test/jalview/ext/ensembl/EnsemblCdnaTest.java

   1 /*
   2  * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
   3  * Copyright (C) $$Year-Rel$$ The Jalview Authors
   4  *
   5  * This file is part of Jalview.
   6  *
   7  * Jalview is free software: you can redistribute it and/or
   8  * modify it under the terms of the GNU General Public License
   9  * as published by the Free Software Foundation, either version 3
  10  * of the License, or (at your option) any later version.
  11  *
  12  * Jalview is distributed in the hope that it will be useful, but
  13  * WITHOUT ANY WARRANTY; without even the implied warranty
  14  * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
  15  * PURPOSE.  See the GNU General Public License for more details.
  16  *
  17  * You should have received a copy of the GNU General Public License
  18  * along with Jalview.  If not, see <http://www.gnu.org/licenses/>.
  19  * The Jalview Authors are detailed in the 'AUTHORS' file.
  20  */
  21 package jalview.ext.ensembl;
  22
  23 import static org.testng.AssertJUnit.assertEquals;
  24 import static org.testng.AssertJUnit.assertFalse;
  25 import static org.testng.AssertJUnit.assertNull;
  26 import static org.testng.AssertJUnit.assertTrue;
  27
  28 import jalview.datamodel.Sequence;
  29 import jalview.datamodel.SequenceDummy;
  30 import jalview.datamodel.SequenceFeature;
  31 import jalview.datamodel.SequenceI;
  32 import jalview.gui.JvOptionPane;
  33 import jalview.io.gff.SequenceOntologyFactory;
  34 import jalview.io.gff.SequenceOntologyLite;
  35 import jalview.util.MapList;
  36
  37 import java.util.List;
  38
  39 import org.testng.Assert;
  40 import org.testng.annotations.AfterClass;
  41 import org.testng.annotations.BeforeClass;
  42 import org.testng.annotations.Test;
  43
  44 public class EnsemblCdnaTest
  45 {
  46
  47   @BeforeClass(alwaysRun = true)
  48   public void setUpJvOptionPane()
  49   {
  50     JvOptionPane.setInteractiveMode(false);
  51     JvOptionPane.setMockResponse(JvOptionPane.CANCEL_OPTION);
  52   }
  53
  54   @BeforeClass(alwaysRun = true)
  55   public void setUp()
  56   {
  57     SequenceOntologyFactory.setInstance(new SequenceOntologyLite());
  58   }
  59
  60   @AfterClass(alwaysRun = true)
  61   public void tearDown()
  62   {
  63     SequenceOntologyFactory.setInstance(null);
  64   }
  65
  66   /**
  67    * Test that the cdna part of genomic sequence is correctly identified by
  68    * 'exon' features (or subtypes) - reverse strand case.
  69    */
  70   @Test(groups = "Functional")
  71   public void testGetGenomicRangesFromFeatures_reverseStrand()
  72   {
  73     EnsemblCdna testee = new EnsemblCdna();
  74     SequenceI genomic = new SequenceDummy("chr7");
  75     genomic.setStart(10000);
  76     genomic.setEnd(50000);
  77     String transcriptId = "ABC123";
  78
  79     // exon at (start+10000) length 501
  80     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
  81             null);
  82     sf.setValue("Parent", "transcript:" + transcriptId);
  83     sf.setStrand("-");
  84     genomic.addSequenceFeature(sf);
  85
  86     // exon (sub-type) at (start + exon_variant) length 101
  87     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
  88     sf.setValue("Parent", "transcript:" + transcriptId);
  89     sf.setStrand("-");
  90     genomic.addSequenceFeature(sf);
  91
  92     // exon belonging to a different transcript doesn't count
  93     sf = new SequenceFeature("exon", "", 11500, 12600, 0f, null);
  94     sf.setValue("Parent", "transcript:anotherOne");
  95     genomic.addSequenceFeature(sf);
  96
  97     // transcript feature doesn't count
  98     sf = new SequenceFeature("transcript", "", 10000, 50000, 0f, null);
  99     genomic.addSequenceFeature(sf);
 100
 101     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 102             transcriptId, 23);
 103     List<int[]> fromRanges = ranges.getFromRanges();
 104     assertEquals(2, fromRanges.size());
 105
 106     /*
