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;
30 import static org.testng.internal.junit.ArrayAsserts.assertArrayEquals;
32 import jalview.datamodel.Alignment;
33 import jalview.datamodel.AlignmentI;
34 import jalview.datamodel.DBRefEntry;
35 import jalview.datamodel.Mapping;
36 import jalview.datamodel.Sequence;
37 import jalview.datamodel.SequenceFeature;
38 import jalview.datamodel.SequenceI;
39 import jalview.util.DBRefUtils;
40 import jalview.util.MapList;
41 import jalview.ws.SequenceFetcher;
42 import jalview.ws.SequenceFetcherFactory;
44 import java.util.ArrayList;
45 import java.util.List;
47 import org.testng.annotations.AfterClass;
48 import org.testng.annotations.Test;
50 public class CrossRefTest
52 @Test(groups = { "Functional" })
53 public void testFindXDbRefs()
55 DBRefEntry ref1 = new DBRefEntry("UNIPROT", "1", "A123");
56 DBRefEntry ref2 = new DBRefEntry("UNIPROTKB/TREMBL", "1", "A123");
57 DBRefEntry ref3 = new DBRefEntry("pdb", "1", "A123");
58 DBRefEntry ref4 = new DBRefEntry("EMBLCDSPROTEIN", "1", "A123");
59 DBRefEntry ref5 = new DBRefEntry("embl", "1", "A123");
60 DBRefEntry ref6 = new DBRefEntry("emblCDS", "1", "A123");
61 DBRefEntry ref7 = new DBRefEntry("GeneDB", "1", "A123");
62 DBRefEntry ref8 = new DBRefEntry("PFAM", "1", "A123");
63 // ENSEMBL is a source of either dna or protein sequence data
64 DBRefEntry ref9 = new DBRefEntry("ENSEMBL", "1", "A123");
65 DBRefEntry[] refs = new DBRefEntry[] { ref1, ref2, ref3, ref4, ref5,
66 ref6, ref7, ref8, ref9 };
71 DBRefEntry[] found = DBRefUtils.selectDbRefs(true, refs);
72 assertEquals(4, found.length);
73 assertSame(ref5, found[0]);
74 assertSame(ref6, found[1]);
75 assertSame(ref7, found[2]);
76 assertSame(ref9, found[3]);
79 * Just the protein refs:
81 found = DBRefUtils.selectDbRefs(false, refs);
82 assertEquals(5, found.length);
83 assertSame(ref1, found[0]);
84 assertSame(ref2, found[1]);
85 assertSame(ref3, found[2]);
86 assertSame(ref4, found[3]);
87 assertSame(ref9, found[4]);
91 * Test the method that finds a sequence's "product" xref source databases,
92 * which may be direct (dbrefs on the sequence), or indirect (dbrefs on
93 * sequences which share a dbref with the sequence
95 @Test(groups = { "Functional" }, enabled = false)
96 public void testFindXrefSourcesForSequence_proteinToDna()
98 SequenceI seq = new Sequence("Seq1", "MGKYQARLSS");
99 List<String> sources = new ArrayList<String>();
100 AlignmentI al = new Alignment(new SequenceI[] {});
103 * first with no dbrefs to search
105 sources = new CrossRef(new SequenceI[] { seq }, al)
106 .findXrefSourcesForSequences(false);
107 assertTrue(sources.isEmpty());
110 * add some dbrefs to sequence
112 // protein db is not a candidate for findXrefSources
113 seq.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
114 // dna coding databatases are
115 seq.addDBRef(new DBRefEntry("EMBL", "0", "E2345"));
116 // a second EMBL xref should not result in a duplicate
117 seq.addDBRef(new DBRefEntry("EMBL", "0", "E2346"));
118 seq.addDBRef(new DBRefEntry("EMBLCDS", "0", "E2347"));
119 seq.addDBRef(new DBRefEntry("GENEDB", "0", "E2348"));
120 seq.addDBRef(new DBRefEntry("ENSEMBL", "0", "E2349"));
121 seq.addDBRef(new DBRefEntry("ENSEMBLGENOMES", "0", "E2350"));
122 sources = new CrossRef(new SequenceI[] { seq }, al)
123 .findXrefSourcesForSequences(false);
124 assertEquals(4, sources.size());
125 assertEquals("[EMBL, EMBLCDS, GENEDB, ENSEMBL]",
129 * add a sequence to the alignment which has a dbref to UNIPROT|A1234
130 * and others to dna coding databases
134 seq.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
135 seq.addDBRef(new DBRefEntry("EMBLCDS", "0", "E2347"));
136 SequenceI seq2 = new Sequence("Seq2", "MGKYQARLSS");
137 seq2.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
138 seq2.addDBRef(new DBRefEntry("EMBL", "0", "E2345"));
139 seq2.addDBRef(new DBRefEntry("GENEDB", "0", "E2348"));
140 // TODO include ENSEMBLGENOMES in DBRefSource.DNACODINGDBS ?
