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.util.DBRefUtils;
41 import jalview.util.MapList;
42 import jalview.ws.SequenceFetcher;
43 import jalview.ws.SequenceFetcherFactory;
45 import java.util.ArrayList;
46 import java.util.List;
48 import org.testng.annotations.AfterClass;
49 import org.testng.annotations.Test;
51 public class CrossRefTest
53 @Test(groups = { "Functional" })
54 public void testFindXDbRefs()
56 DBRefEntry ref1 = new DBRefEntry("UNIPROT", "1", "A123");
57 DBRefEntry ref2 = new DBRefEntry("UNIPROTKB/TREMBL", "1", "A123");
58 DBRefEntry ref3 = new DBRefEntry("pdb", "1", "A123");
59 DBRefEntry ref4 = new DBRefEntry("EMBLCDSPROTEIN", "1", "A123");
60 DBRefEntry ref5 = new DBRefEntry("embl", "1", "A123");
61 DBRefEntry ref6 = new DBRefEntry("emblCDS", "1", "A123");
62 DBRefEntry ref7 = new DBRefEntry("GeneDB", "1", "A123");
63 DBRefEntry ref8 = new DBRefEntry("PFAM", "1", "A123");
64 // ENSEMBL is a source of either dna or protein sequence data
65 DBRefEntry ref9 = new DBRefEntry("ENSEMBL", "1", "A123");
66 DBRefEntry[] refs = new DBRefEntry[] { ref1, ref2, ref3, ref4, ref5,
67 ref6, ref7, ref8, ref9 };
72 DBRefEntry[] found = DBRefUtils.selectDbRefs(true, refs);
73 assertEquals(4, found.length);
74 assertSame(ref5, found[0]);
75 assertSame(ref6, found[1]);
76 assertSame(ref7, found[2]);
77 assertSame(ref9, found[3]);
80 * Just the protein refs:
82 found = DBRefUtils.selectDbRefs(false, refs);
83 assertEquals(4, found.length);
84 assertSame(ref1, found[0]);
85 assertSame(ref2, found[1]);
86 assertSame(ref4, found[2]);
87 assertSame(ref9, found[3]);
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 = true)
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]", sources.toString());
128 * add a sequence to the alignment which has a dbref to UNIPROT|A1234
129 * and others to dna coding databases
133 seq.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
134 seq.addDBRef(new DBRefEntry("EMBLCDS", "0", "E2347"));
135 SequenceI seq2 = new Sequence("Seq2", "MGKYQARLSS");
136 seq2.addDBRef(new DBRefEntry("UNIPROT", "0", "A1234"));
137 seq2.addDBRef(new DBRefEntry("EMBL", "0", "E2345"));
138 seq2.addDBRef(new DBRefEntry("GENEDB", "0", "E2348"));
139 // TODO include ENSEMBLGENOMES in DBRefSource.DNACODINGDBS ?
