@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");
- dna.createDatasetSequence();
- SequenceI protein = new Sequence("Seq1", "-A-L-P-");
- protein.createDatasetSequence();
- AlignedCodonFrame acf = new AlignedCodonFrame();
-
+
/*
* Exons at codon 2 (AAA) and 4 (TTT) mapped to A and P
*/
- MapList map = new MapList(new int[]
+ 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", 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()
+ {
+
/*
- * Align dna as "-A-L-P-". Currently, does nothing (aborts realignment).
- * Change this test first if different behaviour wanted.
+ * Intron GGGAAA followed by exon CCCTTT
*/
- AlignmentUtils.alignSequenceAs(dna, protein, acf, "---", '-', false,
- false);
- assertEquals("GGGAAACCCTTTGGG", dna.getSequenceAsString());
+ MapList map = new MapList(new int[]
+ { 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");
}
}