/*
- * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
- * Copyright (C) $$Year-Rel$$ The Jalview Authors
+ * Jalview - A Sequence Alignment Editor and Viewer (Version 2.9.0b2)
+ * Copyright (C) 2015 The Jalview Authors
*
* This file is part of Jalview.
*
*/
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 java.io.IOException;
-import java.util.ArrayList;
-import java.util.Arrays;
-import java.util.Collections;
-import java.util.HashSet;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
-
-import org.junit.Test;
+import static org.testng.AssertJUnit.assertEquals;
+import static org.testng.AssertJUnit.assertFalse;
+import static org.testng.AssertJUnit.assertNull;
+import static org.testng.AssertJUnit.assertSame;
+import static org.testng.AssertJUnit.assertTrue;
import jalview.datamodel.AlignedCodonFrame;
import jalview.datamodel.Alignment;
import jalview.util.MapList;
import jalview.util.MappingUtils;
-public class AlignmentUtilsTests
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Arrays;
+import java.util.Collections;
+import java.util.HashSet;
+import java.util.Iterator;
+import java.util.LinkedHashSet;
+import java.util.List;
+import java.util.Map;
+import java.util.Set;
+
+import org.testng.annotations.Test;
+
+public class AlignmentUtilsTests
{
// @formatter:off
private static final String TEST_DATA =
"GGGTCAGGCAGT\n";
// @formatter:on
- public static Sequence ts=new Sequence("short","ASDASDASDASDASDASDASDASDASDASDASDASDASD");
+ // public static Sequence ts=new
+ // Sequence("short","ASDASDASDASDASDASDASDASDASDASDASDASDASD");
+ public static Sequence ts = new Sequence("short",
+ "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklm");
- @Test
- public void testExpandFlanks()
+ @Test(groups = { "Functional" })
+ public void testExpandContext()
{
AlignmentI al = new Alignment(new Sequence[] {});
- for (int i=4;i<14;i+=3)
+ for (int i = 4; i < 14; i += 2)
{
- SequenceI s1=ts.deriveSequence().getSubSequence(i, i+7);
+ SequenceI s1 = ts.deriveSequence().getSubSequence(i, i + 7);
al.addSequence(s1);
}
- System.out.println(new AppletFormatAdapter().formatSequences("Clustal", al, true));
- for (int flnk=-1;flnk<25; flnk++)
+ System.out.println(new AppletFormatAdapter().formatSequences("Clustal",
+ al, true));
+ for (int flnk = -1; flnk < 25; flnk++)
{
- AlignmentI exp;
- System.out.println("\nFlank size: "+flnk);
- System.out.println(new AppletFormatAdapter().formatSequences("Clustal", exp=AlignmentUtils.expandContext(al, flnk), true));
- if (flnk==-1) {
- for (SequenceI sq:exp.getSequences())
+ AlignmentI exp = AlignmentUtils.expandContext(al, flnk);
+ System.out.println("\nFlank size: " + flnk);
+ System.out.println(new AppletFormatAdapter().formatSequences(
+ "Clustal", exp, true));
+ if (flnk == -1)
{
+ /*
+ * Full expansion to complete sequences
+ */
+ for (SequenceI sq : exp.getSequences())
+ {
String ung = sq.getSequenceAsString().replaceAll("-+", "");
- assertTrue("Flanking sequence not the same as original dataset sequence.\n"+ung+"\n"+sq.getDatasetSequence().getSequenceAsString(),ung.equalsIgnoreCase(sq.getDatasetSequence().getSequenceAsString()));
+ final String errorMsg = "Flanking sequence not the same as original dataset sequence.\n"
+ + ung
+ + "\n"
+ + sq.getDatasetSequence().getSequenceAsString();
+ assertTrue(errorMsg, ung.equalsIgnoreCase(sq.getDatasetSequence()
+ .getSequenceAsString()));
+ }
}
+ else if (flnk == 24)
+ {
+ /*
+ * Last sequence is fully expanded, others have leading gaps to match
+ */
+ assertTrue(exp.getSequenceAt(4).getSequenceAsString()
