens to be true for Uniprot,PDB,EMBL but not Pfam,Rfam,Ensembl
501 final SequenceI pep1 = new Sequence("UNIPROT|P15056", "MAAL");
502 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
503 final SequenceI pep2 = new Sequence("UNIPROT|H7C5K3", "QALF");
504 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
506 * argument false suppresses adding DAS sources
507 * todo: define an interface type SequenceFetcherI and mock that
509 SequenceFetcher mockFetcher = new SequenceFetcher()
512 public boolean isFetchable(String source)
518 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
520 return new SequenceI[] { pep1, pep2 };
523 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
526 * find UNIPROT xrefs for gene and transcripts
528 * - the two proteins are retrieved but not duplicated
529 * - mappings are built from transcript (CDS) to proteins
530 * - no mappings from gene to proteins
532 SequenceI[] seqs = new SequenceI[] { gene, braf001, braf002 };
533 AlignmentI al = new Alignment(seqs);
534 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("UNIPROT",
536 assertEquals(2, xrefs.getHeight());
537 assertSame(pep1, xrefs.getSequenceAt(0));
538 assertSame(pep2, xrefs.getSequenceAt(1));
543 * Test that emulates this (real but simplified) case:
545 * UNIPROT|P0CE19 EMBL|J03321, EMBL|X06707, EMBL|M19487
546 * UNIPROT|P0CE20 EMBL|J03321, EMBL|X06707, EMBL|X07547
547 * Find cross-references for EMBL. These are mocked here as
548 * EMBL|J03321 with mappings to P0CE18, P0CE19, P0CE20
549 * EMBL|X06707 with mappings to P0CE17, P0CE19, P0CE20
550 * EMBL|M19487 with mappings to P0CE19, Q46432
551 * EMBL|X07547 with mappings to P0CE20, B0BCM4
552 * EMBL sequences are first 'fetched' (mocked here) for P0CE19.
553 * The 3 EMBL sequences are added to the alignment dataset.
554 * Their dbrefs to Uniprot products P0CE19 and P0CE20 should be matched in the
555 * alignment dataset and updated to reference the original Uniprot sequences.
556 * For the second Uniprot sequence, the J03321 and X06707 xrefs should be
557 * resolved from the dataset, and only the X07547 dbref fetched.
558 * So the end state to verify is:
559 * - 4 cross-ref sequences returned: J03321, X06707, M19487, X07547
560 * - P0CE19/20 dbrefs to EMBL sequences now have mappings
561 * - J03321 dbrefs to P0CE19/20 mapped to original Uniprot sequences
562 * - X06707 dbrefs to P0CE19/20 mapped to original Uniprot sequences
565 @Test(groups = { "Functional_Failing" })
566 public void testFindXrefSequences_uniprotEmblManyToMany()
569 * Uniprot sequences, both with xrefs to EMBL|J03321
572 SequenceI p0ce19 = new Sequence("UNIPROT|P0CE19", "KPFG");
573 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
574 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
575 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "M19487"));
576 SequenceI p0ce20 = new Sequence("UNIPROT|P0CE20", "PFGK");
577 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
578 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
579 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X07547"));
582 * EMBL sequences to be 'fetched', complete with dbrefs and mappings
583 * to their protein products (CDS location and translations are provided
584 * in EMBL XML); these should be matched to, and replaced with,
585 * the corresponding uniprot sequences after fetching
589 * J03321 with mappings to P0CE19 and P0CE20
591 final SequenceI j03321 = new Sequence("EMBL|J03321", "AAACCCTTTGGGAAAA");
592 DBRefEntry dbref1 = new DBRefEntry("UNIPROT", "0", "P0CE19");
593 MapList mapList = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 },
595 Mapping map = new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
597 // add a dbref to the mapped to sequence - should get copied to p0ce19
598 map.getTo().addDBRef(new DBRefEntry("PIR", "0", "S01875"));
600 j03321.addDBRef(dbref1);
601 DBRefEntry dbref2 = new DBRefEntry("UNIPROT", "0", "P0CE20");
602 mapList = new MapList(new int[] { 4, 15 }, new int[] { 2, 5 }, 3, 1);
603 dbref2.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
604 new MapList(mapList)));
605 j03321.addDBRef(dbref2);
608 * X06707 with mappings to P0CE19 and P0CE20
610 final SequenceI x06707 = new Sequence("EMBL|X06707", "atgAAACCCTTTGGG");
611 DBRefEntry dbref3 = new DBRefEntry("UNIPROT", "0", "P0CE19");
612 MapList map2 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
