test/jalview/ext/ensembl/EnsemblCdnaTest.java

   1 package jalview.ext.ensembl;
   2
   3 import static org.testng.AssertJUnit.assertEquals;
   4 import static org.testng.AssertJUnit.assertFalse;
   5 import static org.testng.AssertJUnit.assertNull;
   6 import static org.testng.AssertJUnit.assertTrue;
   7
   8 import jalview.datamodel.SequenceDummy;
   9 import jalview.datamodel.SequenceFeature;
  10 import jalview.datamodel.SequenceI;
  11 import jalview.io.gff.SequenceOntologyFactory;
  12 import jalview.io.gff.SequenceOntologyLite;
  13 import jalview.util.MapList;
  14
  15 import java.util.List;
  16
  17 import org.testng.Assert;
  18 import org.testng.annotations.AfterClass;
  19 import org.testng.annotations.BeforeClass;
  20 import org.testng.annotations.Test;
  21
  22 public class EnsemblCdnaTest
  23 {
  24   @BeforeClass
  25   public void setUp()
  26   {
  27     SequenceOntologyFactory.setInstance(new SequenceOntologyLite());
  28   }
  29
  30   @AfterClass
  31   public void tearDown()
  32   {
  33     SequenceOntologyFactory.setInstance(null);
  34   }
  35   /**
  36    * Test that the cdna part of genomic sequence is correctly identified by
  37    * 'exon' features (or subtypes) - reverse strand case.
  38    */
  39   @Test(groups = "Functional")
  40   public void testGetGenomicRangesFromFeatures_reverseStrand()
  41   {
  42     EnsemblCdna testee = new EnsemblCdna();
  43     SequenceI genomic = new SequenceDummy("chr7");
  44     genomic.setStart(10000);
  45     genomic.setEnd(50000);
  46     String transcriptId = "ABC123";
  47
  48     // exon at (start+10000) length 501
  49     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
  50             null);
  51     sf.setValue("Parent", "transcript:" + transcriptId);
  52     sf.setStrand("-");
  53     genomic.addSequenceFeature(sf);
  54
  55     // exon (sub-type) at (start + exon_variant) length 101
  56     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
  57     sf.setValue("Parent", "transcript:" + transcriptId);
  58     sf.setStrand("-");
  59     genomic.addSequenceFeature(sf);
  60
  61     // exon belonging to a different transcript doesn't count
  62     sf = new SequenceFeature("exon", "", 11500, 12600, 0f, null);
  63     sf.setValue("Parent", "transcript:anotherOne");
  64     genomic.addSequenceFeature(sf);
  65
  66     // transcript feature doesn't count
  67     sf = new SequenceFeature("transcript", "", 10000, 50000, 0f, null);
  68     genomic.addSequenceFeature(sf);
  69
  70     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
  71             transcriptId, 23);
  72     List<int[]> fromRanges = ranges.getFromRanges();
  73     assertEquals(2, fromRanges.size());
  74
  75     /*
  76      * from ranges should be sorted by start order descending
  77      * and hold mappings from reverse strand sense
  78      */
  79     assertEquals(20500, fromRanges.get(0)[0]);
  80     assertEquals(20000, fromRanges.get(0)[1]);
  81     assertEquals(10600, fromRanges.get(1)[0]);
  82     assertEquals(10500, fromRanges.get(1)[1]);
  83     // to range should start from given start numbering
  84     List<int[]> toRanges = ranges.getToRanges();
  85     assertEquals(1, toRanges.size());
  86     assertEquals(23, toRanges.get(0)[0]);
  87     assertEquals(624, toRanges.get(0)[1]);
  88   }
  89
  90   /**
  91    * Test that the cdna part of genomic sequence is correctly identified by
  92    * 'exon' features (or subtypes) with the desired transcript as parent
  93    */
  94   @Test(groups = "Functional")
  95   public void testGetGenomicRangesFromFeatures()
  96   {
  97     EnsemblCdna testee = new EnsemblCdna();
  98     SequenceI genomic = new SequenceDummy("chr7");
  99     genomic.setStart(10000);
 100     genomic.setEnd(50000);
 101     String transcriptId = "ABC123";
 102
 103     // exon at (start+10000) length 501
 104     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
 105             null);
 106     sf.setValue("Parent", "transcript:" + transcriptId);
 107     sf.setStrand("+");
 108     genomic.addSequenceFeature(sf);
 109
 110     // exon (sub-type) at (start + exon_variant) length 101
 111     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
 112     sf.setValue("Parent", "transcript:" + transcriptId);
 113     sf.setStrand("+");
 114     genomic.addSequenceFeature(sf);
 115
 116     // exon belonging to a different transcript doesn't count
 117     sf = new SequenceFeature("exon", "", 11500, 12600, 0f, null);
 118     sf.setValue("Parent", "transcript:anotherOne");
 119     genomic.addSequenceFeature(sf);
 120
 121     // transcript feature doesn't count
 122     sf = new SequenceFeature("transcript", "", 10000, 50000, 0f, null);
 123     sf.setStrand("-"); // weird but ignored
 124     genomic.addSequenceFeature(sf);
 125