140 al.addSequence(seq2);
141 sources = new CrossRef(new SequenceI[] { seq, seq2 }, al)
142 .findXrefSourcesForSequences(false);
143 assertEquals(3, sources.size());
144 assertEquals("[EMBLCDS, EMBL, GENEDB]", sources.toString());
148 * Test for finding 'product' sequences for the case where only an indirect
149 * xref is found - not on the nucleotide sequence but on a peptide sequence in
150 * the alignment which which it shares a nucleotide dbref
152 @Test(groups = { "Functional" }, enabled = true)
153 public void testFindXrefSequences_indirectDbrefToProtein()
157 * - nucleotide dbref EMBL|AF039662
158 * - peptide dbrefs EMBL|AF039662, UNIPROT|Q9ZTS2
160 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
161 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
162 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
163 uniprotSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
164 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
167 * Find UNIPROT xrefs for nucleotide
168 * - it has no UNIPROT dbref of its own
169 * - but peptide with matching nucleotide dbref does, so is returned
171 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
172 Alignment xrefs = new CrossRef(new SequenceI[] { emblSeq }, al)
173 .findXrefSequences("UNIPROT", true);
174 assertEquals(1, xrefs.getHeight());
175 assertSame(uniprotSeq, xrefs.getSequenceAt(0));
179 * Test for finding 'product' sequences for the case where only an indirect
180 * xref is found - not on the peptide sequence but on a nucleotide sequence in
181 * the alignment which which it shares a protein dbref
183 @Test(groups = { "Functional" }, enabled = true)
184 public void testFindXrefSequences_indirectDbrefToNucleotide()
188 * - peptide dbref UNIPROT|Q9ZTS2
189 * - nucleotide dbref EMBL|AF039662, UNIPROT|Q9ZTS2
191 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
192 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
193 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
194 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
195 emblSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
198 * find EMBL xrefs for peptide sequence - it has no direct
199 * dbrefs, but the 'corresponding' nucleotide sequence does, so is returned
202 * Find EMBL xrefs for peptide
203 * - it has no EMBL dbref of its own
204 * - but nucleotide with matching peptide dbref does, so is returned
206 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
207 Alignment xrefs = new CrossRef(new SequenceI[] { uniprotSeq }, al)
208 .findXrefSequences("EMBL", false);
209 assertEquals(1, xrefs.getHeight());
210 assertSame(emblSeq, xrefs.getSequenceAt(0));
214 * Test for finding 'product' sequences for the case where the selected
215 * sequence has no dbref to the desired source, and there are no indirect
216 * references via another sequence in the alignment
218 @Test(groups = { "Functional" })
219 public void testFindXrefSequences_noDbrefs()
222 * two nucleotide sequences, one with UNIPROT dbref
224 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
225 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
226 SequenceI dna2 = new Sequence("AJ307031", "AAACCCTTT");
229 * find UNIPROT xrefs for peptide sequence - it has no direct
230 * dbrefs, and the other sequence (which has a UNIPROT dbref) is not
231 * equatable to it, so no results found
233 AlignmentI al = new Alignment(new SequenceI[] { dna1, dna2 });
234 Alignment xrefs = new CrossRef(new SequenceI[] { dna2 }, al)
235 .findXrefSequences("UNIPROT", true);
240 * Tests for the method that searches an alignment (with one sequence
241 * excluded) for protein/nucleotide sequences with a given cross-reference
243 @Test(groups = { "Functional" }, enabled = true)
244 public void testSearchDataset()
247 * nucleotide sequence with UNIPROT AND EMBL dbref
248 * peptide sequence with UNIPROT dbref
250 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
251 Mapping map = new Mapping(new Sequence("pep2", "MLAVSRG"), new MapList(
252 new int[] { 1, 21 }, new int[] {
254 DBRefEntry dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
255 dna1.addDBRef(dbref);
256 dna1.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
257 SequenceI pep1 = new Sequence("Q9ZTS2", "MLAVSRGQ");
258 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
259 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
260 AlignmentI al = new Alignment(new SequenceI[] { dna1, pep1 });
262 List<SequenceI> result = new ArrayList<SequenceI>();