141 al.addSequence(seq2);
142 sources = new CrossRef(new SequenceI[] { seq }, al)
143 .findXrefSourcesForSequences(false);
144 assertEquals(3, sources.size());
145 assertEquals("[EMBLCDS, EMBL, GENEDB]", sources.toString());
149 * Test for finding 'product' sequences for the case where only an indirect
150 * xref is found - not on the nucleotide sequence but on a peptide sequence in
151 * the alignment which which it shares a nucleotide dbref
153 @Test(groups = { "Functional" }, enabled = false)
154 public void testFindXrefSequences_indirectDbrefToProtein()
158 * - nucleotide dbref EMBL|AF039662
159 * - peptide dbrefs EMBL|AF039662, UNIPROT|Q9ZTS2
161 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
162 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
163 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
164 uniprotSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
165 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
168 * Find UNIPROT xrefs for nucleotide
169 * - it has no UNIPROT dbref of its own
170 * - but peptide with matching nucleotide dbref does, so is returned
172 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
173 Alignment xrefs = new CrossRef(new SequenceI[] { emblSeq }, al)
174 .findXrefSequences("UNIPROT", true);
175 assertEquals(1, xrefs.getHeight());
176 assertSame(uniprotSeq, xrefs.getSequenceAt(0));
180 * Test for finding 'product' sequences for the case where only an indirect
181 * xref is found - not on the peptide sequence but on a nucleotide sequence in
182 * the alignment which which it shares a protein dbref
184 @Test(groups = { "Functional" }, enabled = false)
185 public void testFindXrefSequences_indirectDbrefToNucleotide()
189 * - peptide dbref UNIPROT|Q9ZTS2
190 * - nucleotide dbref EMBL|AF039662, UNIPROT|Q9ZTS2
192 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
193 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
194 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
195 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
196 emblSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
199 * find EMBL xrefs for peptide sequence - it has no direct
200 * dbrefs, but the 'corresponding' nucleotide sequence does, so is returned
203 * Find EMBL xrefs for peptide
204 * - it has no EMBL dbref of its own
205 * - but nucleotide with matching peptide dbref does, so is returned
207 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
208 Alignment xrefs = new CrossRef(new SequenceI[] { uniprotSeq },
210 .findXrefSequences("EMBL", true);
211 assertEquals(1, xrefs.getHeight());
212 assertSame(emblSeq, xrefs.getSequenceAt(0));
216 * Test for finding 'product' sequences for the case where the selected
217 * sequence has no dbref to the desired source, and there are no indirect
218 * references via another sequence in the alignment
220 @Test(groups = { "Functional" })
221 public void testFindXrefSequences_noDbrefs()
224 * two nucleotide sequences, one with UNIPROT dbref
226 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
227 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
228 SequenceI dna2 = new Sequence("AJ307031", "AAACCCTTT");
231 * find UNIPROT xrefs for peptide sequence - it has no direct
232 * dbrefs, and the other sequence (which has a UNIPROT dbref) is not
233 * equatable to it, so no results found
235 AlignmentI al = new Alignment(new SequenceI[] { dna1, dna2 });
236 Alignment xrefs = new CrossRef(new SequenceI[] { dna2 }, al)
237 .findXrefSequences("UNIPROT", true);
242 * Tests for the method that searches an alignment (with one sequence
243 * excluded) for protein/nucleotide sequences with a given cross-reference
245 @Test(groups = { "Functional" }, enabled = false)
246 public void testSearchDataset()
249 * nucleotide sequence with UNIPROT AND EMBL dbref
250 * peptide sequence with UNIPROT dbref
252 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
253 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
254 dna1.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
255 SequenceI pep1 = new Sequence("Q9ZTS2", "MLAVSRGQ");
256 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
257 AlignmentI al = new Alignment(new SequenceI[] { dna1, pep1 });
259 List<SequenceI> result = new ArrayList<SequenceI>();
262 * first search for a dbref nowhere on the alignment:
264 DBRefEntry dbref = new DBRefEntry("UNIPROT", "0", "P30419");
265 CrossRef testee = new CrossRef(al.getSequencesArray(), al);
266 boolean found = testee.searchDataset(true, dna1, dbref, result, null,
269 assertTrue(result.isEmpty());
271 // TODO we are setting direct=true here but it is set to
272 // false in Jalview code...
