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 // method is patched to remove EMBL from the sources to match
125 assertEquals(3, sources.size());
126 assertEquals("[EMBLCDS, GENEDB, ENSEMBL]", sources.toString());
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, seq2 }, al)
143 .findXrefSourcesForSequences(false);
144 // method removed EMBL from sources to match
145 assertEquals(2, sources.size());
146 assertEquals("[EMBLCDS, GENEDB]", sources.toString());
150 * Test for finding 'product' sequences for the case where only an indirect
151 * xref is found - not on the nucleotide sequence but on a peptide sequence in
152 * the alignment which which it shares a nucleotide dbref
154 @Test(groups = { "Functional" }, enabled = true)
155 public void testFindXrefSequences_indirectDbrefToProtein()
159 * - nucleotide dbref EMBL|AF039662
160 * - peptide dbrefs EMBL|AF039662, UNIPROT|Q9ZTS2
162 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
163 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
164 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
165 uniprotSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
166 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
169 * Find UNIPROT xrefs for nucleotide
170 * - it has no UNIPROT dbref of its own
171 * - but peptide with matching nucleotide dbref does, so is returned
173 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
174 Alignment xrefs = new CrossRef(new SequenceI[] { emblSeq }, al)
175 .findXrefSequences("UNIPROT", true);
176 assertEquals(1, xrefs.getHeight());
177 assertSame(uniprotSeq, xrefs.getSequenceAt(0));
181 * Test for finding 'product' sequences for the case where only an indirect
182 * xref is found - not on the peptide sequence but on a nucleotide sequence in
183 * the alignment which which it shares a protein dbref
185 @Test(groups = { "Functional" }, enabled = true)
186 public void testFindXrefSequences_indirectDbrefToNucleotide()
190 * - peptide dbref UNIPROT|Q9ZTS2
191 * - nucleotide dbref EMBL|AF039662, UNIPROT|Q9ZTS2
193 SequenceI uniprotSeq = new Sequence("Q9ZTS2", "MASVSATMISTS");
194 uniprotSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
195 SequenceI emblSeq = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
196 emblSeq.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
197 emblSeq.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
200 * find EMBL xrefs for peptide sequence - it has no direct
201 * dbrefs, but the 'corresponding' nucleotide sequence does, so is returned
204 * Find EMBL xrefs for peptide
205 * - it has no EMBL dbref of its own
206 * - but nucleotide with matching peptide dbref does, so is returned
208 AlignmentI al = new Alignment(new SequenceI[] { emblSeq, uniprotSeq });
209 Alignment xrefs = new CrossRef(new SequenceI[] { uniprotSeq }, al)
210 .findXrefSequences("EMBL", false);
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 = true)
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 Mapping map = new Mapping(new Sequence("pep2", "MLAVSRG"), new MapList(
254 new int[] { 1, 21 }, new int[] { 1, 7 }, 3, 1));
255 DBRefEntry dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
256 dna1.addDBRef(dbref);
257 dna1.addDBRef(new DBRefEntry("EMBL", "0", "AF039662"));
258 SequenceI pep1 = new Sequence("Q9ZTS2", "MLAVSRGQ");
259 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
260 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
261 AlignmentI al = new Alignment(new SequenceI[] { dna1, pep1 });
263 List<SequenceI> result = new ArrayList<SequenceI>();
266 * first search for a dbref nowhere on the alignment:
268 dbref = new DBRefEntry("UNIPROT", "0", "P30419");
269 CrossRef testee = new CrossRef(al.getSequencesArray(), al);
270 AlignedCodonFrame acf = new AlignedCodonFrame();
271 boolean found = testee.searchDataset(true, dna1, dbref, result, acf,
274 assertTrue(result.isEmpty());
275 assertTrue(acf.isEmpty());
278 * search for a protein sequence with dbref UNIPROT:Q9ZTS2
280 acf = new AlignedCodonFrame();
281 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
282 found = testee.searchDataset(!dna1.isProtein(), dna1, dbref, result,
283 acf, false); // search dataset with a protein xref from a dna
284 // sequence to locate the protein product