+ .startsWith("abc"));
+ assertTrue(exp.getSequenceAt(3).getSequenceAsString()
+ .startsWith("--abc"));
+ assertTrue(exp.getSequenceAt(2).getSequenceAsString()
+ .startsWith("----abc"));
+ assertTrue(exp.getSequenceAt(1).getSequenceAsString()
+ .startsWith("------abc"));
+ assertTrue(exp.getSequenceAt(0).getSequenceAsString()
+ .startsWith("--------abc"));
}
}
- }
+ }
+
+ /**
+ * Test that annotations are correctly adjusted by expandContext
+ */
+ @Test(groups = { "Functional" })
+ public void testExpandContext_annotation()
+ {
+ AlignmentI al = new Alignment(new Sequence[] {});
+ SequenceI ds = new Sequence("Seq1", "ABCDEFGHI");
+ // subsequence DEF:
+ SequenceI seq1 = ds.deriveSequence().getSubSequence(3, 6);
+ al.addSequence(seq1);
+
+ /*
+ * Annotate DEF with 4/5/6 respectively
+ */
+ Annotation[] anns = new Annotation[] { new Annotation(4),
+ new Annotation(5), new Annotation(6) };
+ AlignmentAnnotation ann = new AlignmentAnnotation("SS",
+ "secondary structure", anns);
+ seq1.addAlignmentAnnotation(ann);
+
+ /*
+ * The annotations array should match aligned positions
+ */
+ assertEquals(3, ann.annotations.length);
+ assertEquals(4, ann.annotations[0].value, 0.001);
+ assertEquals(5, ann.annotations[1].value, 0.001);
+ assertEquals(6, ann.annotations[2].value, 0.001);
+
+ /*
+ * Check annotation to sequence position mappings before expanding the
+ * sequence; these are set up in Sequence.addAlignmentAnnotation ->
+ * Annotation.setSequenceRef -> createSequenceMappings
+ */
+ assertNull(ann.getAnnotationForPosition(1));
+ assertNull(ann.getAnnotationForPosition(2));
+ assertNull(ann.getAnnotationForPosition(3));
+ assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
+ assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
+ assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
+ assertNull(ann.getAnnotationForPosition(7));
+ assertNull(ann.getAnnotationForPosition(8));
+ assertNull(ann.getAnnotationForPosition(9));
+
+ /*
+ * Expand the subsequence to the full sequence abcDEFghi
+ */
+ AlignmentI expanded = AlignmentUtils.expandContext(al, -1);
+ assertEquals("abcDEFghi", expanded.getSequenceAt(0)
+ .getSequenceAsString());
+
+ /*
+ * Confirm the alignment and sequence have the same SS annotation,
+ * referencing the expanded sequence
+ */
+ ann = expanded.getSequenceAt(0).getAnnotation()[0];
+ assertSame(ann, expanded.getAlignmentAnnotation()[0]);
+ assertSame(expanded.getSequenceAt(0), ann.sequenceRef);
+
+ /*
+ * The annotations array should have null values except for annotated
+ * positions
+ */
+ assertNull(ann.annotations[0]);
+ assertNull(ann.annotations[1]);
+ assertNull(ann.annotations[2]);
+ assertEquals(4, ann.annotations[3].value, 0.001);
+ assertEquals(5, ann.annotations[4].value, 0.001);
+ assertEquals(6, ann.annotations[5].value, 0.001);
+ assertNull(ann.annotations[6]);
+ assertNull(ann.annotations[7]);
+ assertNull(ann.annotations[8]);
+
+ /*
+ * sequence position mappings should be unchanged
+ */
+ assertNull(ann.getAnnotationForPosition(1));
+ assertNull(ann.getAnnotationForPosition(2));
+ assertNull(ann.getAnnotationForPosition(3));
+ assertEquals(4, ann.getAnnotationForPosition(4).value, 0.001);
+ assertEquals(5, ann.getAnnotationForPosition(5).value, 0.001);
+ assertEquals(6, ann.getAnnotationForPosition(6).value, 0.001);
+ assertNull(ann.getAnnotationForPosition(7));
+ assertNull(ann.getAnnotationForPosition(8));
+ assertNull(ann.getAnnotationForPosition(9));
+ }
/**
* Test method that returns a map of lists of sequences by sequence name.