275 * search for a protein sequence with dbref UNIPROT:Q9ZTS2
277 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
278 found = testee.searchDataset(true, dna1, dbref, result, null, true);
280 assertEquals(1, result.size());
281 assertSame(pep1, result.get(0));
284 * search for a nucleotide sequence with dbref UNIPROT:Q9ZTS2
287 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
288 found = testee.searchDataset(false, pep1, dbref, result, null, false);
290 assertEquals(1, result.size());
291 assertSame(dna1, result.get(0));
295 * Test for finding 'product' sequences for the case where the selected
296 * sequence has a dbref with a mapping to a sequence
298 @Test(groups = { "Functional" })
299 public void testFindXrefSequences_fromDbRefMap()
302 * two peptide sequences each with a DBRef and SequenceFeature
304 SequenceI pep1 = new Sequence("Q9ZTS2", "MALFQRSV");
305 pep1.addDBRef(new DBRefEntry("Pfam", "0", "PF00111"));
306 pep1.addSequenceFeature(new SequenceFeature("type", "desc", 12, 14, 1f,
308 SequenceI pep2 = new Sequence("P30419", "MTRRSQIF");
309 pep2.addDBRef(new DBRefEntry("PDB", "0", "3JTK"));
310 pep2.addSequenceFeature(new SequenceFeature("type2", "desc2", 13, 15,
314 * nucleotide sequence (to go in the alignment)
316 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
319 * add DBRefEntry's to dna1 with mappings from dna to both peptides
321 MapList mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 },
323 Mapping map = new Mapping(pep1, mapList);
324 DBRefEntry dbRef1 = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
325 dna1.addDBRef(dbRef1);
326 mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 }, 3, 1);
327 map = new Mapping(pep2, mapList);
328 DBRefEntry dbRef2 = new DBRefEntry("UNIPROT", "0", "P30419", map);
329 dna1.addDBRef(dbRef2);
332 * find UNIPROT xrefs for nucleotide sequence - it should pick up
335 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
336 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
337 .findXrefSequences("UNIPROT", true);
338 assertEquals(2, xrefs.getHeight());
341 * cross-refs alignment holds copies of the mapped sequences
342 * including copies of their dbrefs and features
344 checkCopySequence(pep1, xrefs.getSequenceAt(0));
345 checkCopySequence(pep2, xrefs.getSequenceAt(1));
349 * Helper method to assert seq1 looks like a copy of seq2
354 private void checkCopySequence(SequenceI seq1, SequenceI seq2)
356 assertNotSame(seq1, seq2);
357 assertEquals(seq1.getName(), seq2.getName());
358 assertEquals(seq1.getStart(), seq2.getStart());
359 assertEquals(seq1.getEnd(), seq2.getEnd());
360 assertEquals(seq1.getSequenceAsString(), seq2.getSequenceAsString());
365 assertArrayEquals(seq1.getDBRefs(), seq2.getDBRefs());
366 // check one to verify a copy, not the same object
367 if (seq1.getDBRefs().length > 0)
369 assertNotSame(seq1.getDBRefs()[0], seq2.getDBRefs()[0]);
375 assertArrayEquals(seq1.getSequenceFeatures(),
376 seq2.getSequenceFeatures());
377 if (seq1.getSequenceFeatures().length > 0)
379 assertNotSame(seq1.getSequenceFeatures()[0],
380 seq2.getSequenceFeatures()[0]);
385 * Test for finding 'product' sequences for the case where the selected
386 * sequence has a dbref with no mapping, triggering a fetch from database
388 @Test(groups = { "Functional" })
389 public void testFindXrefSequences_withFetch()
391 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
392 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
393 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "P30419"));
394 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "P00314"));
395 final SequenceI pep1 = new Sequence("Q9ZTS2", "MYQLIRSSW");
