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;
45 import jalview.ws.params.InvalidArgumentException;
47 import java.util.ArrayList;
48 import java.util.Arrays;
49 import java.util.List;
51 import org.testng.annotations.AfterClass;
52 import org.testng.annotations.BeforeClass;
53 import org.testng.annotations.Test;
55 public class CrossRefTest
58 @BeforeClass(alwaysRun = true)
59 public void setUpJvOptionPane()
61 JvOptionPane.setInteractiveMode(false);
62 JvOptionPane.setMockResponse(JvOptionPane.CANCEL_OPTION);
65 @Test(groups = { "Functional" })
66 public void testFindXDbRefs()
68 DBRefEntry ref1 = new DBRefEntry("UNIPROT", "1", "A123");
69 DBRefEntry ref2 = new DBRefEntry("UNIPROTKB/TREMBL", "1", "A123");
70 DBRefEntry ref3 = new DBRefEntry("pdb", "1", "A123");
71 DBRefEntry ref4 = new DBRefEntry("EMBLCDSPROTEIN", "1", "A123");
72 DBRefEntry ref5 = new DBRefEntry("embl", "1", "A123");
73 DBRefEntry ref6 = new DBRefEntry("emblCDS", "1", "A123");
74 DBRefEntry ref7 = new DBRefEntry("GeneDB", "1", "A123");
75 DBRefEntry ref8 = new DBRefEntry("PFAM", "1", "A123");
76 // ENSEMBL is a source of either dna or protein sequence data
77 DBRefEntry ref9 = new DBRefEntry("ENSEMBL", "1", "A123");
78 List<DBRefEntry> refs = Arrays.asList(new DBRefEntry[] { ref1, ref2, ref3, ref4, ref5,
79 ref6, ref7, ref8, ref9 });
84 List<DBRefEntry> found = DBRefUtils.selectDbRefs(true, refs);
85 assertEquals(4, found.size());
86 assertSame(ref5, found.get(0));
87 assertSame(ref6, found.get(1));
88 assertSame(ref7, found.get(2));
89 assertSame(ref9, found.get(3));
92 * Just the protein refs:
94 found = DBRefUtils.selectDbRefs(false, refs);
95 assertEquals(4, found.size());
96 assertSame(ref1, found.get(0));
97 assertSame(ref2, found.get(1));
98 assertSame(ref4, found.get(2));
99 assertSame(ref9, found.get(3));
103 * Test the method that finds a sequence's "product" xref source databases,
104 * which may be direct (dbrefs on the sequence), or indirect (dbrefs on
105 * sequences which share a dbref with the sequence
107 @Test(groups = { "Functional" }, enabled = true)
108 public void testFindXrefSourcesForSequence_proteinToDna()
110 SequenceI seq = new Sequence("Seq1", "MGKYQARLSS");
111 List<String> sources = new ArrayList<>();
112 AlignmentI al = new Alignment(new SequenceI[] {});
115 * first with no dbrefs to search
117 sources = new CrossRef(new SequenceI[] { seq }, al)
118 .findXrefSourcesForSequences(false);
119 assertTrue(sources.isEmpty());
122 * add some dbrefs to sequence
124 // protein db is not a candidate for findXrefSources
125 seq.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
126 // dna coding databatases are
127 seq.addDBRef(new DBRefEntry("EMBL", "0", "E2345"));
128 // a second EMBL xref should not result in a duplicate
129 seq.addDBRef(new DBRefEntry("EMBL", "0", "E2346"));
130 seq.addDBRef(new DBRefEntry("EMBLCDS", "0", "E2347"));
131 seq.addDBRef(new DBRefEntry("GENEDB", "0", "E2348"));
132 seq.addDBRef(new DBRefEntry("ENSEMBL", "0", "E2349"));
133 seq.addDBRef(new DBRefEntry("ENSEMBLGENOMES", "0", "E2350"));
134 sources = new CrossRef(new SequenceI[] { seq }, al)
135 .findXrefSourcesForSequences(false);
136 // method is patched to remove EMBL from the sources to match
137 assertEquals(4, sources.size());
138 assertEquals("[EMBLCDS, GENEDB, ENSEMBL, ENSEMBLGENOMES]",
142 * add a sequence to the alignment which has a dbref to UNIPROT|A1234
143 * and others to dna coding databases
147 seq.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
148 seq.addDBRef(new DBRefEntry("EMBLCDS", "0", "E2347"));
149 SequenceI seq2 = new Sequence("Seq2", "MGKYQARLSS");
150 seq2.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
151 seq2.addDBRef(new DBRefEntry("EMBL", "0", "E2345"));
152 seq2.addDBRef(new DBRefEntry("GENEDB", "0", "E2348"));
153 // TODO include ENSEMBLGENOMES in DBRefSource.DNACODINGDBS ?