265 * first search for a dbref nowhere on the alignment:
267 dbref = new DBRefEntry("UNIPROT", "0", "P30419");
268 CrossRef testee = new CrossRef(al.getSequencesArray(), al);
269 AlignedCodonFrame acf = new AlignedCodonFrame();
270 boolean found = testee.searchDataset(true, dna1, dbref, result, acf,
273 assertTrue(result.isEmpty());
274 assertTrue(acf.isEmpty());
277 * search for a protein sequence with dbref UNIPROT:Q9ZTS2
279 acf = new AlignedCodonFrame();
280 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
281 found = testee.searchDataset(!dna1.isProtein(), dna1, dbref, result,
282 acf, false); // search dataset with a protein xref from a dna
283 // sequence to locate the protein product
285 assertEquals(1, result.size());
286 assertSame(pep1, result.get(0));
287 assertTrue(acf.isEmpty());
290 * search for a nucleotide sequence with dbref UNIPROT:Q9ZTS2
293 acf = new AlignedCodonFrame();
294 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
295 found = testee.searchDataset(!pep1.isProtein(), pep1, dbref, result,
296 acf, false); // search dataset with a protein's direct dbref to
297 // locate dna sequences with matching xref
299 assertEquals(1, result.size());
300 assertSame(dna1, result.get(0));
301 // should now have a mapping from dna to pep1
302 List<SequenceToSequenceMapping> mappings = acf.getMappings();
303 assertEquals(1, mappings.size());
304 SequenceToSequenceMapping mapping = mappings.get(0);
305 assertSame(dna1, mapping.getFromSeq());
306 assertSame(pep1, mapping.getMapping().getTo());
307 MapList mapList = mapping.getMapping().getMap();
308 assertEquals(1, mapList.getToRatio());
309 assertEquals(3, mapList.getFromRatio());
310 assertEquals(1, mapList.getFromRanges().size());
311 assertEquals(1, mapList.getFromRanges().get(0)[0]);
312 assertEquals(21, mapList.getFromRanges().get(0)[1]);
313 assertEquals(1, mapList.getToRanges().size());
314 assertEquals(1, mapList.getToRanges().get(0)[0]);
315 assertEquals(7, mapList.getToRanges().get(0)[1]);
319 * Test for finding 'product' sequences for the case where the selected
320 * sequence has a dbref with a mapping to a sequence. This represents the case
323 * <li>a fetched sequence is already decorated with its cross-reference (e.g.
324 * EMBL + translation), or</li>
325 * <li>Get Cross-References has been done once resulting in instantiated
326 * cross-reference mappings</li>
329 @Test(groups = { "Functional" })
330 public void testFindXrefSequences_fromDbRefMap()
333 * scenario: nucleotide sequence AF039662
334 * with dbref + mapping to Q9ZTS2 and P30419
335 * which themselves each have a dbref and feature
337 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
338 SequenceI pep1 = new Sequence("Q9ZTS2", "MALFQRSV");
339 SequenceI pep2 = new Sequence("P30419", "MTRRSQIF");
340 dna1.createDatasetSequence();
341 pep1.createDatasetSequence();
342 pep2.createDatasetSequence();
344 pep1.getDatasetSequence().addDBRef(
345 new DBRefEntry("Pfam", "0", "PF00111"));
346 pep1.addSequenceFeature(new SequenceFeature("type", "desc", 12, 14, 1f,
348 pep2.getDatasetSequence().addDBRef(new DBRefEntry("PDB", "0", "3JTK"));
349 pep2.addSequenceFeature(new SequenceFeature("type2", "desc2", 13, 15,
352 MapList mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 },
354 Mapping map = new Mapping(pep1, mapList);
355 DBRefEntry dbRef1 = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
356 dna1.getDatasetSequence().addDBRef(dbRef1);
357 mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 }, 3, 1);
358 map = new Mapping(pep2, mapList);
359 DBRefEntry dbRef2 = new DBRefEntry("UNIPROT", "0", "P30419", map);
360 dna1.getDatasetSequence().addDBRef(dbRef2);
363 * find UNIPROT xrefs for nucleotide sequence - it should pick up
366 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
367 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
368 .findXrefSequences("UNIPROT", true);
369 assertEquals(2, xrefs.getHeight());
372 * cross-refs alignment holds copies of the mapped sequences
373 * including copies of their dbrefs and features
375 checkCopySequence(pep1, xrefs.getSequenceAt(0));
376 checkCopySequence(pep2, xrefs.getSequenceAt(1));
380 * Helper method that verifies that 'copy' has the same name, start, end,
381 * sequence and dataset sequence object as 'original' (but is not the same
387 private void checkCopySequence(SequenceI copy, SequenceI original)
389 assertNotSame(copy, original);
390 assertSame(copy.getDatasetSequence(), original.getDatasetSequence());