396 final SequenceI pep2 = new Sequence("P00314", "MRKLLAASG");
399 * argument false suppresses adding DAS sources
400 * todo: define an interface type SequenceFetcherI and mock that
402 SequenceFetcher mockFetcher = new SequenceFetcher(false)
405 public boolean isFetchable(String source)
411 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
413 return new SequenceI[] { pep1, pep2 };
416 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
419 * find UNIPROT xrefs for nucleotide sequence
421 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
422 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
423 .findXrefSequences("UNIPROT", true);
424 assertEquals(2, xrefs.getHeight());
425 assertSame(pep1, xrefs.getSequenceAt(0));
426 assertSame(pep2, xrefs.getSequenceAt(1));
430 public void tearDown()
432 SequenceFetcherFactory.setSequenceFetcher(null);
436 * Test for finding 'product' sequences for the case where both gene and
437 * transcript sequences have dbrefs to Uniprot.
439 @Test(groups = { "Functional" })
440 public void testFindXrefSequences_forGeneAndTranscripts()
445 SequenceI gene = new Sequence("ENSG00000157764", "CGCCTCCCTTCCCC");
446 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
447 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
450 * 'transcript' with CDS feature (supports mapping to protein)
452 SequenceI braf001 = new Sequence("ENST00000288602", "taagATGGCGGCGCTGa");
453 braf001.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
454 braf001.addSequenceFeature(new SequenceFeature("CDS", "", 5, 16, 0f,
458 * 'spliced transcript' with CDS ranges
460 SequenceI braf002 = new Sequence("ENST00000497784", "gCAGGCtaTCTGTTCaa");
461 braf002.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
462 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 2, 6, 0f,
464 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 9, 15, 0f,
468 * TODO code is fragile - use of SequenceIdMatcher depends on fetched
469 * sequences having a name starting Source|Accession
470 * which happens to be true for Uniprot,PDB,EMBL but not Pfam,Rfam,Ensembl
472 final SequenceI pep1 = new Sequence("UNIPROT|P15056", "MAAL");
473 final SequenceI pep2 = new Sequence("UNIPROT|H7C5K3", "QALF");
476 * argument false suppresses adding DAS sources
477 * todo: define an interface type SequenceFetcherI and mock that
479 SequenceFetcher mockFetcher = new SequenceFetcher(false)
482 public boolean isFetchable(String source)
488 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
490 return new SequenceI[] { pep1, pep2 };
493 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
496 * find UNIPROT xrefs for gene and transcripts
498 * - the two proteins are retrieved but not duplicated
499 * - mappings are built from transcript (CDS) to proteins
500 * - no mappings from gene to proteins
502 SequenceI[] seqs = new SequenceI[] { gene, braf001, braf002 };
503 AlignmentI al = new Alignment(seqs);
504 Alignment xrefs = new CrossRef(seqs, al)
505 .findXrefSequences("UNIPROT",
507 assertEquals(2, xrefs.getHeight());
508 assertSame(pep1, xrefs.getSequenceAt(0));
509 assertSame(pep2, xrefs.getSequenceAt(1));
514 * Test that emulates this (real but simplified) case:
516 * UNIPROT|P0CE19 EMBL|J03321, EMBL|X06707, EMBL|M19487
517 * UNIPROT|P0CE20 EMBL|J03321, EMBL|X06707, EMBL|X07547
518 * Find cross-references for EMBL. These are mocked here as
519 * EMBL|J03321 with mappings to P0CE18, P0CE19, P0CE20
520 * EMBL|X06707 with mappings to P0CE17, P0CE19, P0CE20
521 * EMBL|M19487 with mappings to P0CE19, Q46432
522 * EMBL|X07547 with mappings to P0CE20, B0BCM4
523 * EMBL sequences are first 'fetched' (mocked here) for P0CE19.
