*/
package jalview.analysis;
+import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertTrue;
-
-import org.junit.Test;
-
+import jalview.analysis.AlignmentUtils.MappingResult;
+import jalview.datamodel.AlignedCodonFrame;
import jalview.datamodel.Alignment;
import jalview.datamodel.AlignmentI;
+import jalview.datamodel.Mapping;
import jalview.datamodel.Sequence;
import jalview.datamodel.SequenceI;
import jalview.io.AppletFormatAdapter;
+import jalview.io.FormatAdapter;
+import jalview.util.MapList;
+
+import java.io.IOException;
+import java.util.Arrays;
+import java.util.List;
+import java.util.Map;
+
+import org.junit.Test;
public class AlignmentUtilsTests
{
+ // @formatter:off
+ private static final String TEST_DATA =
+ "# STOCKHOLM 1.0\n" +
+ "#=GS D.melanogaster.1 AC AY119185.1/838-902\n" +
+ "#=GS D.melanogaster.2 AC AC092237.1/57223-57161\n" +
+ "#=GS D.melanogaster.3 AC AY060611.1/560-627\n" +
+ "D.melanogaster.1 G.AGCC.CU...AUGAUCGA\n" +
+ "#=GR D.melanogaster.1 SS ................((((\n" +
+ "D.melanogaster.2 C.AUUCAACU.UAUGAGGAU\n" +
+ "#=GR D.melanogaster.2 SS ................((((\n" +
+ "D.melanogaster.3 G.UGGCGCU..UAUGACGCA\n" +
+ "#=GR D.melanogaster.3 SS (.(((...(....(((((((\n" +
+ "//";
+
+ private static final String AA_SEQS_1 =
+ ">Seq1Name\n" +
+ "K-QY--L\n" +
+ ">Seq2Name\n" +
+ "-R-FP-W-\n";
+
+ private static final String CDNA_SEQS_1 =
+ ">Seq1Name\n" +
+ "AC-GG--CUC-CAA-CT\n" +
+ ">Seq2Name\n" +
+ "-CG-TTA--ACG---AAGT\n";
+
+ private static final String CDNA_SEQS_2 =
+ ">Seq1Name\n" +
+ "GCTCGUCGTACT\n" +
+ ">Seq2Name\n" +
+ "GGGTCAGGCAGT\n";
+ // @formatter:on
+
public static Sequence ts=new Sequence("short","ASDASDASDASDASDASDASDASDASDASDASDASDASD");
+
@Test
public void testExpandFlanks()
{
assertTrue("Flanking sequence not the same as original dataset sequence.\n"+ung+"\n"+sq.getDatasetSequence().getSequenceAsString(),ung.equalsIgnoreCase(sq.getDatasetSequence().getSequenceAsString()));
}
}
+ }
}
+
+ /**
+ * Test method that returns a map of lists of sequences by sequence name.
+ *
+ * @throws IOException
+ */
+ @Test
+ public void testGetSequencesByName() throws IOException
+ {
+ final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
+ + ">Seq1Name\nABCD\n";
+ AlignmentI al = loadAlignment(data, "FASTA");
+ Map<String, List<SequenceI>> map = AlignmentUtils
+ .getSequencesByName(al);
+ assertEquals(2, map.keySet().size());
+ assertEquals(2, map.get("Seq1Name").size());
+ assertEquals("KQYL", map.get("Seq1Name").get(0).getSequenceAsString());
+ assertEquals("ABCD", map.get("Seq1Name").get(1).getSequenceAsString());
+ assertEquals(1, map.get("Seq2Name").size());
+ assertEquals("RFPW", map.get("Seq2Name").get(0).getSequenceAsString());
+ }
+ /**
+ * Helper method to load an alignment and ensure dataset sequences are set up.
+ *
+ * @param data
+ * @param format TODO
+ * @return
+ * @throws IOException
+ */
+ protected AlignmentI loadAlignment(final String data, String format) throws IOException
+ {
+ Alignment a = new FormatAdapter().readFile(data,
+ AppletFormatAdapter.PASTE, format);
+ a.setDataset(null);
+ return a;
+ }
+ /**
+ * Test mapping of protein to cDNA.