286 assertEquals(1, result.size());
287 assertSame(pep1, result.get(0));
288 assertTrue(acf.isEmpty());
291 * search for a nucleotide sequence with dbref UNIPROT:Q9ZTS2
294 acf = new AlignedCodonFrame();
295 dbref = new DBRefEntry("UNIPROT", "0", "Q9ZTS2");
296 found = testee.searchDataset(!pep1.isProtein(), pep1, dbref, result,
297 acf, false); // search dataset with a protein's direct dbref to
298 // locate dna sequences with matching xref
300 assertEquals(1, result.size());
301 assertSame(dna1, result.get(0));
302 // should now have a mapping from dna to pep1
303 List<SequenceToSequenceMapping> mappings = acf.getMappings();
304 assertEquals(1, mappings.size());
305 SequenceToSequenceMapping mapping = mappings.get(0);
306 assertSame(dna1, mapping.getFromSeq());
307 assertSame(pep1, mapping.getMapping().getTo());
308 MapList mapList = mapping.getMapping().getMap();
309 assertEquals(1, mapList.getToRatio());
310 assertEquals(3, mapList.getFromRatio());
311 assertEquals(1, mapList.getFromRanges().size());
312 assertEquals(1, mapList.getFromRanges().get(0)[0]);
313 assertEquals(21, mapList.getFromRanges().get(0)[1]);
314 assertEquals(1, mapList.getToRanges().size());
315 assertEquals(1, mapList.getToRanges().get(0)[0]);
316 assertEquals(7, mapList.getToRanges().get(0)[1]);
320 * Test for finding 'product' sequences for the case where the selected
321 * sequence has a dbref with a mapping to a sequence. This represents the case
324 * <li>a fetched sequence is already decorated with its cross-reference (e.g.
325 * EMBL + translation), or</li>
326 * <li>Get Cross-References has been done once resulting in instantiated
327 * cross-reference mappings</li>
330 @Test(groups = { "Functional" })
331 public void testFindXrefSequences_fromDbRefMap()
334 * scenario: nucleotide sequence AF039662
335 * with dbref + mapping to Q9ZTS2 and P30419
336 * which themselves each have a dbref and feature
338 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
339 SequenceI pep1 = new Sequence("Q9ZTS2", "MALFQRSV");
340 SequenceI pep2 = new Sequence("P30419", "MTRRSQIF");
341 dna1.createDatasetSequence();
342 pep1.createDatasetSequence();
343 pep2.createDatasetSequence();
345 pep1.getDatasetSequence().addDBRef(
346 new DBRefEntry("Pfam", "0", "PF00111"));
347 pep1.addSequenceFeature(new SequenceFeature("type", "desc", 12, 14, 1f,
349 pep2.getDatasetSequence().addDBRef(new DBRefEntry("PDB", "0", "3JTK"));
350 pep2.addSequenceFeature(new SequenceFeature("type2", "desc2", 13, 15,
353 MapList mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 },
355 Mapping map = new Mapping(pep1, mapList);
356 DBRefEntry dbRef1 = new DBRefEntry("UNIPROT", "0", "Q9ZTS2", map);
357 dna1.getDatasetSequence().addDBRef(dbRef1);
358 mapList = new MapList(new int[] { 1, 24 }, new int[] { 1, 3 }, 3, 1);
359 map = new Mapping(pep2, mapList);
360 DBRefEntry dbRef2 = new DBRefEntry("UNIPROT", "0", "P30419", map);
361 dna1.getDatasetSequence().addDBRef(dbRef2);
364 * find UNIPROT xrefs for nucleotide sequence - it should pick up
367 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
368 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
369 .findXrefSequences("UNIPROT", true);
370 assertEquals(2, xrefs.getHeight());
373 * cross-refs alignment holds copies of the mapped sequences
374 * including copies of their dbrefs and features
376 checkCopySequence(pep1, xrefs.getSequenceAt(0));
377 checkCopySequence(pep2, xrefs.getSequenceAt(1));
381 * Helper method that verifies that 'copy' has the same name, start, end,
382 * sequence and dataset sequence object as 'original' (but is not the same
388 private void checkCopySequence(SequenceI copy, SequenceI original)
390 assertNotSame(copy, original);
391 assertSame(copy.getDatasetSequence(), original.getDatasetSequence());
392 assertEquals(copy.getName(), original.getName());
393 assertEquals(copy.getStart(), original.getStart());
394 assertEquals(copy.getEnd(), original.getEnd());
395 assertEquals(copy.getSequenceAsString(), original.getSequenceAsString());
399 * Test for finding 'product' sequences for the case where the selected
400 * sequence has a dbref with no mapping, triggering a fetch from database
402 @Test(groups = { "Functional" })
403 public void testFindXrefSequences_withFetch()