154 al.addSequence(seq2);
155 sources = new CrossRef(new SequenceI[] { seq, seq2 }, al)
156 .findXrefSourcesForSequences(false);
157 // method removed EMBL from sources to match
158 assertEquals(2, sources.size());
159 assertEquals("[EMBLCDS, GENEDB]", sources.toString());
163 * Test for finding 'product' sequences for the case where only an indirect
164 * xref is found - not on the nucleotide sequence but on a peptide sequence in
165 * the alignment which which it shares a nucleotide dbref
167 @Test(groups = { "Functional" }, enabled = true)
168 public void testFindXrefSequences_indirectDbrefToProtein()
172 * - nucleotide dbref EMBL|AF039662
173 * - peptide dbrefs EMBL|AF039662, UNIPROT|Q9ZTS2
175 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
176 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
177 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
178 uniprotSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
179 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
182 * Find UNIPROT xrefs for nucleotide
183 * - it has no UNIPROT dbref of its own
184 * - but peptide with matching nucleotide dbref does, so is returned
186 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
187 Alignment xrefs = new CrossRef(new SequenceI[] { emblSeq }, al)
188 .findXrefSequences("UNIPROT", true);
189 assertEquals(1, xrefs.getHeight());
190 assertSame(uniprotSeq, xrefs.getSequenceAt(0));
194 * Test for finding 'product' sequences for the case where only an indirect
195 * xref is found - not on the peptide sequence but on a nucleotide sequence in
196 * the alignment which which it shares a protein dbref
198 @Test(groups = { "Functional" }, enabled = true)
199 public void testFindXrefSequences_indirectDbrefToNucleotide()
203 * - peptide dbref UNIPROT|Q9ZTS2
204 * - nucleotide dbref EMBL|AF039662, UNIPROT|Q9ZTS2
206 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
207 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
208 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
209 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
210 emblSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
213 * find EMBL xrefs for peptide sequence - it has no direct
214 * dbrefs, but the 'corresponding' nucleotide sequence does, so is returned
217 * Find EMBL xrefs for peptide
218 * - it has no EMBL dbref of its own
219 * - but nucleotide with matching peptide dbref does, so is returned
221 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
222 Alignment xrefs = new CrossRef(new SequenceI[] { uniprotSeq }, al)
223 .findXrefSequences("EMBL", false);
224 assertEquals(1, xrefs.getHeight());
225 assertSame(emblSeq, xrefs.getSequenceAt(0));
229 * Test for finding 'product' sequences for the case where the selected
230 * sequence has no dbref to the desired source, and there are no indirect
231 * references via another sequence in the alignment
233 @Test(groups = { "Functional" })
234 public void testFindXrefSequences_noDbrefs()
237 * two nucleotide sequences, one with UNIPROT dbref
239 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
240 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
241 SequenceI dna2 = new Sequence("AJ307031", "AAACCCTTT");
244 * find UNIPROT xrefs for peptide sequence - it has no direct
245 * dbrefs, and the other sequence (which has a UNIPROT dbref) is not
246 * equatable to it, so no results found
248 AlignmentI al = new Alignment(new SequenceI[] { dna1, dna2 });
249 Alignment xrefs = new CrossRef(new SequenceI[] { dna2 }, al)
250 .findXrefSequences("UNIPROT", true);
255 * Tests for the method that searches an alignment (with one sequence
256 * excluded) for protein/nucleotide sequences with a given cross-reference
258 @Test(groups = { "Functional" }, enabled = true)
259 public void testSearchDataset()
262 * nucleotide sequence with UNIPROT AND EMBL dbref
263 * peptide sequence with UNIPROT dbref
265 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
266 Mapping map = new Mapping(new Sequence("pep2", "MLAVSRG"), new MapList(
267 new int[] { 1, 21 }, new int[] { 1, 7 }, 3, 1));
268 DBRefEntry dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
269 dna1.addDBRef(dbref);
270 dna1.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
271 SequenceI pep1 = new Sequence("Q9ZTS2", "MLAVSRGQ");
272 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
273 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
274 AlignmentI al = new Alignment(new SequenceI[] { dna1, pep1 });
276 List<SequenceI> result = new ArrayList<>();
279 * first search for a dbref nowhere on the alignment:
281 dbref = new DBRefEntry("UNIPROT", "0", "P30419");