614 dbref3.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"), map2));
615 x06707.addDBRef(dbref3);
616 DBRefEntry dbref4 = new DBRefEntry("UNIPROT", "0", "P0CE20");
617 MapList map3 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
619 dbref4.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"), map3));
620 x06707.addDBRef(dbref4);
623 * M19487 with mapping to P0CE19 and Q46432
625 final SequenceI m19487 = new Sequence("EMBL|M19487", "AAACCCTTTGGG");
626 DBRefEntry dbref5 = new DBRefEntry("UNIPROT", "0", "P0CE19");
627 dbref5.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
628 new MapList(mapList)));
629 m19487.addDBRef(dbref5);
630 DBRefEntry dbref6 = new DBRefEntry("UNIPROT", "0", "Q46432");
631 dbref6.setMap(new Mapping(new Sequence("UNIPROT|Q46432", "KPFG"),
632 new MapList(mapList)));
633 m19487.addDBRef(dbref6);
636 * X07547 with mapping to P0CE20 and B0BCM4
638 final SequenceI x07547 = new Sequence("EMBL|X07547", "cccAAACCCTTTGGG");
639 DBRefEntry dbref7 = new DBRefEntry("UNIPROT", "0", "P0CE20");
640 dbref7.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
642 x07547.addDBRef(dbref7);
643 DBRefEntry dbref8 = new DBRefEntry("UNIPROT", "0", "B0BCM4");
644 dbref8.setMap(new Mapping(new Sequence("UNIPROT|B0BCM4", "KPFG"),
646 x07547.addDBRef(dbref8);
649 * mock sequence fetcher to 'return' the EMBL sequences
650 * TODO: Mockito would allow .thenReturn().thenReturn() here,
651 * and also capture and verification of the parameters
652 * passed in calls to getSequences() - important to verify that
653 * duplicate sequence fetches are not requested
655 SequenceFetcher mockFetcher = new SequenceFetcher()
660 public boolean isFetchable(String source)
666 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
671 assertEquals("Expected 3 embl seqs in first fetch", 3,
673 return new SequenceI[] { j03321, x06707, m19487 };
677 assertEquals("Expected 1 embl seq in second fetch", 1,
679 return new SequenceI[] { x07547 };
683 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
686 * find EMBL xrefs for Uniprot seqs and verify that
687 * - the EMBL xref'd sequences are retrieved without duplicates
688 * - mappings are added to the Uniprot dbrefs
689 * - mappings in the EMBL-to-Uniprot dbrefs are updated to the
690 * alignment sequences
691 * - dbrefs on the EMBL sequences are added to the original dbrefs
693 SequenceI[] seqs = new SequenceI[] { p0ce19, p0ce20 };
694 AlignmentI al = new Alignment(seqs);
695 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("EMBL",
699 * verify retrieved sequences
701 assertNotNull(xrefs);
702 assertEquals(4, xrefs.getHeight());
703 assertSame(j03321, xrefs.getSequenceAt(0));
704 assertSame(x06707, xrefs.getSequenceAt(1));
705 assertSame(m19487, xrefs.getSequenceAt(2));
706 assertSame(x07547, xrefs.getSequenceAt(3));
709 * verify mappings added to Uniprot-to-EMBL dbrefs
711 Mapping mapping = p0ce19.getDBRefs().get(0).getMap();
712 assertSame(j03321, mapping.getTo());
713 mapping = p0ce19.getDBRefs().get(1).getMap();
714 assertSame(x06707, mapping.getTo());
715 mapping = p0ce20.getDBRefs().get(0).getMap();
716 assertSame(j03321, mapping.getTo());
717 mapping = p0ce20.getDBRefs().get(1).getMap();
718 assertSame(x06707, mapping.getTo());
721 * verify dbrefs on EMBL are mapped to alignment seqs
724 assertSame(p0ce19, j03321.getDBRefs().get(0).getMap().getTo());
725 assertSame(p0ce20, j03321.getDBRefs().get(1).getMap().getTo());
726 assertSame(p0ce19, x06707.getDBRefs().get(0).getMap().getTo());
727 assertSame(p0ce20, x06707.getDBRefs().get(1).getMap().getTo());
730 * verify new dbref on EMBL dbref mapping is copied to the
731 * original Uniprot sequence
733 assertEquals(4, p0ce19.getDBRefs().size());
734 assertEquals("PIR", p0ce19.getDBRefs().get(3).getSource());
735 assertEquals("S01875", p0ce19.getDBRefs().get(3).getAccessionId());
738 @Test(groups = "Functional")
739 public void testSameSequence()
741 assertTrue(CrossRef.sameSequence(null, null));
742 SequenceI seq1 = new Sequence("seq1", "ABCDEF");
743 assertFalse(CrossRef.sameSequence(seq1, null));
744 assertFalse(CrossRef.sameSequence(null, seq1));
745 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "ABCDEF")));
746 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "abcdef")));
748 .sameSequence(seq1, new Sequence("seq2", "ABCDE-F")));
749 assertFalse(CrossRef.sameSequence(seq1, new Sequence("seq2", "BCDEF")));