391 assertEquals(copy.getName(), original.getName());
392 assertEquals(copy.getStart(), original.getStart());
393 assertEquals(copy.getEnd(), original.getEnd());
394 assertEquals(copy.getSequenceAsString(), original.getSequenceAsString());
398 * Test for finding 'product' sequences for the case where the selected
399 * sequence has a dbref with no mapping, triggering a fetch from database
401 @Test(groups = { "Functional" })
402 public void testFindXrefSequences_withFetch()
404 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
405 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
406 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "P30419"));
407 dna1.addDBRef(new DBRefEntry("UNIPROT", "0", "P00314"));
408 final SequenceI pep1 = new Sequence("Q9ZTS2", "MYQLIRSSW");
409 final SequenceI pep2 = new Sequence("P00314", "MRKLLAASG");
412 * argument false suppresses adding DAS sources
413 * todo: define an interface type SequenceFetcherI and mock that
415 SequenceFetcher mockFetcher = new SequenceFetcher(false)
418 public boolean isFetchable(String source)
424 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
426 return new SequenceI[] { pep1, pep2 };
429 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
432 * find UNIPROT xrefs for nucleotide sequence
434 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
435 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
436 .findXrefSequences("UNIPROT", true);
437 assertEquals(2, xrefs.getHeight());
438 assertSame(pep1, xrefs.getSequenceAt(0));
439 assertSame(pep2, xrefs.getSequenceAt(1));
443 public void tearDown()
445 SequenceFetcherFactory.setSequenceFetcher(null);
449 * Test for finding 'product' sequences for the case where both gene and
450 * transcript sequences have dbrefs to Uniprot.
452 @Test(groups = { "Functional" })
453 public void testFindXrefSequences_forGeneAndTranscripts()
458 SequenceI gene = new Sequence("ENSG00000157764", "CGCCTCCCTTCCCC");
459 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
460 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
463 * 'transcript' with CDS feature (supports mapping to protein)
465 SequenceI braf001 = new Sequence("ENST00000288602", "taagATGGCGGCGCTGa");
466 braf001.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
467 braf001.addSequenceFeature(new SequenceFeature("CDS", "", 5, 16, 0f,
471 * 'spliced transcript' with CDS ranges
473 SequenceI braf002 = new Sequence("ENST00000497784", "gCAGGCtaTCTGTTCaa");
474 braf002.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
475 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 2, 6, 0f,
477 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 9, 15, 0f,
481 * TODO code is fragile - use of SequenceIdMatcher depends on fetched
482 * sequences having a name starting Source|Accession
483 * which happens to be true for Uniprot,PDB,EMBL but not Pfam,Rfam,Ensembl
485 final SequenceI pep1 = new Sequence("UNIPROT|P15056", "MAAL");
486 final SequenceI pep2 = new Sequence("UNIPROT|H7C5K3", "QALF");
489 * argument false suppresses adding DAS sources
490 * todo: define an interface type SequenceFetcherI and mock that
492 SequenceFetcher mockFetcher = new SequenceFetcher(false)
495 public boolean isFetchable(String source)
501 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
503 return new SequenceI[] { pep1, pep2 };
506 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
509 * find UNIPROT xrefs for gene and transcripts
511 * - the two proteins are retrieved but not duplicated
512 * - mappings are built from transcript (CDS) to proteins
513 * - no mappings from gene to proteins
515 SequenceI[] seqs = new SequenceI[] { gene, braf001, braf002 };
516 AlignmentI al = new Alignment(seqs);
517 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("UNIPROT",
519 assertEquals(2, xrefs.getHeight());
520 assertSame(pep1, xrefs.getSequenceAt(0));
521 assertSame(pep2, xrefs.getSequenceAt(1));
526 * Test that emulates this (real but simplified) case:
528 * UNIPROT|P0CE19 EMBL|J03321, EMBL|X06707, EMBL|M19487
529 * UNIPROT|P0CE20 EMBL|J03321, EMBL|X06707, EMBL|X07547
530 * Find cross-references for EMBL. These are mocked here as
531 * EMBL|J03321 with mappings to P0CE18, P0CE19, P0CE20
532 * EMBL|X06707 with mappings to P0CE17, P0CE19, P0CE20
533 * EMBL|M19487 with mappings to P0CE19, Q46432
534 * EMBL|X07547 with mappings to P0CE20, B0BCM4
535 * EMBL sequences are first 'fetched' (mocked here) for P0CE19.
