package jalview.analysis;
import static org.junit.Assert.assertEquals;
+import static org.junit.Assert.assertFalse;
+import static org.junit.Assert.assertSame;
import static org.junit.Assert.assertTrue;
import jalview.analysis.AlignmentUtils.MappingResult;
import jalview.datamodel.AlignedCodonFrame;
import java.io.IOException;
import java.util.Arrays;
+import java.util.Collections;
import java.util.List;
import java.util.Map;
/*
* Check two mappings (one for Mouse, one for Human)
*/
- assertEquals(2, protein.getCodonFrames().length);
- assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).length);
- assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).length);
+ assertEquals(2, protein.getCodonFrames().size());
+ assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
+ assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
/*
* Inspect mapping for Human protein
*/
- AlignedCodonFrame humanMapping = protein.getCodonFrame(protein
- .getSequenceAt(0))[0];
+ AlignedCodonFrame humanMapping = protein.getCodonFrame(
+ protein.getSequenceAt(0)).get(0);
assertEquals(1, humanMapping.getdnaSeqs().length);
assertEquals(cdna1.getSequenceAt(1).getDatasetSequence(),
humanMapping.getdnaSeqs()[0]);
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
assertTrue(Arrays.equals(new int[]
- { 1, 9 }, mapList.getFromRanges()));
+ { 1, 9 }, mapList.getFromRanges().get(0)));
+ assertEquals(1, mapList.getFromRanges().size());
assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges()));
+ { 1, 3 }, mapList.getToRanges().get(0)));
+ assertEquals(1, mapList.getToRanges().size());
/*
* Inspect mappings for Mouse protein
*/
- AlignedCodonFrame mouseMapping1 = protein.getCodonFrame(protein
- .getSequenceAt(1))[0];
+ AlignedCodonFrame mouseMapping1 = protein.getCodonFrame(
+ protein.getSequenceAt(1)).get(0);
assertEquals(2, mouseMapping1.getdnaSeqs().length);
assertEquals(cdna1.getSequenceAt(0).getDatasetSequence(),
mouseMapping1.getdnaSeqs()[0]);
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
assertTrue(Arrays.equals(new int[]
- { 1, 9 }, mapList.getFromRanges()));
+ { 1, 9 }, mapList.getFromRanges().get(0)));
+ assertEquals(1, mapList.getFromRanges().size());
assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges()));
+ { 1, 3 }, mapList.getToRanges().get(0)));
+ assertEquals(1, mapList.getToRanges().size());
}
}
/*
* Check two mappings (one for Mouse, one for Human)
*/
- assertEquals(2, protein.getCodonFrames().length);
- assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).length);
- assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).length);
+ assertEquals(2, protein.getCodonFrames().size());
+ assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
+ assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
/*
* Inspect mapping for Human protein - should map to 2nd and 4th cDNA seqs
*/
- AlignedCodonFrame humanMapping = protein.getCodonFrame(protein
- .getSequenceAt(0))[0];
+ AlignedCodonFrame humanMapping = protein.getCodonFrame(
+ protein.getSequenceAt(0)).get(0);
assertEquals(2, humanMapping.getdnaSeqs().length);
assertEquals(cdna1.getSequenceAt(1).getDatasetSequence(),
humanMapping.getdnaSeqs()[0]);
Mapping[] protMappings = humanMapping.getProtMappings();
// two mappings, both to cDNA with stop codon
assertEquals(2, protMappings.length);
+
MapList mapList = protMappings[0].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
assertTrue(Arrays.equals(new int[]
- { 1, 9 }, mapList.getFromRanges()));
+ { 1, 9 }, mapList.getFromRanges().get(0)));
+ assertEquals(1, mapList.getFromRanges().size());
assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges()));
+ { 1, 3 }, mapList.getToRanges().get(0)));
+ assertEquals(1, mapList.getToRanges().size());
+
mapList = protMappings[1].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
assertTrue(Arrays.equals(new int[]
- { 1, 9 }, mapList.getFromRanges()));
+ { 1, 9 }, mapList.getFromRanges().get(0)));
+ assertEquals(1, mapList.getFromRanges().size());
assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges()));
+ { 1, 3 }, mapList.getToRanges().get(0)));
+ assertEquals(1, mapList.getToRanges().size());
/*
* Inspect mapping for Mouse protein - should map to 1st/3rd/5th cDNA seqs
*/
- AlignedCodonFrame mouseMapping = protein.getCodonFrame(protein
- .getSequenceAt(1))[0];