524 * The 3 EMBL sequences are added to the alignment dataset.
525 * Their dbrefs to Uniprot products P0CE19 and P0CE20 should be matched in the
526 * alignment dataset and updated to reference the original Uniprot sequences.
527 * For the second Uniprot sequence, the J03321 and X06707 xrefs should be
528 * resolved from the dataset, and only the X07547 dbref fetched.
529 * So the end state to verify is:
530 * - 4 cross-ref sequences returned: J03321, X06707, M19487, X07547
531 * - P0CE19/20 dbrefs to EMBL sequences now have mappings
532 * - J03321 dbrefs to P0CE19/20 mapped to original Uniprot sequences
533 * - X06707 dbrefs to P0CE19/20 mapped to original Uniprot sequences
536 @Test(groups = { "Functional" })
537 public void testFindXrefSequences_uniprotEmblManyToMany()
540 * Uniprot sequences, both with xrefs to EMBL|J03321
543 SequenceI p0ce19 = new Sequence("UNIPROT|P0CE19", "KPFG");
544 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
545 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
546 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "M19487"));
547 SequenceI p0ce20 = new Sequence("UNIPROT|P0CE20", "PFGK");
548 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
549 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
550 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X07547"));
553 * EMBL sequences to be 'fetched', complete with dbrefs and mappings
554 * to their protein products (CDS location and translations are provided
555 * in EMBL XML); these should be matched to, and replaced with,
556 * the corresponding uniprot sequences after fetching
560 * J03321 with mappings to P0CE19 and P0CE20
562 final SequenceI j03321 = new Sequence("EMBL|J03321", "AAACCCTTTGGGAAAA");
563 DBRefEntry dbref1 = new DBRefEntry("UNIPROT", "0", "P0CE19");
564 MapList mapList = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 },
566 Mapping map = new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"), mapList);
567 // add a dbref to the mapped to sequence - should get copied to p0ce19
568 map.getTo().addDBRef(new DBRefEntry("PIR", "0", "S01875"));
570 j03321.addDBRef(dbref1);
571 DBRefEntry dbref2 = new DBRefEntry("UNIPROT", "0", "P0CE20");
572 mapList = new MapList(new int[] { 4, 15 }, new int[] { 2, 5 }, 3, 1);
573 dbref2.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
574 new MapList(mapList)));
575 j03321.addDBRef(dbref2);
578 * X06707 with mappings to P0CE19 and P0CE20
580 final SequenceI x06707 = new Sequence("EMBL|X06707", "atgAAACCCTTTGGG");
581 DBRefEntry dbref3 = new DBRefEntry("UNIPROT", "0", "P0CE19");
582 MapList map2 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
584 dbref3.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"), map2));
585 x06707.addDBRef(dbref3);
586 DBRefEntry dbref4 = new DBRefEntry("UNIPROT", "0", "P0CE20");
587 MapList map3 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
589 dbref4.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"), map3));
590 x06707.addDBRef(dbref4);
593 * M19487 with mapping to P0CE19 and Q46432
595 final SequenceI m19487 = new Sequence("EMBL|M19487", "AAACCCTTTGGG");
596 DBRefEntry dbref5 = new DBRefEntry("UNIPROT", "0", "P0CE19");
597 dbref5.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
598 new MapList(mapList)));
599 m19487.addDBRef(dbref5);
600 DBRefEntry dbref6 = new DBRefEntry("UNIPROT", "0", "Q46432");
601 dbref6.setMap(new Mapping(new Sequence("UNIPROT|Q46432", "KPFG"),
602 new MapList(mapList)));
603 m19487.addDBRef(dbref6);