405 SequenceI dna1 = new Sequence("AF039662", "GGGGCAGCACAAGAAC");
406 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "Q9ZTS2"));
407 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "P30419"));
408 dna1.addDBRef(new DBRefEntry("UNIPROT", "ENA:0", "P00314"));
409 final SequenceI pep1 = new Sequence("Q9ZTS2", "MYQLIRSSW");
410 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "Q9ZTS2"));
412 final SequenceI pep2 = new Sequence("P00314", "MRKLLAASG");
413 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "P00314"));
416 * argument false suppresses adding DAS sources
417 * todo: define an interface type SequenceFetcherI and mock that
419 SequenceFetcher mockFetcher = new SequenceFetcher(false)
422 public boolean isFetchable(String source)
428 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
430 return new SequenceI[] { pep1, pep2 };
433 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
436 * find UNIPROT xrefs for nucleotide sequence
438 AlignmentI al = new Alignment(new SequenceI[] { dna1 });
439 Alignment xrefs = new CrossRef(new SequenceI[] { dna1 }, al)
440 .findXrefSequences("UNIPROT", true);
441 assertEquals(2, xrefs.getHeight());
442 assertSame(pep1, xrefs.getSequenceAt(0));
443 assertSame(pep2, xrefs.getSequenceAt(1));
447 public void tearDown()
449 SequenceFetcherFactory.setSequenceFetcher(null);
453 * Test for finding 'product' sequences for the case where both gene and
454 * transcript sequences have dbrefs to Uniprot.
456 @Test(groups = { "Functional" })
457 public void testFindXrefSequences_forGeneAndTranscripts()
462 SequenceI gene = new Sequence("ENSG00000157764", "CGCCTCCCTTCCCC");
463 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
464 gene.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
467 * 'transcript' with CDS feature (supports mapping to protein)
469 SequenceI braf001 = new Sequence("ENST00000288602", "taagATGGCGGCGCTGa");
470 braf001.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
471 braf001.addSequenceFeature(new SequenceFeature("CDS", "", 5, 16, 0f,
475 * 'spliced transcript' with CDS ranges
477 SequenceI braf002 = new Sequence("ENST00000497784", "gCAGGCtaTCTGTTCaa");
478 braf002.addDBRef(new DBRefEntry("UNIPROT", "ENSEMBL|0", "H7C5K3"));
479 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 2, 6, 0f,
481 braf002.addSequenceFeature(new SequenceFeature("CDS", "", 9, 15, 0f,
485 * TODO code is fragile - use of SequenceIdMatcher depends on fetched
486 * sequences having a name starting Source|Accession
487 * which happens to be true for Uniprot,PDB,EMBL but not Pfam,Rfam,Ensembl
489 final SequenceI pep1 = new Sequence("UNIPROT|P15056", "MAAL");
490 pep1.addDBRef(new DBRefEntry("UNIPROT", "0", "P15056"));
491 final SequenceI pep2 = new Sequence("UNIPROT|H7C5K3", "QALF");
492 pep2.addDBRef(new DBRefEntry("UNIPROT", "0", "H7C5K3"));
494 * argument false suppresses adding DAS sources
495 * todo: define an interface type SequenceFetcherI and mock that
497 SequenceFetcher mockFetcher = new SequenceFetcher(false)
500 public boolean isFetchable(String source)
506 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
508 return new SequenceI[] { pep1, pep2 };
511 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
514 * find UNIPROT xrefs for gene and transcripts
516 * - the two proteins are retrieved but not duplicated
517 * - mappings are built from transcript (CDS) to proteins
518 * - no mappings from gene to proteins
520 SequenceI[] seqs = new SequenceI[] { gene, braf001, braf002 };
521 AlignmentI al = new Alignment(seqs);
522 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("UNIPROT",
524 assertEquals(2, xrefs.getHeight());
525 assertSame(pep1, xrefs.getSequenceAt(0));
526 assertSame(pep2, xrefs.getSequenceAt(1));
531 * Test that emulates this (real but simplified) case:
533 * UNIPROT|P0CE19 EMBL|J03321, EMBL|X06707, EMBL|M19487
534 * UNIPROT|P0CE20 EMBL|J03321, EMBL|X06707, EMBL|X07547
535 * Find cross-references for EMBL. These are mocked here as
536 * EMBL|J03321 with mappings to P0CE18, P0CE19, P0CE20
537 * EMBL|X06707 with mappings to P0CE17, P0CE19, P0CE20
538 * EMBL|M19487 with mappings to P0CE19, Q46432
539 * EMBL|X07547 with mappings to P0CE20, B0BCM4
540 * EMBL sequences are first 'fetched' (mocked here) for P0CE19.