536 * The 3 EMBL sequences are added to the alignment dataset.
537 * Their dbrefs to Uniprot products P0CE19 and P0CE20 should be matched in the
538 * alignment dataset and updated to reference the original Uniprot sequences.
539 * For the second Uniprot sequence, the J03321 and X06707 xrefs should be
540 * resolved from the dataset, and only the X07547 dbref fetched.
541 * So the end state to verify is:
542 * - 4 cross-ref sequences returned: J03321, X06707, M19487, X07547
543 * - P0CE19/20 dbrefs to EMBL sequences now have mappings
544 * - J03321 dbrefs to P0CE19/20 mapped to original Uniprot sequences
545 * - X06707 dbrefs to P0CE19/20 mapped to original Uniprot sequences
548 @Test(groups = { "Functional" })
549 public void testFindXrefSequences_uniprotEmblManyToMany()
552 * Uniprot sequences, both with xrefs to EMBL|J03321
555 SequenceI p0ce19 = new Sequence("UNIPROT|P0CE19", "KPFG");
556 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
557 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
558 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "M19487"));
559 SequenceI p0ce20 = new Sequence("UNIPROT|P0CE20", "PFGK");
560 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
561 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
562 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X07547"));
565 * EMBL sequences to be 'fetched', complete with dbrefs and mappings
566 * to their protein products (CDS location and translations are provided
567 * in EMBL XML); these should be matched to, and replaced with,
568 * the corresponding uniprot sequences after fetching
572 * J03321 with mappings to P0CE19 and P0CE20
574 final SequenceI j03321 = new Sequence("EMBL|J03321", "AAACCCTTTGGGAAAA");
575 DBRefEntry dbref1 = new DBRefEntry("UNIPROT", "0", "P0CE19");
576 MapList mapList = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 },
578 Mapping map = new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
580 // add a dbref to the mapped to sequence - should get copied to p0ce19
581 map.getTo().addDBRef(new DBRefEntry("PIR", "0", "S01875"));
583 j03321.addDBRef(dbref1);
584 DBRefEntry dbref2 = new DBRefEntry("UNIPROT", "0", "P0CE20");
585 mapList = new MapList(new int[] { 4, 15 }, new int[] { 2, 5 }, 3, 1);
586 dbref2.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
587 new MapList(mapList)));
588 j03321.addDBRef(dbref2);
591 * X06707 with mappings to P0CE19 and P0CE20
593 final SequenceI x06707 = new Sequence("EMBL|X06707", "atgAAACCCTTTGGG");
594 DBRefEntry dbref3 = new DBRefEntry("UNIPROT", "0", "P0CE19");
595 MapList map2 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
597 dbref3.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"), map2));
598 x06707.addDBRef(dbref3);
599 DBRefEntry dbref4 = new DBRefEntry("UNIPROT", "0", "P0CE20");
600 MapList map3 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
602 dbref4.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"), map3));
603 x06707.addDBRef(dbref4);
606 * M19487 with mapping to P0CE19 and Q46432
608 final SequenceI m19487 = new Sequence("EMBL|M19487", "AAACCCTTTGGG");
609 DBRefEntry dbref5 = new DBRefEntry("UNIPROT", "0", "P0CE19");
610 dbref5.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
611 new MapList(mapList)));
612 m19487.addDBRef(dbref5);
613 DBRefEntry dbref6 = new DBRefEntry("UNIPROT", "0", "Q46432");
614 dbref6.setMap(new Mapping(new Sequence("UNIPROT|Q46432", "KPFG"),
615 new MapList(mapList)));
616 m19487.addDBRef(dbref6);