282 CrossRef testee = new CrossRef(al.getSequencesArray(), al);
283 AlignedCodonFrame acf = new AlignedCodonFrame();
284 boolean found = testee.searchDataset(true, dna1, dbref, result, acf,
285 true, DBRefUtils.SEARCH_MODE_FULL);
287 assertTrue(result.isEmpty());
288 assertTrue(acf.isEmpty());
291 * search for a protein sequence with dbref UNIPROT:Q9ZTS2
293 acf = new AlignedCodonFrame();
294 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
295 found = testee.searchDataset(!dna1.isProtein(), dna1, dbref, result,
296 acf, false, DBRefUtils.SEARCH_MODE_FULL); // search dataset with a protein xref from a dna
297 // sequence to locate the protein product
299 assertEquals(1, result.size());
300 assertSame(pep1, result.get(0));
301 assertTrue(acf.isEmpty());
304 * search for a nucleotide sequence with dbref UNIPROT:Q9ZTS2
307 acf = new AlignedCodonFrame();
308 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
309 found = testee.searchDataset(!pep1.isProtein(), pep1, dbref, result,
310 acf, false, DBRefUtils.SEARCH_MODE_FULL); // search dataset with a protein's direct dbref to
311 // locate dna sequences with matching xref
313 assertEquals(1, result.size());
314 assertSame(dna1, result.get(0));
315 // should now have a mapping from dna to pep1
316 List<SequenceToSequenceMapping> mappings = acf.getMappings();
317 assertEquals(1, mappings.size());
318 SequenceToSequenceMapping mapping = mappings.get(0);
319 assertSame(dna1, mapping.getFromSeq());
320 assertSame(pep1, mapping.getMapping().getTo());
321 MapList mapList = mapping.getMapping().getMap();
322 assertEquals(1, mapList.getToRatio());
323 assertEquals(3, mapList.getFromRatio());
324 assertEquals(1, mapList.getFromRanges().size());
325 assertEquals(1, mapList.getFromRanges().get(0)[0]);
326 assertEquals(21, mapList.getFromRanges().get(0)[1]);
327 assertEquals(1, mapList.getToRanges().size());
328 assertEquals(1, mapList.getToRanges().get(0)[0]);
329 assertEquals(7, mapList.getToRanges().get(0)[1]);
333 * Test for finding 'product' sequences for the case where the selected
334 * sequence has a dbref with a mapping to a sequence. This represents the case
337 * <li>a fetched sequence is already decorated with its cross-reference (e.g.
338 * EMBL + translation), or</li>
339 * <li>Get Cross-References has been done once resulting in instantiated
340 * cross-reference mappings</li>
343 @Test(groups = { "Functional" })
344 public void testFindXrefSequences_fromDbRefMap()
347 * scenario: nucleotide sequence AF039662
348 * with dbref + mapping to Q9ZTS2 and P30419
349 * which themselves each have a dbref and feature
351 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
352 SequenceI pep1 = new Sequence("Q9ZTS2", "MALFQRSV");
353 SequenceI pep2 = new Sequence("P30419", "MTRRSQIF");
354 dna1.createDatasetSequence();
355 pep1.createDatasetSequence();
356 pep2.createDatasetSequence();
358 pep1.getDatasetSequence().addDBRef(
359 new DBRefEntry("Pfam", "0", "PF00111"));
360 pep1.addSequenceFeature(new SequenceFeature("type", "desc", 12, 14, 1f,
362 pep2.getDatasetSequence().addDBRef(new DBRefEntry("PDB", "0", "3JTK"));
363 pep2.addSequenceFeature(new SequenceFeature("type2", "desc2", 13, 15,
366 MapList mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 },
368 Mapping map = new Mapping(pep1, mapList);
369 DBRefEntry dbRef1 = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
370 dna1.getDatasetSequence().addDBRef(dbRef1);
371 mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 }, 3, 1);
372 map = new Mapping(pep2, mapList);
373 DBRefEntry dbRef2 = new DBRefEntry("UNIPROT", "0", "P30419", map);
374 dna1.getDatasetSequence().addDBRef(dbRef2);
377 * find UNIPROT xrefs for nucleotide sequence - it should pick up
380 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
381 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
382 .findXrefSequences("UNIPROT", true);
383 assertEquals(2, xrefs.getHeight());
386 * cross-refs alignment holds copies of the mapped sequences
387 * including copies of their dbrefs and features
389 checkCopySequence(pep1, xrefs.getSequenceAt(0));
390 checkCopySequence(pep2, xrefs.getSequenceAt(1));
394 * Helper method that verifies that 'copy' has the same name, start, end,
395 * sequence and dataset sequence object as 'original' (but is not the same
401 private void checkCopySequence(SequenceI copy, SequenceI original)
403 assertNotSame(copy, original);
404 assertSame(copy.getDatasetSequence(), original.getDatasetSequence());
405 assertEquals(copy.getName(), original.getName());
406 assertEquals(copy.getStart(), original.getStart());