+ *
+ * @throws IOException
+ */
+ @Test
+ public void testMapProteinToCdna() throws IOException
+ {
+ // protein: Human + Mouse, 3 residues
+ AlignmentI protein = loadAlignment(
+ ">Human\nKQY\n>Mouse\nAFP\n>Worm\nRST\n",
+ "FASTA");
+ // cDNA: Mouse, Human, Mouse, 9 bases
+ // @formatter:off
+ String dnaData =
+ ">Mouse\nGAAATCCAG\n" +
+ ">Human\nTTCGATTAC\n" +
+ ">Mouse\nGTCGTTTGC\n" +
+ ">Mouse\nGTCGTTTGCgac\n" + // not mapped - wrong length
+ ">Fly\nGTCGTTTGC\n"; // not mapped - no name match
+ // @formatter:on
+ AlignmentI cdna1 = loadAlignment(
+ dnaData,
+ "FASTA");
+ MappingResult mapped = AlignmentUtils.mapProteinToCdna(protein, cdna1);
+ assertEquals(mapped, MappingResult.Mapped);
+
+ /*
+ * Check two mappings (one for Mouse, one for Human)
+ */
+ 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)).get(0);
+ assertEquals(1, humanMapping.getdnaSeqs().length);
+ assertEquals(cdna1.getSequenceAt(1).getDatasetSequence(),
+ humanMapping.getdnaSeqs()[0]);
+ Mapping[] protMappings = humanMapping.getProtMappings();
+ assertEquals(1, protMappings.length);
+ MapList mapList = protMappings[0].getMap();
+ assertEquals(3, mapList.getFromRatio());
+ assertEquals(1, mapList.getToRatio());
+ assertTrue(Arrays.equals(new int[]
+ { 1, 9 }, mapList.getFromRanges()));
+ assertTrue(Arrays.equals(new int[]
+ { 1, 3 }, mapList.getToRanges()));
+
+ /*
+ * Inspect mappings for Mouse protein
+ */
+ AlignedCodonFrame mouseMapping1 = protein.getCodonFrame(
+ protein.getSequenceAt(1)).get(0);
+ assertEquals(2, mouseMapping1.getdnaSeqs().length);
+ assertEquals(cdna1.getSequenceAt(0).getDatasetSequence(),
+ mouseMapping1.getdnaSeqs()[0]);
+ assertEquals(cdna1.getSequenceAt(2).getDatasetSequence(),
+ mouseMapping1.getdnaSeqs()[1]);
+ protMappings = mouseMapping1.getProtMappings();
+ assertEquals(2, protMappings.length);
+ for (int i = 0; i < 2; i++)
+ {
+ mapList = protMappings[i].getMap();
+ assertEquals(3, mapList.getFromRatio());
+ assertEquals(1, mapList.getToRatio());
+ assertTrue(Arrays.equals(new int[]
+ { 1, 9 }, mapList.getFromRanges()));
+ assertTrue(Arrays.equals(new int[]
+ { 1, 3 }, mapList.getToRanges()));
+ }
+ }
+
+ /**
+ * Test mapping of protein to cDNA which may include start and/or stop codons.
+ *
+ * @throws IOException
+ */
+ @Test
+ public void testMapProteinToCdna_stopStartCodons() throws IOException
+ {
+ // protein: Human + Mouse, 3 residues
+ AlignmentI protein = loadAlignment(
+ ">Human\nKQY\n>Mouse\nAFP\n>Worm\nRST\n", "FASTA");
+ // @formatter:off
+ String dnaData =
+ ">Mouse\natgGAAATCCAG\n" + // Mouse with start codon
+ ">Human\nTTCGATtactaa\n" + // Human with stop codon TAA
+ ">Mouse\nGTCGTTTGctaG\n" + // Mouse with stop codon TAG
+ ">Human\nGTCGTTTgctGa\n" + // Human with stop codon TGA
+ ">Mouse\nATGGTCGTTTGCtag\n"; // Mouse with start and stop codons
+ // @formatter:on
+ AlignmentI cdna1 = loadAlignment(
+ dnaData,
+ "FASTA");
+ MappingResult mapped = AlignmentUtils.mapProteinToCdna(protein, cdna1);
+ assertEquals(mapped, MappingResult.Mapped);
+
+ /*
+ * Check two mappings (one for Mouse, one for Human)
+ */
+ 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)).get(0);
+ assertEquals(2, humanMapping.getdnaSeqs().length);
+ assertEquals(cdna1.getSequenceAt(1).getDatasetSequence(),
+ humanMapping.getdnaSeqs()[0]);
+ assertEquals(cdna1.getSequenceAt(3).getDatasetSequence(),
+ humanMapping.getdnaSeqs()[1]);
+ 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()));