*
* @throws IOException
*/
- @Test
+ @Test(groups = { "Functional" })
public void testGetSequencesByName() throws IOException
{
final String data = ">Seq1Name\nKQYL\n" + ">Seq2Name\nRFPW\n"
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
+ * @param format
+ * TODO
* @return
* @throws IOException
*/
- protected AlignmentI loadAlignment(final String data, String format) throws IOException
+ protected AlignmentI loadAlignment(final String data, String format)
+ throws IOException
{
- Alignment a = new FormatAdapter().readFile(data,
+ AlignmentI a = new FormatAdapter().readFile(data,
AppletFormatAdapter.PASTE, format);
a.setDataset(null);
return a;
}
-
+
/**
* Test mapping of protein to cDNA, for the case where we have no sequence
* cross-references, so mappings are made first-served 1-1 where sequences
*
* @throws IOException
*/
- @Test
- public void testMapProteinToCdna_noXrefs() throws IOException
+ @Test(groups = { "Functional" })
+ public void testMapProteinAlignmentToCdna_noXrefs() throws IOException
{
List<SequenceI> protseqs = new ArrayList<SequenceI>();
protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
cdna.setDataset(null);
- assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
+ assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
// 3 mappings made, each from 1 to 1 sequence
assertEquals(3, protein.getCodonFrames().size());
assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
// V12345 mapped to A22222
- AlignedCodonFrame acf = protein.getCodonFrame(
- protein.getSequenceAt(0)).get(0);
+ AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
+ .get(0);
assertEquals(1, acf.getdnaSeqs().length);
assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
acf.getdnaSeqs()[0]);
MapList mapList = protMappings[0].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
- assertTrue(Arrays.equals(new int[]
- { 1, 9 }, mapList.getFromRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
+ .get(0)));
assertEquals(1, mapList.getFromRanges().size());
- assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 1, 3 },
+ mapList.getToRanges().get(0)));
assertEquals(1, mapList.getToRanges().size());
// V12346 mapped to A33333
/**
* Test for the alignSequenceAs method that takes two sequences and a mapping.
*/
- @Test
+ @Test(groups = { "Functional" })
public void testAlignSequenceAs_withMapping_noIntrons()
{
- MapList map = new MapList(new int[]
- { 1, 6 }, new int[]
- { 1, 2 }, 3, 1);
+ MapList map = new MapList(new int[] { 1, 6 }, new int[] { 1, 2 }, 3, 1);
/*
* No existing gaps in dna:
/**
* Test for the alignSequenceAs method that takes two sequences and a mapping.
*/
- @Test
+ @Test(groups = { "Functional" })
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);
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 2 }, 3, 1);
/*
* Simple case: no gaps in dna
/**
* Test for the case where not all of the protein sequence is mapped to cDNA.
*/
- @Test
+ @Test(groups = { "Functional" })
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);
-
+ 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");
+ checkAlignSequenceAs("GGGAAACCCTTTGGG", "-A-L-P-", false, false, map,
+ "GGGAAACCCTTTGGG");
}
/**
/**
* Test for the alignSequenceAs method where we preserve gaps in introns only.
*/
- @Test
+ @Test(groups = { "Functional" })
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");
+ 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");
}
/**
* Test for the method that generates an aligned translated sequence from one
* mapping.
*/
- @Test
+ @Test(groups = { "Functional" })
public void testGetAlignedTranslation_dnaLikeProtein()
{
// dna alignment will be replaced
// 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);
+ 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());
+ 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
+ @Test(groups = { "Functional" })
public void testAlignProteinAsDna()
{
// seq1 codons are [1,2,3] [4,5,6] [7,8,9] [10,11,12]
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 });
+ 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 });
+ SequenceI prot4 = new Sequence("Seq4", "R-QSV"); // unmapped, unchanged
+ AlignmentI protein = new Alignment(new SequenceI[] { prot1, prot2,
+ prot3, prot4 });
protein.setDataset(null);
- MapList map = new MapList(new int[]
- { 1, 12 }, new int[]
- { 1, 4 }, 3, 1);
+ 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);
assertEquals("C-H--Y-Q-", prot1.getSequenceAsString());
assertEquals("-C--H-Y-Q", prot2.getSequenceAsString());
assertEquals("C--H--Y-Q", prot3.getSequenceAsString());
+ assertEquals("R-QSV", prot4.getSequenceAsString());
}
/**