407 assertEquals(copy.getEnd(), original.getEnd());
408 assertEquals(copy.getSequenceAsString(), original.getSequenceAsString());
412 * Test for finding 'product' sequences for the case where the selected
413 * sequence has a dbref with no mapping, triggering a fetch from database
415 @Test(groups = { "Functional_Failing" })
416 public void testFindXrefSequences_withFetch()
418 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
419 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "Q9ZTS2"));
420 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "P30419"));
421 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "P00314"));
422 final SequenceI pep1 = new Sequence("Q9ZTS2", "MYQLIRSSW");
423 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
425 final SequenceI pep2 = new Sequence("P00314", "MRKLLAASG");
426 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "P00314"));
429 * argument false suppresses adding DAS sources
430 * todo: define an interface type SequenceFetcherI and mock that
432 SequenceFetcher mockFetcher = new SequenceFetcher()
435 public boolean isFetchable(String source)
441 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
443 return new SequenceI[] { pep1, pep2 };
446 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
449 * find UNIPROT xrefs for nucleotide sequence
451 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
452 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
453 .findXrefSequences("UNIPROT", true);
454 assertEquals(2, xrefs.getHeight());
455 assertSame(pep1, xrefs.getSequenceAt(0));
456 assertSame(pep2, xrefs.getSequenceAt(1));
459 @AfterClass(alwaysRun = true)
460 public void tearDown()
462 SequenceFetcherFactory.setSequenceFetcher(null);
466 * Test for finding 'product' sequences for the case where both gene and
467 * transcript sequences have dbrefs to Uniprot.
469 @Test(groups = { "Functional_Failing" })
470 public void testFindXrefSequences_forGeneAndTranscripts()
475 SequenceI gene = new Sequence("ENSG00000157764", "CGCCTCCCTTCCCC");
476 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
477 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
480 * 'transcript' with CDS feature (supports mapping to protein)
482 SequenceI braf001 = new Sequence("ENST00000288602", "taagATGGCGGCGCTGa");
483 braf001.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
484 braf001.addSequenceFeature(new SequenceFeature("CDS", "", 5, 16, 0f,
488 * 'spliced transcript' with CDS ranges
490 SequenceI braf002 = new Sequence("ENST00000497784", "gCAGGCtaTCTGTTCaa");
491 braf002.addDBRef(new DBRefEntry("UNIPROT", "ENSEMBL|0", "H7C5K3"));
492 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 2, 6, 0f,
494 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 9, 15, 0f,
498 * TODO code is fragile - use of SequenceIdMatcher depends on fetched
499 * sequences having a name starting Source|Accession
500 * which happens to be true for Uniprot,PDB,EMBL but not Pfam,Rfam,Ensembl
502 final SequenceI pep1 = new Sequence("UNIPROT|P15056", "MAAL");
503 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
504 final SequenceI pep2 = new Sequence("UNIPROT|H7C5K3", "QALF");
505 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
507 * argument false suppresses adding DAS sources
508 * todo: define an interface type SequenceFetcherI and mock that
510 SequenceFetcher mockFetcher = new SequenceFetcher()
513 public boolean isFetchable(String source)
519 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
521 return new SequenceI[] { pep1, pep2 };
524 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
527 * find UNIPROT xrefs for gene and transcripts
529 * - the two proteins are retrieved but not duplicated
530 * - mappings are built from transcript (CDS) to proteins
531 * - no mappings from gene to proteins
533 SequenceI[] seqs = new SequenceI[] { gene, braf001, braf002 };
534 AlignmentI al = new Alignment(seqs);
535 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("UNIPROT",
537 assertEquals(2, xrefs.getHeight());
538 assertSame(pep1, xrefs.getSequenceAt(0));
539 assertSame(pep2, xrefs.getSequenceAt(1));
544 * Test that emulates this (real but simplified) case:
546 * UNIPROT|P0CE19 EMBL|J03321, EMBL|X06707, EMBL|M19487
547 * UNIPROT|P0CE20 EMBL|J03321, EMBL|X06707, EMBL|X07547
548 * Find cross-references for EMBL. These are mocked here as
549 * EMBL|J03321 with mappings to P0CE18, P0CE19, P0CE20
550 * EMBL|X06707 with mappings to P0CE17, P0CE19, P0CE20
551 * EMBL|M19487 with mappings to P0CE19, Q46432
552 * EMBL|X07547 with mappings to P0CE20, B0BCM4
553 * EMBL sequences are first 'fetched' (mocked here) for P0CE19.