606 * X07547 with mapping to P0CE20 and B0BCM4
608 final SequenceI x07547 = new Sequence("EMBL|X07547", "cccAAACCCTTTGGG");
609 DBRefEntry dbref7 = new DBRefEntry("UNIPROT", "0", "P0CE20");
610 dbref7.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
612 x07547.addDBRef(dbref7);
613 DBRefEntry dbref8 = new DBRefEntry("UNIPROT", "0", "B0BCM4");
614 dbref8.setMap(new Mapping(new Sequence("UNIPROT|B0BCM4", "KPFG"),
616 x07547.addDBRef(dbref8);
619 * mock sequence fetcher to 'return' the EMBL sequences
620 * TODO: Mockito would allow .thenReturn().thenReturn() here,
621 * and also capture and verification of the parameters
622 * passed in calls to getSequences() - important to verify that
623 * duplicate sequence fetches are not requested
625 SequenceFetcher mockFetcher = new SequenceFetcher(false)
629 public boolean isFetchable(String source)
634 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
638 assertEquals("Expected 3 embl seqs in first fetch", 3, refs.size());
639 return new SequenceI[] { j03321, x06707, m19487 };
641 assertEquals("Expected 1 embl seq in second fetch", 1, refs.size());
642 return new SequenceI[] { x07547 };
646 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
649 * find EMBL xrefs for Uniprot seqs and verify that
650 * - the EMBL xref'd sequences are retrieved without duplicates
651 * - mappings are added to the Uniprot dbrefs
652 * - mappings in the EMBL-to-Uniprot dbrefs are updated to the
653 * alignment sequences
654 * - dbrefs on the EMBL sequences are added to the original dbrefs
656 SequenceI[] seqs = new SequenceI[] { p0ce19, p0ce20 };
657 AlignmentI al = new Alignment(seqs);
658 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("EMBL",
662 * verify retrieved sequences
664 assertNotNull(xrefs);
665 assertEquals(4, xrefs.getHeight());
666 assertSame(j03321, xrefs.getSequenceAt(0));
667 assertSame(x06707, xrefs.getSequenceAt(1));
668 assertSame(m19487, xrefs.getSequenceAt(2));
669 assertSame(x07547, xrefs.getSequenceAt(3));
672 * verify mappings added to Uniprot-to-EMBL dbrefs
674 Mapping mapping = p0ce19.getDBRefs()[0].getMap();
675 assertSame(j03321, mapping.getTo());
676 mapping = p0ce19.getDBRefs()[1].getMap();
677 assertSame(x06707, mapping.getTo());
678 mapping = p0ce20.getDBRefs()[0].getMap();
679 assertSame(j03321, mapping.getTo());
680 mapping = p0ce20.getDBRefs()[1].getMap();
681 assertSame(x06707, mapping.getTo());
684 * verify dbrefs on EMBL are mapped to alignment seqs
686 assertSame(p0ce19, j03321.getDBRefs()[0].getMap().getTo());
687 assertSame(p0ce20, j03321.getDBRefs()[1].getMap().getTo());
688 assertSame(p0ce19, x06707.getDBRefs()[0].getMap().getTo());
689 assertSame(p0ce20, x06707.getDBRefs()[1].getMap().getTo());
692 * verify new dbref on EMBL dbref mapping is copied to the
693 * original Uniprot sequence
695 assertEquals(4, p0ce19.getDBRefs().length);
696 assertEquals("PIR", p0ce19.getDBRefs()[3].getSource());
697 assertEquals("S01875", p0ce19.getDBRefs()[3].getAccessionId());
700 @Test(groups = "Functional")
701 public void testSameSequence()
703 assertTrue(CrossRef.sameSequence(null, null));
704 SequenceI seq1 = new Sequence("seq1", "ABCDEF");
705 assertFalse(CrossRef.sameSequence(seq1, null));
706 assertFalse(CrossRef.sameSequence(null, seq1));
707 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "ABCDEF")));
708 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "abcdef")));
710 .sameSequence(seq1, new Sequence("seq2", "ABCDE-F")));
711 assertFalse(CrossRef.sameSequence(seq1, new Sequence("seq2", "BCDEF")));