* Test the method that tests whether a CDNA sequence translates to a protein
* sequence
*/
- @Test
+ @Test(groups = { "Functional" })
public void testTranslatesAs()
{
assertTrue(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(), 0,
"FPKG".toCharArray()));
- // with start codon
+ // with start codon (not in protein)
assertTrue(AlignmentUtils.translatesAs("atgtttcccaaaggg".toCharArray(),
3, "FPKG".toCharArray()));
- // with stop codon1
+ // with stop codon1 (not in protein)
assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
0, "FPKG".toCharArray()));
- // with stop codon2
+ // with stop codon1 (in protein as *)
+ assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtaa".toCharArray(),
+ 0, "FPKG*".toCharArray()));
+ // with stop codon2 (not in protein)
assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtag".toCharArray(),
0, "FPKG".toCharArray()));
- // with stop codon3
+ // with stop codon3 (not in protein)
assertTrue(AlignmentUtils.translatesAs("tttcccaaagggtga".toCharArray(),
0, "FPKG".toCharArray()));
// with start and stop codon1
assertTrue(AlignmentUtils.translatesAs(
- "atgtttcccaaaggtaa".toCharArray(), 3, "FPKG".toCharArray()));
+ "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG".toCharArray()));
+ // with start and stop codon1 (in protein as *)
+ assertTrue(AlignmentUtils.translatesAs(
+ "atgtttcccaaagggtaa".toCharArray(), 3, "FPKG*".toCharArray()));
// with start and stop codon2
assertTrue(AlignmentUtils.translatesAs(
- "atgtttcccaaaggtag".toCharArray(), 3, "FPKG".toCharArray()));
+ "atgtttcccaaagggtag".toCharArray(), 3, "FPKG".toCharArray()));
// with start and stop codon3
assertTrue(AlignmentUtils.translatesAs(
- "atgtttcccaaaggtga".toCharArray(), 3, "FPKG".toCharArray()));
+ "atgtttcccaaagggtga".toCharArray(), 3, "FPKG".toCharArray()));
+
+ // with embedded stop codon
+ assertTrue(AlignmentUtils.translatesAs(
+ "atgtttTAGcccaaaTAAgggtga".toCharArray(), 3,
+ "F*PK*G".toCharArray()));
// wrong protein
assertFalse(AlignmentUtils.translatesAs("tttcccaaaggg".toCharArray(),
- 0,
- "FPMG".toCharArray()));
+ 0, "FPMG".toCharArray()));
}
/**
*
* @throws IOException
*/
- @Test
- public void testMapProteinToCdna_withStartAndStopCodons()
+ @Test(groups = { "Functional" })
+ public void testMapProteinAlignmentToCdna_withStartAndStopCodons()
throws IOException
{
List<SequenceI> protseqs = new ArrayList<SequenceI>();
protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
protein.setDataset(null);
-
+
List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
// start + SAR:
dnaseqs.add(new Sequence("EMBL|A11111", "ATGTCAGCACGC"));
dnaseqs.add(new Sequence("EMBL|A44444", "GAAATTCAG"));
AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[4]));
cdna.setDataset(null);
-
- assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
+
+ assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
// 3 mappings made, each from 1 to 1 sequence
assertEquals(3, protein.getCodonFrames().size());
assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(2)).size());
-
+
// V12345 mapped from A22222
- AlignedCodonFrame acf = protein.getCodonFrame(
- protein.getSequenceAt(0)).get(0);
+ AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
+ .get(0);
assertEquals(1, acf.getdnaSeqs().length);
assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
acf.getdnaSeqs()[0]);
MapList mapList = protMappings[0].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
- assertTrue(Arrays.equals(new int[]
- { 1, 9 }, mapList.getFromRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 1, 9 }, mapList.getFromRanges()
+ .get(0)));
assertEquals(1, mapList.getFromRanges().size());
- assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 1, 3 },
+ mapList.getToRanges().get(0)));
assertEquals(1, mapList.getToRanges().size());
// V12346 mapped from A33333 starting position 4
mapList = protMappings[0].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
- assertTrue(Arrays.equals(new int[]
- { 4, 12 }, mapList.getFromRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
+ .get(0)));
assertEquals(1, mapList.getFromRanges().size());
- assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 1, 3 },
+ mapList.getToRanges().get(0)));
assertEquals(1, mapList.getToRanges().size());
-
+
// V12347 mapped to A11111 starting position 4
acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
assertEquals(1, acf.getdnaSeqs().length);
mapList = protMappings[0].getMap();
assertEquals(3, mapList.getFromRatio());
assertEquals(1, mapList.getToRatio());
- assertTrue(Arrays.equals(new int[]
- { 4, 12 }, mapList.getFromRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 4, 12 }, mapList.getFromRanges()
+ .get(0)));