541 * The 3 EMBL sequences are added to the alignment dataset.
542 * Their dbrefs to Uniprot products P0CE19 and P0CE20 should be matched in the
543 * alignment dataset and updated to reference the original Uniprot sequences.
544 * For the second Uniprot sequence, the J03321 and X06707 xrefs should be
545 * resolved from the dataset, and only the X07547 dbref fetched.
546 * So the end state to verify is:
547 * - 4 cross-ref sequences returned: J03321, X06707, M19487, X07547
548 * - P0CE19/20 dbrefs to EMBL sequences now have mappings
549 * - J03321 dbrefs to P0CE19/20 mapped to original Uniprot sequences
550 * - X06707 dbrefs to P0CE19/20 mapped to original Uniprot sequences
553 @Test(groups = { "Functional" })
554 public void testFindXrefSequences_uniprotEmblManyToMany()
557 * Uniprot sequences, both with xrefs to EMBL|J03321
560 SequenceI p0ce19 = new Sequence("UNIPROT|P0CE19", "KPFG");
561 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
562 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
563 p0ce19.addDBRef(new DBRefEntry("EMBL", "0", "M19487"));
564 SequenceI p0ce20 = new Sequence("UNIPROT|P0CE20", "PFGK");
565 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "J03321"));
566 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X06707"));
567 p0ce20.addDBRef(new DBRefEntry("EMBL", "0", "X07547"));
570 * EMBL sequences to be 'fetched', complete with dbrefs and mappings
571 * to their protein products (CDS location and translations are provided
572 * in EMBL XML); these should be matched to, and replaced with,
573 * the corresponding uniprot sequences after fetching
577 * J03321 with mappings to P0CE19 and P0CE20
579 final SequenceI j03321 = new Sequence("EMBL|J03321", "AAACCCTTTGGGAAAA");
580 DBRefEntry dbref1 = new DBRefEntry("UNIPROT", "0", "P0CE19");
581 MapList mapList = new MapList(new int[] { 1, 12 }, new int[] { 1, 4 },
583 Mapping map = new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
585 // add a dbref to the mapped to sequence - should get copied to p0ce19
586 map.getTo().addDBRef(new DBRefEntry("PIR", "0", "S01875"));
588 j03321.addDBRef(dbref1);
589 DBRefEntry dbref2 = new DBRefEntry("UNIPROT", "0", "P0CE20");
590 mapList = new MapList(new int[] { 4, 15 }, new int[] { 2, 5 }, 3, 1);
591 dbref2.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
592 new MapList(mapList)));
593 j03321.addDBRef(dbref2);
596 * X06707 with mappings to P0CE19 and P0CE20
598 final SequenceI x06707 = new Sequence("EMBL|X06707", "atgAAACCCTTTGGG");
599 DBRefEntry dbref3 = new DBRefEntry("UNIPROT", "0", "P0CE19");
600 MapList map2 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
602 dbref3.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"), map2));
603 x06707.addDBRef(dbref3);
604 DBRefEntry dbref4 = new DBRefEntry("UNIPROT", "0", "P0CE20");
605 MapList map3 = new MapList(new int[] { 4, 15 }, new int[] { 1, 4 }, 3,
607 dbref4.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"), map3));
608 x06707.addDBRef(dbref4);
611 * M19487 with mapping to P0CE19 and Q46432
613 final SequenceI m19487 = new Sequence("EMBL|M19487", "AAACCCTTTGGG");
614 DBRefEntry dbref5 = new DBRefEntry("UNIPROT", "0", "P0CE19");
615 dbref5.setMap(new Mapping(new Sequence("UNIPROT|P0CE19", "KPFG"),
616 new MapList(mapList)));
617 m19487.addDBRef(dbref5);
618 DBRefEntry dbref6 = new DBRefEntry("UNIPROT", "0", "Q46432");
619 dbref6.setMap(new Mapping(new Sequence("UNIPROT|Q46432", "KPFG"),
620 new MapList(mapList)));
621 m19487.addDBRef(dbref6);
624 * X07547 with mapping to P0CE20 and B0BCM4