 107      * from ranges should be sorted by start order descending
 108      * and hold mappings from reverse strand sense
 109      */
 110     assertEquals(20500, fromRanges.get(0)[0]);
 111     assertEquals(20000, fromRanges.get(0)[1]);
 112     assertEquals(10600, fromRanges.get(1)[0]);
 113     assertEquals(10500, fromRanges.get(1)[1]);
 114     // to range should start from given start numbering
 115     List<int[]> toRanges = ranges.getToRanges();
 116     assertEquals(1, toRanges.size());
 117     assertEquals(23, toRanges.get(0)[0]);
 118     assertEquals(624, toRanges.get(0)[1]);
 119   }
 120
 121   /**
 122    * Test that the cdna part of genomic sequence is correctly identified by
 123    * 'exon' features (or subtypes) with the desired transcript as parent
 124    */
 125   @Test(groups = "Functional")
 126   public void testGetGenomicRangesFromFeatures()
 127   {
 128     EnsemblCdna testee = new EnsemblCdna();
 129     SequenceI genomic = new SequenceDummy("chr7");
 130     genomic.setStart(10000);
 131     genomic.setEnd(50000);
 132     String transcriptId = "ABC123";
 133
 134     // exon at (start+10000) length 501
 135     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
 136             null);
 137     sf.setValue("Parent", "transcript:" + transcriptId);
 138     sf.setStrand("+");
 139     genomic.addSequenceFeature(sf);
 140
 141     // exon (sub-type) at (start + exon_variant) length 101
 142     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
 143     sf.setValue("Parent", "transcript:" + transcriptId);
 144     sf.setStrand("+");
 145     genomic.addSequenceFeature(sf);
 146
 147     // exon belonging to a different transcript doesn't count
 148     sf = new SequenceFeature("exon", "", 11500, 12600, 0f, null);
 149     sf.setValue("Parent", "transcript:anotherOne");
 150     genomic.addSequenceFeature(sf);
 151
 152     // transcript feature doesn't count
 153     sf = new SequenceFeature("transcript", "", 10000, 50000, 0f, null);
 154     sf.setStrand("-"); // weird but ignored
 155     genomic.addSequenceFeature(sf);
 156
 157     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 158             transcriptId, 23);
 159     List<int[]> fromRanges = ranges.getFromRanges();
 160     assertEquals(2, fromRanges.size());
 161     // from ranges should be sorted by start order
 162     assertEquals(10500, fromRanges.get(0)[0]);
 163     assertEquals(10600, fromRanges.get(0)[1]);
 164     assertEquals(20000, fromRanges.get(1)[0]);
 165     assertEquals(20500, fromRanges.get(1)[1]);
 166     // to range should start from given start numbering
 167     List<int[]> toRanges = ranges.getToRanges();
 168     assertEquals(1, toRanges.size());
 169     assertEquals(23, toRanges.get(0)[0]);
 170     assertEquals(624, toRanges.get(0)[1]);
 171   }
 172
 173   /**
 174    * The method under test should give up and return null if both forward and
 175    * reverse strands are present in the features of interest
 176    */
 177   @Test(groups = "Functional")
 178   public void testGetGenomicRangesFromFeatures_mixedStrand()
 179   {
 180     EnsemblCdna testee = new EnsemblCdna();
 181     SequenceI genomic = new SequenceDummy("chr7");
 182     genomic.setStart(10000);
 183     genomic.setEnd(50000);
 184     String transcriptId = "ABC123";
 185
 186     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
 187             null);
 188     sf.setValue("Parent", "transcript:" + transcriptId);
 189     sf.setStrand("-");
 190     genomic.addSequenceFeature(sf);
 191
 192     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
 193     sf.setValue("Parent", "transcript:" + transcriptId);
 194     sf.setStrand("+");
 195     genomic.addSequenceFeature(sf);
 196
 197     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 198             transcriptId, 23);
 199     assertNull(ranges);
 200   }
 201
 202   /**
 203    * Test the method that retains features except for 'transcript' (or
 204    * subtypes), or features with parent other than the given id