assertEquals(1, mapList.getFromRanges().size());
- assertTrue(Arrays.equals(new int[]
- { 1, 3 }, mapList.getToRanges().get(0)));
+ assertTrue(Arrays.equals(new int[] { 1, 3 },
+ mapList.getToRanges().get(0)));
assertEquals(1, mapList.getToRanges().size());
-
+
// no mapping involving the 'extra' A44444
assertTrue(protein.getCodonFrame(cdna.getSequenceAt(3)).isEmpty());
}
*
* @throws IOException
*/
- @Test
- public void testMapProteinToCdna_withXrefs() throws IOException
+ @Test(groups = { "Functional" })
+ public void testMapProteinAlignmentToCdna_withXrefs() throws IOException
{
List<SequenceI> protseqs = new ArrayList<SequenceI>();
protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
protseqs.add(new Sequence("UNIPROT|V12347", "SAR"));
AlignmentI protein = new Alignment(protseqs.toArray(new SequenceI[3]));
protein.setDataset(null);
-
+
List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
dnaseqs.add(new Sequence("EMBL|A11111", "TCAGCACGC")); // = SAR
dnaseqs.add(new Sequence("EMBL|A22222", "ATGGAGATACAA")); // = start + EIQ
dnaseqs.add(new Sequence("EMBL|A55555", "GAGATTCAG")); // = EIQ
AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[5]));
cdna.setDataset(null);
-
+
// Xref A22222 to V12345 (should get mapped)
dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
// Xref V12345 to A44444 (should get mapped)
// A11111 should be mapped to V12347
// A55555 is spare and has no xref so is not mapped
- assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
+ assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
// 4 protein mappings made for 3 proteins, 2 to V12345, 1 each to V12346/7
assertEquals(3, protein.getCodonFrames().size());
assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(2)).size());
assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(3)).size());
-
+
// V12345 mapped to A22222 and A44444
- AlignedCodonFrame acf = protein.getCodonFrame(
- protein.getSequenceAt(0)).get(0);
+ AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
+ .get(0);
assertEquals(2, acf.getdnaSeqs().length);
assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
acf.getdnaSeqs()[0]);
assertEquals(cdna.getSequenceAt(3).getDatasetSequence(),
acf.getdnaSeqs()[1]);
-
+
// V12346 mapped to A33333
acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
assertEquals(1, acf.getdnaSeqs().length);
assertEquals(cdna.getSequenceAt(2).getDatasetSequence(),
acf.getdnaSeqs()[0]);
-
+
// V12347 mapped to A11111
acf = protein.getCodonFrame(protein.getSequenceAt(2)).get(0);
assertEquals(1, acf.getdnaSeqs().length);
assertEquals(cdna.getSequenceAt(0).getDatasetSequence(),
acf.getdnaSeqs()[0]);
-
+
// no mapping involving the 'extra' A55555
assertTrue(protein.getCodonFrame(cdna.getSequenceAt(4)).isEmpty());
}
*
* @throws IOException
*/
- @Test
- public void testMapProteinToCdna_prioritiseXrefs() throws IOException
+ @Test(groups = { "Functional" })
+ public void testMapProteinAlignmentToCdna_prioritiseXrefs()
+ throws IOException
{
List<SequenceI> protseqs = new ArrayList<SequenceI>();
protseqs.add(new Sequence("UNIPROT|V12345", "EIQ"));
AlignmentI protein = new Alignment(
protseqs.toArray(new SequenceI[protseqs.size()]));
protein.setDataset(null);
-
+
List<SequenceI> dnaseqs = new ArrayList<SequenceI>();
dnaseqs.add(new Sequence("EMBL|A11111", "GAAATCCAG")); // = EIQ
dnaseqs.add(new Sequence("EMBL|A22222", "GAAATTCAG")); // = EIQ
AlignmentI cdna = new Alignment(dnaseqs.toArray(new SequenceI[dnaseqs
.size()]));
cdna.setDataset(null);
-
+
// Xref A22222 to V12345 (should get mapped)
// A11111 should then be mapped to the unmapped V12346
dnaseqs.get(1).addDBRef(new DBRefEntry("UNIPROT", "1", "V12345"));
-
- assertTrue(AlignmentUtils.mapProteinToCdna(protein, cdna));
-
+
+ assertTrue(AlignmentUtils.mapProteinAlignmentToCdna(protein, cdna));
+
// 2 protein mappings made
assertEquals(2, protein.getCodonFrames().size());
assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(0)).size());
assertEquals(1, protein.getCodonFrame(protein.getSequenceAt(1)).size());
-
+
// one mapping for each of the cDNA sequences
assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(0)).size());
assertEquals(1, protein.getCodonFrame(cdna.getSequenceAt(1)).size());
-
+
// V12345 mapped to A22222
AlignedCodonFrame acf = protein.getCodonFrame(protein.getSequenceAt(0))
.get(0);
assertEquals(1, acf.getdnaSeqs().length);
assertEquals(cdna.getSequenceAt(1).getDatasetSequence(),
acf.getdnaSeqs()[0]);
-
+
// V12346 mapped to A11111
acf = protein.getCodonFrame(protein.getSequenceAt(1)).get(0);
assertEquals(1, acf.getdnaSeqs().length);
* Test the method that shows or hides sequence annotations by type(s) and
* selection group.