+ AlignedCodonFrame mouseMapping = protein.getCodonFrame(
+ protein.getSequenceAt(1)).get(0);
assertEquals(3, mouseMapping.getdnaSeqs().length);
assertEquals(cdna1.getSequenceAt(0).getDatasetSequence(),
mouseMapping.getdnaSeqs()[0]);
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
assertTrue(Arrays.equals(new int[]
- { 4, 12 }, mapList.getFromRanges()));
+ { 4, 12 }, mapList.getFromRanges().get(0)));
+ assertEquals(1, mapList.getFromRanges().size());
assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges()));
+ { 1, 3 }, mapList.getToRanges().get(0)));
+ assertEquals(1, mapList.getToRanges().size());
// second mapping to cDNA with stop codon
mapList = protMappings[1].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
assertTrue(Arrays.equals(new int[]
- { 1, 9 }, mapList.getFromRanges()));
+ { 1, 9 }, mapList.getFromRanges().get(0)));
+ assertEquals(1, mapList.getFromRanges().size());
assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges()));
+ { 1, 3 }, mapList.getToRanges().get(0)));
+ assertEquals(1, mapList.getToRanges().size());
// third mapping to cDNA with start and stop codon
mapList = protMappings[2].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
assertTrue(Arrays.equals(new int[]
- { 4, 12 }, mapList.getFromRanges()));
+ { 4, 12 }, mapList.getFromRanges().get(0)));
+ assertEquals(1, mapList.getFromRanges().size());
assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges()));
+ { 1, 3 }, mapList.getToRanges().get(0)));
+ assertEquals(1, mapList.getToRanges().size());
}
/**
@Test
public void testAlignSequenceAs_withMapping_noIntrons()
{
- /*
- * Simple case: no gaps in dna
- */
- SequenceI dna = new Sequence("Seq1", "GGGAAA");
- dna.createDatasetSequence();
- SequenceI protein = new Sequence("Seq1", "-A-L-");
- protein.createDatasetSequence();
- AlignedCodonFrame acf = new AlignedCodonFrame();
MapList map = new MapList(new int[]
{ 1, 6 }, new int[]
{ 1, 2 }, 3, 1);
- acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map);
/*
* No existing gaps in dna:
*/
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- false);
- assertEquals("---GGG---AAA", dna.getSequenceAsString());
+ checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
+ "---GGG---AAA");
/*
* Now introduce gaps in dna but ignore them when realigning.
*/
- dna.setSequence("-G-G-G-A-A-A-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- false);
- assertEquals("---GGG---AAA", dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-G-G-A-A-A-", "-A-L-", false, false, map,
+ "---GGG---AAA");
/*
* Now include gaps in dna when realigning. First retaining 'mapped' gaps
* only, i.e. those within the exon region.
*/
- dna.setSequence("-G-G--G-A--A-A-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', true,
- false);
- assertEquals("---G-G--G---A--A-A", dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-G--G-A--A-A-", "-A-L-", true, false, map,
+ "---G-G--G---A--A-A");
/*
* Include all gaps in dna when realigning (within and without the exon
* region). The leading gap, and the gaps between codons, are subsumed by
* the protein alignment gap.
*/
- dna.setSequence("-G-GG--AA-A-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', true,
- true);
- assertEquals("---G-GG---AA-A-", dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", true, true, map,
+ "---G-GG---AA-A-");
/*
* Include only unmapped gaps in dna when realigning (outside the exon
* region). The leading gap, and the gaps between codons, are subsumed by
* the protein alignment gap.
*/
- dna.setSequence("-G-GG--AA-A-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- true);
- assertEquals("---GGG---AAA-", dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
+ "---GGG---AAA-");
}
/**
public void testAlignSequenceAs_withMapping_withIntrons()
{
/*
- * Simple case: no gaps in dna
- */
- SequenceI dna = new Sequence("Seq1", "GGGAAACCCTTTGGG");
- dna.createDatasetSequence();
- SequenceI protein = new Sequence("Seq1", "-A-L-");
- protein.createDatasetSequence();
- AlignedCodonFrame acf = new AlignedCodonFrame();
-
- /*
* Exons at codon 2 (AAA) and 4 (TTT)
*/
MapList map = new MapList(new int[]
{ 4, 6, 10, 12 }, new int[]
{ 1, 2 }, 3, 1);
- acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map);
/*
- * Align dna as "-A-L-". The protein 'gaps' follow the introns, i.e are
- * placed immediately before the mapped codons.