626 final SequenceI x07547 = new Sequence("EMBL|X07547", "cccAAACCCTTTGGG");
627 DBRefEntry dbref7 = new DBRefEntry("UNIPROT", "0", "P0CE20");
628 dbref7.setMap(new Mapping(new Sequence("UNIPROT|P0CE20", "PFGK"),
630 x07547.addDBRef(dbref7);
631 DBRefEntry dbref8 = new DBRefEntry("UNIPROT", "0", "B0BCM4");
632 dbref8.setMap(new Mapping(new Sequence("UNIPROT|B0BCM4", "KPFG"),
634 x07547.addDBRef(dbref8);
637 * mock sequence fetcher to 'return' the EMBL sequences
638 * TODO: Mockito would allow .thenReturn().thenReturn() here,
639 * and also capture and verification of the parameters
640 * passed in calls to getSequences() - important to verify that
641 * duplicate sequence fetches are not requested
643 SequenceFetcher mockFetcher = new SequenceFetcher(false)
648 public boolean isFetchable(String source)
654 public SequenceI[] getSequences(List<DBRefEntry> refs, boolean dna)
659 assertEquals("Expected 3 embl seqs in first fetch", 3,
661 return new SequenceI[] { j03321, x06707, m19487 };
665 assertEquals("Expected 1 embl seq in second fetch", 1,
667 return new SequenceI[] { x07547 };
671 SequenceFetcherFactory.setSequenceFetcher(mockFetcher);
674 * find EMBL xrefs for Uniprot seqs and verify that
675 * - the EMBL xref'd sequences are retrieved without duplicates
676 * - mappings are added to the Uniprot dbrefs
677 * - mappings in the EMBL-to-Uniprot dbrefs are updated to the
678 * alignment sequences
679 * - dbrefs on the EMBL sequences are added to the original dbrefs
681 SequenceI[] seqs = new SequenceI[] { p0ce19, p0ce20 };
682 AlignmentI al = new Alignment(seqs);
683 Alignment xrefs = new CrossRef(seqs, al).findXrefSequences("EMBL",
687 * verify retrieved sequences
689 assertNotNull(xrefs);
690 assertEquals(4, xrefs.getHeight());
691 assertSame(j03321, xrefs.getSequenceAt(0));
692 assertSame(x06707, xrefs.getSequenceAt(1));
693 assertSame(m19487, xrefs.getSequenceAt(2));
694 assertSame(x07547, xrefs.getSequenceAt(3));
697 * verify mappings added to Uniprot-to-EMBL dbrefs
699 Mapping mapping = p0ce19.getDBRefs()[0].getMap();
700 assertSame(j03321, mapping.getTo());
701 mapping = p0ce19.getDBRefs()[1].getMap();
702 assertSame(x06707, mapping.getTo());
703 mapping = p0ce20.getDBRefs()[0].getMap();
704 assertSame(j03321, mapping.getTo());
705 mapping = p0ce20.getDBRefs()[1].getMap();
706 assertSame(x06707, mapping.getTo());
709 * verify dbrefs on EMBL are mapped to alignment seqs
711 assertSame(p0ce19, j03321.getDBRefs()[0].getMap().getTo());
712 assertSame(p0ce20, j03321.getDBRefs()[1].getMap().getTo());
713 assertSame(p0ce19, x06707.getDBRefs()[0].getMap().getTo());
714 assertSame(p0ce20, x06707.getDBRefs()[1].getMap().getTo());
717 * verify new dbref on EMBL dbref mapping is copied to the
718 * original Uniprot sequence
720 assertEquals(4, p0ce19.getDBRefs().length);
721 assertEquals("PIR", p0ce19.getDBRefs()[3].getSource());
722 assertEquals("S01875", p0ce19.getDBRefs()[3].getAccessionId());
725 @Test(groups = "Functional")
726 public void testSameSequence()
728 assertTrue(CrossRef.sameSequence(null, null));
729 SequenceI seq1 = new Sequence("seq1", "ABCDEF");
730 assertFalse(CrossRef.sameSequence(seq1, null));
731 assertFalse(CrossRef.sameSequence(null, seq1));
732 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "ABCDEF")));
733 assertTrue(CrossRef.sameSequence(seq1, new Sequence("seq2", "abcdef")));
735 .sameSequence(seq1, new Sequence("seq2", "ABCDE-F")));
736 assertFalse(CrossRef.sameSequence(seq1, new Sequence("seq2", "BCDEF")));