*/
- @Test
+ @Test(groups = { "Functional" })
public void testShowOrHideSequenceAnnotations()
{
SequenceI seq1 = new Sequence("Seq1", "AAA");
SequenceI seq2 = new Sequence("Seq2", "BBB");
SequenceI seq3 = new Sequence("Seq3", "CCC");
- Annotation[] anns = new Annotation[]
- { new Annotation(2f) };
+ Annotation[] anns = new Annotation[] { new Annotation(2f) };
AlignmentAnnotation ann1 = new AlignmentAnnotation("Structure", "ann1",
anns);
ann1.setSequenceRef(seq1);
AlignmentAnnotation ann5 = new AlignmentAnnotation("Temp", "ann5", anns);
ann5.setSequenceRef(seq2);
AlignmentAnnotation ann6 = new AlignmentAnnotation("Temp", "ann6", anns);
- AlignmentI al = new Alignment(new SequenceI[] {seq1, seq2, seq3});
+ AlignmentI al = new Alignment(new SequenceI[] { seq1, seq2, seq3 });
al.addAnnotation(ann1); // Structure for Seq1
al.addAnnotation(ann2); // Structure for Seq2
al.addAnnotation(ann3); // Structure for no sequence
/**
* Tests for the method that checks if one sequence cross-references another
*/
- @Test
+ @Test(groups = { "Functional" })
public void testHasCrossRef()
{
assertFalse(AlignmentUtils.hasCrossRef(null, null));
assertFalse(AlignmentUtils.hasCrossRef(null, seq1));
SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
-
+
// different ref
seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20193"));
assertFalse(AlignmentUtils.hasCrossRef(seq1, seq2));
-
+
// case-insensitive; version number is ignored
seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "v20192"));
assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
-
+
// right case!
seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
assertTrue(AlignmentUtils.hasCrossRef(seq1, seq2));
* Tests for the method that checks if either sequence cross-references the
* other
*/
- @Test
+ @Test(groups = { "Functional" })
public void testHaveCrossRef()
{
assertFalse(AlignmentUtils.hasCrossRef(null, null));
assertFalse(AlignmentUtils.haveCrossRef(null, seq1));
SequenceI seq2 = new Sequence("UNIPROT|V20192", "ABCDEF");
assertFalse(AlignmentUtils.haveCrossRef(seq1, seq2));
-
+
seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
// next is true for haveCrossRef, false for hasCrossRef
assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
-
+
// now the other way round
seq1.setDBRef(null);
seq2.addDBRef(new DBRefEntry("EMBL", "1", "A12345"));
assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
assertTrue(AlignmentUtils.haveCrossRef(seq2, seq1));
-
+
// now both ways
seq1.addDBRef(new DBRefEntry("UNIPROT", "1", "V20192"));
assertTrue(AlignmentUtils.haveCrossRef(seq1, seq2));
/**
* Test the method that extracts the exon-only part of a dna alignment.
*/
- @Test
+ @Test(groups = { "Functional" })
public void testMakeExonAlignment()
{
SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
pep2.createDatasetSequence();
Set<AlignedCodonFrame> mappings = new HashSet<AlignedCodonFrame>();
- MapList map = new MapList(new int[]
- { 4, 6, 10, 12 }, new int[]
- { 1, 2 }, 3, 1);
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 2 }, 3, 1);
AlignedCodonFrame acf = new AlignedCodonFrame();
acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
mappings.add(acf);
- map = new MapList(new int[]
- { 1, 3, 7, 9, 13, 15 }, new int[]
- { 1, 3 }, 3, 1);
+ map = new MapList(new int[] { 1, 3, 7, 9, 13, 15 }, new int[] { 1, 3 },
+ 3, 1);
acf = new AlignedCodonFrame();
acf.addMap(dna2.getDatasetSequence(), pep2.getDatasetSequence(), map);
mappings.add(acf);
-
- AlignmentI exons = AlignmentUtils.makeExonAlignment(new SequenceI[]
- { dna1, dna2 }, mappings);
+
+ AlignmentI exons = AlignmentUtils.makeExonAlignment(new SequenceI[] {
+ dna1, dna2 }, mappings);
assertEquals(2, exons.getSequences().size());
assertEquals("GGGTTT", exons.getSequenceAt(0).getSequenceAsString());
assertEquals("GGGTTTCCC", exons.getSequenceAt(1).getSequenceAsString());
* already has a protein product (Uniprot translation) which in turn has an
* x-ref to the EMBLCDS record.
*/
- @Test
- public void testMakeExonSequence()
+ @Test(groups = { "Functional" })
+ public void testMakeExonSequences()
{
SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
SequenceI pep1 = new Sequence("pep1", "GF");
* EMBLCDS|A12345.