+ * Simple case: no gaps in dna
*/
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- false);
- assertEquals("GGG---AAACCC---TTTGGG", dna.getSequenceAsString());
+ checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
+ "GGG---AAACCCTTTGGG");
/*
* Add gaps to dna - but ignore when realigning.
*/
- dna.setSequence("-G-G-G--A--A---AC-CC-T-TT-GG-G-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- false);
- assertEquals("GGG---AAACCC---TTTGGG", dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-G-G--A--A---AC-CC-T-TT-GG-G-", "--A-L-",
+ false, false, map, "GGG---AAACCCTTTGGG");
/*
* Add gaps to dna - include within exons only when realigning.
*/
- dna.setSequence("-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', true,
- false);
- assertEquals("GGG---A--A---ACCC---T-TTGGG", dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
+ true, false, map, "GGG---A--A---ACCCT-TTGGG");
/*
* Include gaps outside exons only when realigning.
*/
- dna.setSequence("-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- true);
- assertEquals("-G-G-G---AAA-C-CC---TTT-GG-G-", dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
+ false, true, map, "-G-G-GAAAC-CCTTT-GG-G-");
+
+ /*
+ * Include gaps following first intron if we are 'preserving mapped gaps'
+ */
+ checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
+ true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
/*
* Include all gaps in dna when realigning.
*/
- dna.setSequence("-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', true,
- true);
- assertEquals("-G-G-G---A--A---A-C-CC---T-TT-GG-G-",
- dna.getSequenceAsString());
+ checkAlignSequenceAs("-G-G-G--A--A---A-C-CC-T-TT-GG-G-", "--A-L-",
+ true, true, map, "-G-G-G--A--A---A-C-CC-T-TT-GG-G-");
}
/**
@Test
public void testAlignSequenceAs_withMapping_withUnmappedProtein()
{
- SequenceI dna = new Sequence("Seq1", "GGGAAACCCTTTGGG");
+
+ /*
+ * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
+ */
+ final MapList map = new MapList(new int[]
+ { 4, 6, 10, 12 }, new int[]
+ { 1, 1, 3, 3 }, 3, 1);
+
+
+ /*
+ * Expect alignment does nothing (aborts realignment). Change this test
+ * first if different behaviour wanted.
+ */
+ checkAlignSequenceAs("GGGAAACCCTTTGGG", "-A-L-P-", false,
+ false, map, "GGGAAACCCTTTGGG");
+ }
+
+ /**
+ * Helper method that performs and verifies the method under test.
+ *
+ * @param dnaSeq
+ * @param proteinSeq
+ * @param preserveMappedGaps
+ * @param preserveUnmappedGaps
+ * @param map
+ * @param expected
+ */
+ protected void checkAlignSequenceAs(final String dnaSeq,
+ final String proteinSeq, final boolean preserveMappedGaps,
+ final boolean preserveUnmappedGaps, MapList map,
+ final String expected)
+ {
+ SequenceI dna = new Sequence("Seq1", dnaSeq);
dna.createDatasetSequence();
- SequenceI protein = new Sequence("Seq1", "-A-L-P-");
+ SequenceI protein = new Sequence("Seq1", proteinSeq);
protein.createDatasetSequence();
AlignedCodonFrame acf = new AlignedCodonFrame();
+ acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map);
+
+ AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-',
+ preserveMappedGaps, preserveUnmappedGaps);
+ assertEquals(expected, dna.getSequenceAsString());
+ }
+
+ /**
+ * Test for the alignSequenceAs method where we preserve gaps in introns only.
+ */
+ @Test
+ public void testAlignSequenceAs_keepIntronGapsOnly()
+ {
/*
- * Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
+ * Intron GGGAAA followed by exon CCCTTT
*/
MapList map = new MapList(new int[]
- { 4, 6, 10, 12 }, new int[]
- { 1, 1, 3, 3 }, 3, 1);
+ { 7, 12 }, new int[]
+ { 1, 2 }, 3, 1);
+
+ checkAlignSequenceAs("GG-G-AA-A-C-CC-T-TT", "AL",
+ false, true, map, "GG-G-AA-ACCCTTT");
+ }
+
+ /**
+ * Test for the method that generates an aligned translated sequence from one
+ * mapping.