619 * X07547 with mapping to P0CE20 and B0BCM4
621 final SequenceI x07547 = new Sequence("EMBL|X07547", "cccAAACCCTTTGGG");
622 DBRefEntry dbref7 = new DBRefEntry("UNIPROT", "0", "P0CE20");
623 dbref7.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
625 x07547.addDBRef(dbref7);
626 DBRefEntry dbref8 = new DBRefEntry("UNIPROT", "0", "B0BCM4");
627 dbref8.setMap(new Mapping(new Sequence("UNIPROT|B0BCM4", "KPFG"),
629 x07547.addDBRef(dbref8);
632 * mock sequence fetcher to 'return' the EMBL sequences
633 * TODO: Mockito would allow .thenReturn().thenReturn() here,
634 * and also capture and verification of the parameters
635 * passed in calls to getSequences() - important to verify that
636 * duplicate sequence fetches are not requested
638 SequenceFetcher mockFetcher = new SequenceFetcher(false)
643 public boolean isFetchable(String source)
649 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
654 assertEquals("Expected 3 embl seqs in first fetch", 3,
656 return new SequenceI[] { j03321, x06707, m19487 };
660 assertEquals("Expected 1 embl seq in second fetch", 1,
662 return new SequenceI[] { x07547 };
666 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
669 * find EMBL xrefs for Uniprot seqs and verify that
670 * - the EMBL xref'd sequences are retrieved without duplicates
671 * - mappings are added to the Uniprot dbrefs
672 * - mappings in the EMBL-to-Uniprot dbrefs are updated to the
673 * alignment sequences
674 * - dbrefs on the EMBL sequences are added to the original dbrefs
676 SequenceI[] seqs = new SequenceI[] { p0ce19, p0ce20 };
677 AlignmentI al = new Alignment(seqs);
678 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("EMBL",
682 * verify retrieved sequences
684 assertNotNull(xrefs);
685 assertEquals(4, xrefs.getHeight());
686 assertSame(j03321, xrefs.getSequenceAt(0));
687 assertSame(x06707, xrefs.getSequenceAt(1));
688 assertSame(m19487, xrefs.getSequenceAt(2));
689 assertSame(x07547, xrefs.getSequenceAt(3));
692 * verify mappings added to Uniprot-to-EMBL dbrefs
694 Mapping mapping = p0ce19.getDBRefs()[0].getMap();
695 assertSame(j03321, mapping.getTo());
696 mapping = p0ce19.getDBRefs()[1].getMap();
697 assertSame(x06707, mapping.getTo());
698 mapping = p0ce20.getDBRefs()[0].getMap();
699 assertSame(j03321, mapping.getTo());
700 mapping = p0ce20.getDBRefs()[1].getMap();
701 assertSame(x06707, mapping.getTo());
704 * verify dbrefs on EMBL are mapped to alignment seqs
706 assertSame(p0ce19, j03321.getDBRefs()[0].getMap().getTo());
707 assertSame(p0ce20, j03321.getDBRefs()[1].getMap().getTo());
708 assertSame(p0ce19, x06707.getDBRefs()[0].getMap().getTo());
709 assertSame(p0ce20, x06707.getDBRefs()[1].getMap().getTo());
712 * verify new dbref on EMBL dbref mapping is copied to the
713 * original Uniprot sequence
715 assertEquals(4, p0ce19.getDBRefs().length);
716 assertEquals("PIR", p0ce19.getDBRefs()[3].getSource());
717 assertEquals("S01875", p0ce19.getDBRefs()[3].getAccessionId());
720 @Test(groups = "Functional")
721 public void testSameSequence()
723 assertTrue(CrossRef.sameSequence(null, null));
724 SequenceI seq1 = new Sequence("seq1", "ABCDEF");
725 assertFalse(CrossRef.sameSequence(seq1, null));
726 assertFalse(CrossRef.sameSequence(null, seq1));
727 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "ABCDEF")));
728 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "abcdef")));
730 .sameSequence(seq1, new Sequence("seq2", "ABCDE-F")));
731 assertFalse(CrossRef.sameSequence(seq1, new Sequence("seq2", "BCDEF")));