554 * The 3 EMBL sequences are added to the alignment dataset.
555 * Their dbrefs to Uniprot products P0CE19 and P0CE20 should be matched in the
556 * alignment dataset and updated to reference the original Uniprot sequences.
557 * For the second Uniprot sequence, the J03321 and X06707 xrefs should be
558 * resolved from the dataset, and only the X07547 dbref fetched.
559 * So the end state to verify is:
560 * - 4 cross-ref sequences returned: J03321, X06707, M19487, X07547
561 * - P0CE19/20 dbrefs to EMBL sequences now have mappings
562 * - J03321 dbrefs to P0CE19/20 mapped to original Uniprot sequences
563 * - X06707 dbrefs to P0CE19/20 mapped to original Uniprot sequences
566 @Test(groups = { "Functional_Failing" })
567 public void testFindXrefSequences_uniprotEmblManyToMany()
570 * Uniprot sequences, both with xrefs to EMBL|J03321
573 SequenceI p0ce19 = new Sequence("UNIPROT|P0CE19", "KPFG");
574 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
575 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
576 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "M19487"));
577 SequenceI p0ce20 = new Sequence("UNIPROT|P0CE20", "PFGK");
578 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
579 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
580 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X07547"));
583 * EMBL sequences to be 'fetched', complete with dbrefs and mappings
584 * to their protein products (CDS location and translations are provided
585 * in EMBL XML); these should be matched to, and replaced with,
586 * the corresponding uniprot sequences after fetching
590 * J03321 with mappings to P0CE19 and P0CE20
592 final SequenceI j03321 = new Sequence("EMBL|J03321", "AAACCCTTTGGGAAAA");
593 DBRefEntry dbref1 = new DBRefEntry("UNIPROT", "0", "P0CE19");
594 MapList mapList = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 },
596 Mapping map = new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
598 // add a dbref to the mapped to sequence - should get copied to p0ce19
599 map.getTo().addDBRef(new DBRefEntry("PIR", "0", "S01875"));
601 j03321.addDBRef(dbref1);
602 DBRefEntry dbref2 = new DBRefEntry("UNIPROT", "0", "P0CE20");
603 mapList = new MapList(new int[] { 4, 15 }, new int[] { 2, 5 }, 3, 1);
604 dbref2.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
605 new MapList(mapList)));
606 j03321.addDBRef(dbref2);
609 * X06707 with mappings to P0CE19 and P0CE20
611 final SequenceI x06707 = new Sequence("EMBL|X06707", "atgAAACCCTTTGGG");
612 DBRefEntry dbref3 = new DBRefEntry("UNIPROT", "0", "P0CE19");
613 MapList map2 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
615 dbref3.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"), map2));
616 x06707.addDBRef(dbref3);
617 DBRefEntry dbref4 = new DBRefEntry("UNIPROT", "0", "P0CE20");
618 MapList map3 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
620 dbref4.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"), map3));
621 x06707.addDBRef(dbref4);
624 * M19487 with mapping to P0CE19 and Q46432
626 final SequenceI m19487 = new Sequence("EMBL|M19487", "AAACCCTTTGGG");
627 DBRefEntry dbref5 = new DBRefEntry("UNIPROT", "0", "P0CE19");
628 dbref5.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
629 new MapList(mapList)));
630 m19487.addDBRef(dbref5);
631 DBRefEntry dbref6 = new DBRefEntry("UNIPROT", "0", "Q46432");
632 dbref6.setMap(new Mapping(new Sequence("UNIPROT|Q46432", "KPFG"),
633 new MapList(mapList)));
634 m19487.addDBRef(dbref6);
637 * X07547 with mapping to P0CE20 and B0BCM4