 126     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 127             transcriptId, 23);
 128     List<int[]> fromRanges = ranges.getFromRanges();
 129     assertEquals(2, fromRanges.size());
 130     // from ranges should be sorted by start order
 131     assertEquals(10500, fromRanges.get(0)[0]);
 132     assertEquals(10600, fromRanges.get(0)[1]);
 133     assertEquals(20000, fromRanges.get(1)[0]);
 134     assertEquals(20500, fromRanges.get(1)[1]);
 135     // to range should start from given start numbering
 136     List<int[]> toRanges = ranges.getToRanges();
 137     assertEquals(1, toRanges.size());
 138     assertEquals(23, toRanges.get(0)[0]);
 139     assertEquals(624, toRanges.get(0)[1]);
 140   }
 141
 142   /**
 143    * The method under test should give up and return null if both forward and
 144    * reverse strands are present in the features of interest
 145    */
 146   @Test(groups = "Functional")
 147   public void testGetGenomicRangesFromFeatures_mixedStrand()
 148   {
 149     EnsemblCdna testee = new EnsemblCdna();
 150     SequenceI genomic = new SequenceDummy("chr7");
 151     genomic.setStart(10000);
 152     genomic.setEnd(50000);
 153     String transcriptId = "ABC123";
 154
 155     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
 156             null);
 157     sf.setValue("Parent", "transcript:" + transcriptId);
 158     sf.setStrand("-");
 159     genomic.addSequenceFeature(sf);
 160
 161     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
 162     sf.setValue("Parent", "transcript:" + transcriptId);
 163     sf.setStrand("+");
 164     genomic.addSequenceFeature(sf);
 165
 166     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 167             transcriptId, 23);
 168     assertNull(ranges);
 169   }
 170
 171   /**
 172    * Test the method that retains features except for 'transcript' (or
 173    * subtypes), or features with parent other than the given id
 174    */
 175   @Test(groups = "Functional")
 176   public void testRetainFeature()
 177   {
 178     String accId = "ABC123";
 179     EnsemblCdna testee = new EnsemblCdna();
 180
 181     SequenceFeature sf = new SequenceFeature("transcript", "", 20000,
 182             20500, 0f, null);
 183     assertFalse(testee.retainFeature(sf, accId));
 184
 185     sf.setType("aberrant_processed_transcript");
 186     assertFalse(testee.retainFeature(sf, accId));
 187
 188     sf.setType("NMD_transcript_variant");
 189     assertFalse(testee.retainFeature(sf, accId));
 190
 191     // other feature with no parent is retained
 192     sf.setType("sequence_variant");
 193     assertTrue(testee.retainFeature(sf, accId));
 194
 195     // other feature with desired parent is retained
 196     sf.setValue("Parent", "transcript:" + accId);
 197     assertTrue(testee.retainFeature(sf, accId));
 198
 199     // feature with wrong parent is not retained
 200     sf.setValue("Parent", "transcript:XYZ");
 201     assertFalse(testee.retainFeature(sf, accId));
 202   }
 203
 204   /**
 205    * Test the method that picks out 'exon' (or subtype) features with the
 206    * accession id as parent
 207    */
 208   @Test(groups = "Functional")
 209   public void testIdentifiesSequence()
 210   {
 211     String accId = "ABC123";
 212     EnsemblCdna testee = new EnsemblCdna();
 213
 214     // exon with no parent not valid
 215     SequenceFeature sf = new SequenceFeature("exon", "", 1, 2, 0f, null);
 216     assertFalse(testee.identifiesSequence(sf, accId));
 217
 218     // exon with wrong parent not valid
 219     sf.setValue("Parent", "transcript:XYZ");
 220     assertFalse(testee.identifiesSequence(sf, accId));
 221
 222     // exon with right parent is valid
 223     sf.setValue("Parent", "transcript:" + accId);
 224     assertTrue(testee.identifiesSequence(sf, accId));
 225
 226     // exon sub-type with right parent is valid
 227     sf.setType("coding_exon");
 228     assertTrue(testee.identifiesSequence(sf, accId));
 229
 230     // transcript not valid:
 231     sf.setType("transcript");
 232     assertFalse(testee.identifiesSequence(sf, accId));
 233
 234     // CDS not valid:
 235     sf.setType("CDS");
 236     assertFalse(testee.identifiesSequence(sf, accId));
 237   }
 238
 239   @Test(groups = "Functional")
 240   public void testIsValidReference() throws Exception
 241   {
 242     EnsemblSequenceFetcher esq = new EnsemblCdna();
 243     Assert.assertTrue(esq.isValidReference("CCDS5863.1"));
 244     Assert.assertTrue(esq.isValidReference("ENST00000288602"));
 245     Assert.assertTrue(esq.isValidReference("ENSG00000288602"));
 246     Assert.assertFalse(esq.isValidReference("ENSP00000288602"));
 247     Assert.assertFalse(esq.isValidReference("ENST0000288602"));
 248     // non-human species having a 3 character identifier included:
 249     Assert.assertTrue(esq.isValidReference("ENSMUSG00000099398"));
 250   }
 251 }