 205    */
 206   @Test(groups = "Functional")
 207   public void testRetainFeature()
 208   {
 209     String accId = "ABC123";
 210     EnsemblCdna testee = new EnsemblCdna();
 211
 212     SequenceFeature sf = new SequenceFeature("transcript", "", 20000,
 213             20500, 0f, null);
 214     assertFalse(testee.retainFeature(sf, accId));
 215
 216     sf = new SequenceFeature("aberrant_processed_transcript", "", 20000,
 217             20500, 0f, null);
 218     assertFalse(testee.retainFeature(sf, accId));
 219
 220     sf = new SequenceFeature("NMD_transcript_variant", "", 20000, 20500,
 221             0f, null);
 222     assertFalse(testee.retainFeature(sf, accId));
 223
 224     // other feature with no parent is retained
 225     sf = new SequenceFeature("sequence_variant", "", 20000, 20500, 0f, null);
 226     assertTrue(testee.retainFeature(sf, accId));
 227
 228     // other feature with desired parent is retained
 229     sf.setValue("Parent", "transcript:" + accId);
 230     assertTrue(testee.retainFeature(sf, accId));
 231
 232     // test is not case-sensitive
 233     assertTrue(testee.retainFeature(sf, accId.toLowerCase()));
 234
 235     // feature with wrong parent is not retained
 236     sf.setValue("Parent", "transcript:XYZ");
 237     assertFalse(testee.retainFeature(sf, accId));
 238   }
 239
 240   /**
 241    * Test the method that picks out 'exon' (or subtype) features with the
 242    * accession id as parent
 243    */
 244   @Test(groups = "Functional")
 245   public void testGetIdentifyingFeatures()
 246   {
 247     String accId = "ABC123";
 248     SequenceI seq = new Sequence(accId, "MKLNFRQIE");
 249
 250     // exon with no parent: not valid
 251     SequenceFeature sf1 = new SequenceFeature("exon", "", 1, 2, 0f, null);
 252     seq.addSequenceFeature(sf1);
 253
 254     // exon with wrong parent: not valid
 255     SequenceFeature sf2 = new SequenceFeature("exon", "", 1, 2, 0f, null);
 256     sf2.setValue("Parent", "transcript:XYZ");
 257     seq.addSequenceFeature(sf2);
 258
 259     // exon with right parent is valid
 260     SequenceFeature sf3 = new SequenceFeature("exon", "", 1, 2, 0f, null);
 261     sf3.setValue("Parent", "transcript:" + accId);
 262     seq.addSequenceFeature(sf3);
 263
 264     // exon sub-type with right parent is valid
 265     SequenceFeature sf4 = new SequenceFeature("coding_exon", "", 1, 2, 0f,
 266             null);
 267     sf4.setValue("Parent", "transcript:" + accId);
 268     seq.addSequenceFeature(sf4);
 269
 270     // transcript not valid:
 271     SequenceFeature sf5 = new SequenceFeature("transcript", "", 1, 2, 0f,
 272             null);
 273     sf5.setValue("Parent", "transcript:" + accId);
 274     seq.addSequenceFeature(sf5);
 275
 276     // CDS not valid:
 277     SequenceFeature sf6 = new SequenceFeature("transcript", "", 1, 2, 0f,
 278             null);
 279     sf6.setValue("Parent", "transcript:" + accId);
 280     seq.addSequenceFeature(sf6);
 281
 282     List<SequenceFeature> sfs = new EnsemblCdna()
 283             .getIdentifyingFeatures(seq, accId);
 284     assertFalse(sfs.contains(sf1));
 285     assertFalse(sfs.contains(sf2));
 286     assertTrue(sfs.contains(sf3));
 287     assertTrue(sfs.contains(sf4));
 288     assertFalse(sfs.contains(sf5));
 289     assertFalse(sfs.contains(sf6));
 290   }
 291
 292   @Test(groups = "Functional")
 293   public void testIsValidReference() throws Exception
 294   {
 295     EnsemblSequenceFetcher esq = new EnsemblCdna();
 296     Assert.assertTrue(esq.isValidReference("CCDS5863.1"));
 297     Assert.assertTrue(esq.isValidReference("ENST00000288602"));
 298     Assert.assertTrue(esq.isValidReference("ENSG00000288602"));
 299     Assert.assertFalse(esq.isValidReference("ENSP00000288602"));
 300     Assert.assertFalse(esq.isValidReference("ENST0000288602"));
 301     // non-human species having a 3 character identifier included:
 302     Assert.assertTrue(esq.isValidReference("ENSMUSG00000099398"));
 303   }
 304 }