*/
Set<AlignedCodonFrame> mappings = new HashSet<AlignedCodonFrame>();
- MapList map = new MapList(new int[]
- { 4, 6, 10, 12 }, new int[]
- { 1, 2 }, 3, 1);
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 2 }, 3, 1);
AlignedCodonFrame acf = new AlignedCodonFrame();
acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
mappings.add(acf);
AlignedCodonFrame newMapping = new AlignedCodonFrame();
- SequenceI exon = AlignmentUtils.makeExonSequence(dna1, acf, newMapping);
+ List<SequenceI> exons = AlignmentUtils.makeExonSequences(dna1, acf,
+ newMapping);
+ assertEquals(1, exons.size());
+ SequenceI exon = exons.get(0);
+
+ assertEquals("GGGTTT", exon.getSequenceAsString());
+ assertEquals("dna1|A12345", exon.getName());
+ assertEquals(1, exon.getDBRef().length);
+ DBRefEntry cdsRef = exon.getDBRef()[0];
+ assertEquals("EMBLCDS", cdsRef.getSource());
+ assertEquals("2", cdsRef.getVersion());
+ assertEquals("A12345", cdsRef.getAccessionId());
+ }
+
+ /**
+ * Test the method that makes an exon-only alignment from a DNA sequence and
+ * its product mappings, for the case where there are multiple exon mappings
+ * to different protein products.
+ */
+ @Test(groups = { "Functional" })
+ public void testMakeExonAlignment_multipleProteins()
+ {
+ SequenceI dna1 = new Sequence("dna1", "aaaGGGcccTTTaaa");
+ SequenceI pep1 = new Sequence("pep1", "GF"); // GGGTTT
+ SequenceI pep2 = new Sequence("pep2", "KP"); // aaaccc
+ SequenceI pep3 = new Sequence("pep3", "KF"); // aaaTTT
+ dna1.createDatasetSequence();
+ pep1.createDatasetSequence();
+ pep2.createDatasetSequence();
+ pep3.createDatasetSequence();
+ pep1.getDatasetSequence().addDBRef(
+ new DBRefEntry("EMBLCDS", "2", "A12345"));
+ pep2.getDatasetSequence().addDBRef(
+ new DBRefEntry("EMBLCDS", "3", "A12346"));
+ pep3.getDatasetSequence().addDBRef(
+ new DBRefEntry("EMBLCDS", "4", "A12347"));
+
+ /*
+ * Make the mappings from dna to protein. Using LinkedHashset is a
+ * convenience so results are in the input order. There is no assertion that
+ * the generated exon sequences are in any particular order.
+ */
+ Set<AlignedCodonFrame> mappings = new LinkedHashSet<AlignedCodonFrame>();
+ // map ...GGG...TTT to GF
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 2 }, 3, 1);
+ AlignedCodonFrame acf = new AlignedCodonFrame();
+ acf.addMap(dna1.getDatasetSequence(), pep1.getDatasetSequence(), map);
+ mappings.add(acf);
+
+ // map aaa...ccc to KP
+ map = new MapList(new int[] { 1, 3, 7, 9 }, new int[] { 1, 2 }, 3, 1);
+ acf = new AlignedCodonFrame();
+ acf.addMap(dna1.getDatasetSequence(), pep2.getDatasetSequence(), map);
+ mappings.add(acf);
+ // map aaa......TTT to KF
+ map = new MapList(new int[] { 1, 3, 10, 12 }, new int[] { 1, 2 }, 3, 1);
+ acf = new AlignedCodonFrame();
+ acf.addMap(dna1.getDatasetSequence(), pep3.getDatasetSequence(), map);
+ mappings.add(acf);
+
+ /*
+ * Create the Exon alignment; also replaces the dna-to-protein mappings with
+ * exon-to-protein and exon-to-dna mappings
+ */
+ AlignmentI exal = AlignmentUtils.makeExonAlignment(
+ new SequenceI[] { dna1 }, mappings);
+
+ /*
+ * Verify we have 3 exon sequences, mapped to pep1/2/3 respectively
+ */
+ List<SequenceI> exons = exal.getSequences();
+ assertEquals(3, exons.size());
+
+ SequenceI exon = exons.get(0);
assertEquals("GGGTTT", exon.getSequenceAsString());
assertEquals("dna1|A12345", exon.getName());
assertEquals(1, exon.getDBRef().length);
assertEquals("2", cdsRef.getVersion());
assertEquals("A12345", cdsRef.getAccessionId());
+ exon = exons.get(1);
+ assertEquals("aaaccc", exon.getSequenceAsString());
+ assertEquals("dna1|A12346", exon.getName());
+ assertEquals(1, exon.getDBRef().length);
+ cdsRef = exon.getDBRef()[0];