+ */
+ @Test
+ public void testGetAlignedTranslation_dnaLikeProtein()
+ {
+ // dna alignment will be replaced
+ SequenceI dna = new Sequence("Seq1", "T-G-CC-A--T-TAC-CAG-");
+ dna.createDatasetSequence();
+ // protein alignment will be 'applied' to dna
+ SequenceI protein = new Sequence("Seq1", "-CH-Y--Q-");
+ protein.createDatasetSequence();
+ MapList map = new MapList(new int[]
+ { 1, 12 }, new int[]
+ { 1, 4 }, 3, 1);
+ AlignedCodonFrame acf = new AlignedCodonFrame();
acf.addMap(dna.getDatasetSequence(), protein.getDatasetSequence(), map);
+ final SequenceI aligned = AlignmentUtils
+ .getAlignedTranslation(protein, '-', acf);
+ assertEquals("---TGCCAT---TAC------CAG---", aligned.getSequenceAsString());
+ assertSame(aligned.getDatasetSequence(), dna.getDatasetSequence());
+ }
+
+ /**
+ * Test the method that realigns protein to match mapped codon alignment.
+ */
+ @Test
+ public void testAlignProteinAsDna()
+ {
+ // seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
+ SequenceI dna1 = new Sequence("Seq1", "TGCCATTACCAG-");
+ // seq2 codons are [1,3,4] [5,6,7] [8,9,10] [11,12,13]
+ SequenceI dna2 = new Sequence("Seq2", "T-GCCATTACCAG");
+ // seq3 codons are [1,2,3] [4,5,7] [8,9,10] [11,12,13]
+ SequenceI dna3 = new Sequence("Seq3", "TGCCA-TTACCAG");
+ AlignmentI dna = new Alignment(new SequenceI[]
+ { dna1, dna2, dna3 });
+ dna.setDataset(null);
+
+ // protein alignment will be realigned like dna
+ SequenceI prot1 = new Sequence("Seq1", "CHYQ");
+ SequenceI prot2 = new Sequence("Seq2", "CHYQ");
+ SequenceI prot3 = new Sequence("Seq3", "CHYQ");
+ AlignmentI protein = new Alignment(new SequenceI[]
+ { prot1, prot2, prot3 });
+ protein.setDataset(null);
+
+ MapList map = new MapList(new int[]
+ { 1, 12 }, new int[]
+ { 1, 4 }, 3, 1);
+ AlignedCodonFrame acf = new AlignedCodonFrame();
+ acf.addMap(dna1.getDatasetSequence(), prot1.getDatasetSequence(), map);
+ acf.addMap(dna2.getDatasetSequence(), prot2.getDatasetSequence(), map);
+ acf.addMap(dna3.getDatasetSequence(), prot3.getDatasetSequence(), map);
+ protein.setCodonFrames(Collections.singleton(acf));
+
/*
- * Align dna as "-A-L-P-". Currently, does nothing (aborts realignment).
- * Change this test first if different behaviour wanted.
+ * Translated codon order is [1,2,3] [1,3,4] [4,5,6] [4,5,7] [5,6,7] [7,8,9]
+ * [8,9,10] [10,11,12] [11,12,13]
*/
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- false);
- assertEquals("GGGAAACCCTTTGGG", dna.getSequenceAsString());
+ AlignmentUtils.alignProteinAsDna(protein, dna);
+ assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
+ assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
+ assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
+ }
+
+ /**
+ * Test the method that tests whether a CDNA sequence translates to a protein
+ * sequence
+ */
+ @Test
+ public void testTranslatesAs()
+ {
+ assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
+ "FPKG".toCharArray()));
+ // with start codon
+ assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
+ 3, "FPKG".toCharArray()));
+ // with stop codon1
+ assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
+ 0, "FPKG".toCharArray()));
+ // with stop codon2
+ assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
+ 0, "FPKG".toCharArray()));
+ // with stop codon3
+ assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
+ 0, "FPKG".toCharArray()));
+ // with start and stop codon1
+ assertTrue(AlignmentUtils.translatesAs(
+ "atgtttcccaaaggtaa".toCharArray(), 3, "FPKG".toCharArray()));
+ // with start and stop codon2
+ assertTrue(AlignmentUtils.translatesAs(
+ "atgtttcccaaaggtag".toCharArray(), 3, "FPKG".toCharArray()));
+ // with start and stop codon3
+ assertTrue(AlignmentUtils.translatesAs(
+ "atgtttcccaaaggtga".toCharArray(), 3, "FPKG".toCharArray()));
+
+ // wrong protein
+ assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
+ 0,
+ "FPMG".toCharArray()));
+ }
+
+ @Test
+ public void testTranslatesAs_withAmbiguityCodes()
+ {
+ // R means A or G
+ assertTrue(AlignmentUtils.translatesAs("car".toCharArray(), 0,
+ "Q".toCharArray()));
}
}