639 final SequenceI x07547 = new Sequence("EMBL|X07547", "cccAAACCCTTTGGG");
640 DBRefEntry dbref7 = new DBRefEntry("UNIPROT", "0", "P0CE20");
641 dbref7.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
643 x07547.addDBRef(dbref7);
644 DBRefEntry dbref8 = new DBRefEntry("UNIPROT", "0", "B0BCM4");
645 dbref8.setMap(new Mapping(new Sequence("UNIPROT|B0BCM4", "KPFG"),
647 x07547.addDBRef(dbref8);
650 * mock sequence fetcher to 'return' the EMBL sequences
651 * TODO: Mockito would allow .thenReturn().thenReturn() here,
652 * and also capture and verification of the parameters
653 * passed in calls to getSequences() - important to verify that
654 * duplicate sequence fetches are not requested
656 SequenceFetcher mockFetcher = new SequenceFetcher()
661 public boolean isFetchable(String source)
667 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
672 assertEquals("Expected 3 embl seqs in first fetch", 3,
674 return new SequenceI[] { j03321, x06707, m19487 };
678 assertEquals("Expected 1 embl seq in second fetch", 1,
680 return new SequenceI[] { x07547 };
684 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
687 * find EMBL xrefs for Uniprot seqs and verify that
688 * - the EMBL xref'd sequences are retrieved without duplicates
689 * - mappings are added to the Uniprot dbrefs
690 * - mappings in the EMBL-to-Uniprot dbrefs are updated to the
691 * alignment sequences
692 * - dbrefs on the EMBL sequences are added to the original dbrefs
694 SequenceI[] seqs = new SequenceI[] { p0ce19, p0ce20 };
695 AlignmentI al = new Alignment(seqs);
696 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("EMBL",
700 * verify retrieved sequences
702 assertNotNull(xrefs);
703 assertEquals(4, xrefs.getHeight());
704 assertSame(j03321, xrefs.getSequenceAt(0));
705 assertSame(x06707, xrefs.getSequenceAt(1));
706 assertSame(m19487, xrefs.getSequenceAt(2));
707 assertSame(x07547, xrefs.getSequenceAt(3));
710 * verify mappings added to Uniprot-to-EMBL dbrefs
712 Mapping mapping = p0ce19.getDBRefs().get(0).getMap();
713 assertSame(j03321, mapping.getTo());
714 mapping = p0ce19.getDBRefs().get(1).getMap();
715 assertSame(x06707, mapping.getTo());
716 mapping = p0ce20.getDBRefs().get(0).getMap();
717 assertSame(j03321, mapping.getTo());
718 mapping = p0ce20.getDBRefs().get(1).getMap();
719 assertSame(x06707, mapping.getTo());
722 * verify dbrefs on EMBL are mapped to alignment seqs
725 assertSame(p0ce19, j03321.getDBRefs().get(0).getMap().getTo());
726 assertSame(p0ce20, j03321.getDBRefs().get(1).getMap().getTo());
727 assertSame(p0ce19, x06707.getDBRefs().get(0).getMap().getTo());
728 assertSame(p0ce20, x06707.getDBRefs().get(1).getMap().getTo());
731 * verify new dbref on EMBL dbref mapping is copied to the
732 * original Uniprot sequence
734 assertEquals(4, p0ce19.getDBRefs().size());
735 assertEquals("PIR", p0ce19.getDBRefs().get(3).getSource());
736 assertEquals("S01875", p0ce19.getDBRefs().get(3).getAccessionId());
739 @Test(groups = "Functional")
740 public void testSameSequence()
742 assertTrue(CrossRef.sameSequence(null, null));
743 SequenceI seq1 = new Sequence("seq1", "ABCDEF");
744 assertFalse(CrossRef.sameSequence(seq1, null));
745 assertFalse(CrossRef.sameSequence(null, seq1));
746 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "ABCDEF")));
747 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "abcdef")));
749 .sameSequence(seq1, new Sequence("seq2", "ABCDE-F")));
750 assertFalse(CrossRef.sameSequence(seq1, new Sequence("seq2", "BCDEF")));