+ assertTrue(Arrays.equals(new int[]
+ { 1, 3 }, mapList.getToRanges()));
+ mapList = protMappings[1].getMap();
+ assertEquals(3, mapList.getFromRatio());
+ assertEquals(1, mapList.getToRatio());
+ assertTrue(Arrays.equals(new int[]
+ { 1, 9 }, mapList.getFromRanges()));
+ assertTrue(Arrays.equals(new int[]
+ { 1, 3 }, mapList.getToRanges()));
+
+ /*
+ * Inspect mapping for Mouse protein - should map to 1st/3rd/5th cDNA seqs
+ */
+ AlignedCodonFrame mouseMapping = protein.getCodonFrame(
+ protein.getSequenceAt(1)).get(0);
+ assertEquals(3, mouseMapping.getdnaSeqs().length);
+ assertEquals(cdna1.getSequenceAt(0).getDatasetSequence(),
+ mouseMapping.getdnaSeqs()[0]);
+ assertEquals(cdna1.getSequenceAt(2).getDatasetSequence(),
+ mouseMapping.getdnaSeqs()[1]);
+ assertEquals(cdna1.getSequenceAt(4).getDatasetSequence(),
+ mouseMapping.getdnaSeqs()[2]);
+
+ // three mappings
+ protMappings = mouseMapping.getProtMappings();
+ assertEquals(3, protMappings.length);
+
+ // first mapping to cDNA with start codon
+ mapList = protMappings[0].getMap();
+ assertEquals(3, mapList.getFromRatio());
+ assertEquals(1, mapList.getToRatio());
+ assertTrue(Arrays.equals(new int[]
+ { 4, 12 }, mapList.getFromRanges()));
+ assertTrue(Arrays.equals(new int[]
+ { 1, 3 }, mapList.getToRanges()));
+
+ // 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()));
+ assertTrue(Arrays.equals(new int[]
+ { 1, 3 }, mapList.getToRanges()));
+
+ // 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()));
+ assertTrue(Arrays.equals(new int[]
+ { 1, 3 }, mapList.getToRanges()));
+ }
+
+ /**
+ * Test for the alignSequenceAs method that takes two sequences and a mapping.
+ */
+ @Test
+ public void testAlignSequenceAs_withMapping_noIntrons()
+ {
+ MapList map = new MapList(new int[]
+ { 1, 6 }, new int[]
+ { 1, 2 }, 3, 1);
+
+ /*
+ * No existing gaps in dna:
+ */
+ checkAlignSequenceAs("GGGAAA", "-A-L-", false, false, map,
+ "---GGG---AAA");
+
+ /*
+ * Now introduce gaps in dna but ignore them when realigning.
+ */
+ 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.
+ */
+ 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.
+ */
+ 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.
+ */
+ checkAlignSequenceAs("-G-GG--AA-A-", "-A-L-", false, true, map,
+ "---GGG---AAA-");
+ }
+
+ /**
+ * Test for the alignSequenceAs method that takes two sequences and a mapping.
+ */
+ @Test
+ public void testAlignSequenceAs_withMapping_withIntrons()
+ {
+ /*
+ * Exons at codon 2 (AAA) and 4 (TTT)
+ */
+ MapList map = new MapList(new int[]
+ { 4, 6, 10, 12 }, new int[]
+ { 1, 2 }, 3, 1);
+
+ /*
+ * Simple case: no gaps in dna
+ */
+ checkAlignSequenceAs("GGGAAACCCTTTGGG", "--A-L-", false, false, map,
+ "GGG---AAACCCTTTGGG");
+
+ /*
+ * Add gaps to dna - but ignore when realigning.
+ */
+ 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.
+ */
+ 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.
+ */
+ 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.
+ */
+ 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 for the case where not all of the protein sequence is mapped to cDNA.
+ */
+ @Test
+ public void testAlignSequenceAs_withMapping_withUnmappedProtein()
+ {
+
+ /*
+ * 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", 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()
+ {
+
+ /*
+ * Intron GGGAAA followed by exon CCCTTT
+ */
+ 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");
}
}