+ assertEquals("EMBLCDS", cdsRef.getSource());
+ assertEquals("3", cdsRef.getVersion());
+ assertEquals("A12346", cdsRef.getAccessionId());
+
+ exon = exons.get(2);
+ assertEquals("aaaTTT", exon.getSequenceAsString());
+ assertEquals("dna1|A12347", exon.getName());
+ assertEquals(1, exon.getDBRef().length);
+ cdsRef = exon.getDBRef()[0];
+ assertEquals("EMBLCDS", cdsRef.getSource());
+ assertEquals("4", cdsRef.getVersion());
+ assertEquals("A12347", cdsRef.getAccessionId());
+
+ /*
+ * Verify there are mappings from each exon sequence to its protein product
+ * and also to its dna source
+ */
+ Iterator<AlignedCodonFrame> newMappingsIterator = mappings.iterator();
+
+ // mappings for dna1 - exon1 - pep1
+ AlignedCodonFrame exonMapping = newMappingsIterator.next();
+ List<Mapping> dnaMappings = exonMapping.getMappingsForSequence(dna1);
+ assertEquals(1, dnaMappings.size());
+ assertSame(exons.get(0).getDatasetSequence(), dnaMappings.get(0)
+ .getTo());
+ assertEquals("G(1) in CDS should map to G(4) in DNA", 4, dnaMappings
+ .get(0).getMap().getToPosition(1));
+ List<Mapping> peptideMappings = exonMapping
+ .getMappingsForSequence(pep1);
+ assertEquals(1, peptideMappings.size());
+ assertSame(pep1.getDatasetSequence(), peptideMappings.get(0).getTo());
+
+ // mappings for dna1 - exon2 - pep2
+ exonMapping = newMappingsIterator.next();
+ dnaMappings = exonMapping.getMappingsForSequence(dna1);
+ assertEquals(1, dnaMappings.size());
+ assertSame(exons.get(1).getDatasetSequence(), dnaMappings.get(0)
+ .getTo());
+ assertEquals("c(4) in CDS should map to c(7) in DNA", 7, dnaMappings
+ .get(0).getMap().getToPosition(4));
+ peptideMappings = exonMapping.getMappingsForSequence(pep2);
+ assertEquals(1, peptideMappings.size());
+ assertSame(pep2.getDatasetSequence(), peptideMappings.get(0).getTo());
+
+ // mappings for dna1 - exon3 - pep3
+ exonMapping = newMappingsIterator.next();
+ dnaMappings = exonMapping.getMappingsForSequence(dna1);
+ assertEquals(1, dnaMappings.size());
+ assertSame(exons.get(2).getDatasetSequence(), dnaMappings.get(0)
+ .getTo());
+ assertEquals("T(4) in CDS should map to T(10) in DNA", 10, dnaMappings
+ .get(0).getMap().getToPosition(4));
+ peptideMappings = exonMapping.getMappingsForSequence(pep3);
+ assertEquals(1, peptideMappings.size());
+ assertSame(pep3.getDatasetSequence(), peptideMappings.get(0).getTo());
+ }
+
+ @Test(groups = { "Functional" })
+ public void testIsMappable()
+ {
+ SequenceI dna1 = new Sequence("dna1", "cgCAGtgGT");
+ SequenceI aa1 = new Sequence("aa1", "RSG");
+ AlignmentI al1 = new Alignment(new SequenceI[] { dna1 });
+ AlignmentI al2 = new Alignment(new SequenceI[] { aa1 });
+
+ assertFalse(AlignmentUtils.isMappable(null, null));
+ assertFalse(AlignmentUtils.isMappable(al1, null));
+ assertFalse(AlignmentUtils.isMappable(null, al1));
+ assertFalse(AlignmentUtils.isMappable(al1, al1));
+ assertFalse(AlignmentUtils.isMappable(al2, al2));
+
+ assertTrue(AlignmentUtils.isMappable(al1, al2));
+ assertTrue(AlignmentUtils.isMappable(al2, al1));
+ }
+
+ /**
+ * Test creating a mapping when the sequences involved do not start at residue
+ * 1
+ *
+ * @throws IOException
+ */
+ @Test(groups = { "Functional" })
+ public void testMapProteinSequenceToCdna_forSubsequence()
+ throws IOException
+ {
+ SequenceI prot = new Sequence("UNIPROT|V12345", "E-I--Q", 10, 12);
+ prot.createDatasetSequence();
+
+ SequenceI dna = new Sequence("EMBL|A33333", "GAA--AT-C-CAG", 40, 48);
+ dna.createDatasetSequence();
+
+ MapList map = AlignmentUtils.mapProteinSequenceToCdna(prot, dna);
+ assertEquals(10, map.getToLowest());
+ assertEquals(12, map.getToHighest());
+ assertEquals(40, map.getFromLowest());
+ assertEquals(48, map.getFromHighest());
}
}