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 java.util.Locale;
  24
  25 import static org.testng.AssertJUnit.assertEquals;
  26 import static org.testng.AssertJUnit.assertFalse;
  27 import static org.testng.AssertJUnit.assertNull;
  28 import static org.testng.AssertJUnit.assertTrue;
  29
  30 import jalview.datamodel.Sequence;
  31 import jalview.datamodel.SequenceDummy;
  32 import jalview.datamodel.SequenceFeature;
  33 import jalview.datamodel.SequenceI;
  34 import jalview.gui.JvOptionPane;
  35 import jalview.io.gff.SequenceOntologyFactory;
  36 import jalview.io.gff.SequenceOntologyLite;
  37 import jalview.util.MapList;
  38
  39 import java.util.List;
  40
  41 import org.testng.Assert;
  42 import org.testng.annotations.AfterClass;
  43 import org.testng.annotations.BeforeClass;
  44 import org.testng.annotations.Test;
  45
  46 public class EnsemblCdnaTest
  47 {
  48
  49   @BeforeClass(alwaysRun = true)
  50   public void setUpJvOptionPane()
  51   {
  52     JvOptionPane.setInteractiveMode(false);
  53     JvOptionPane.setMockResponse(JvOptionPane.CANCEL_OPTION);
  54   }
  55
  56   @BeforeClass(alwaysRun = true)
  57   public void setUp()
  58   {
  59     SequenceOntologyFactory.setInstance(new SequenceOntologyLite());
  60   }
  61
  62   @AfterClass(alwaysRun = true)
  63   public void tearDown()
  64   {
  65     SequenceOntologyFactory.setInstance(null);
  66   }
  67
  68   /**
  69    * Test that the cdna part of genomic sequence is correctly identified by
  70    * 'exon' features (or subtypes) - reverse strand case.
  71    */
  72   @Test(groups = "Functional")
  73   public void testGetGenomicRangesFromFeatures_reverseStrand()
  74   {
  75     EnsemblCdna testee = new EnsemblCdna();
  76     SequenceI genomic = new SequenceDummy("chr7");
  77     genomic.setStart(10000);
  78     genomic.setEnd(50000);
  79     String transcriptId = "ABC123";
  80
  81     // exon at (start+10000) length 501
  82     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
  83             null);
  84     sf.setValue("Parent", transcriptId);
  85     sf.setStrand("-");
  86     genomic.addSequenceFeature(sf);
  87
  88     // exon (sub-type) at (start + exon_variant) length 101
  89     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
  90     sf.setValue("Parent", transcriptId);
  91     sf.setStrand("-");
  92     genomic.addSequenceFeature(sf);
  93
  94     // exon belonging to a different transcript doesn't count
  95     sf = new SequenceFeature("exon", "", 11500, 12600, 0f, null);
  96     sf.setValue("Parent", "anotherOne");
  97     genomic.addSequenceFeature(sf);
  98
  99     // transcript feature doesn't count
 100     sf = new SequenceFeature("transcript", "", 10000, 50000, 0f, null);
 101     genomic.addSequenceFeature(sf);
 102
 103     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 104             transcriptId, 23);
 105     List<int[]> fromRanges = ranges.getFromRanges();
 106     assertEquals(2, fromRanges.size());
 107
 108     /*
 109      * from ranges should be sorted by start order descending
 110      * and hold mappings from reverse strand sense
 111      */
 112     assertEquals(20500, fromRanges.get(0)[0]);
 113     assertEquals(20000, fromRanges.get(0)[1]);
 114     assertEquals(10600, fromRanges.get(1)[0]);
 115     assertEquals(10500, fromRanges.get(1)[1]);
 116     // to range should start from given start numbering
 117     List<int[]> toRanges = ranges.getToRanges();
 118     assertEquals(1, toRanges.size());
 119     assertEquals(23, toRanges.get(0)[0]);
 120     assertEquals(624, toRanges.get(0)[1]);
 121   }
 122
 123   /**
 124    * Test that the cdna part of genomic sequence is correctly identified by
 125    * 'exon' features (or subtypes) with the desired transcript as parent
 126    */
 127   @Test(groups = "Functional")
 128   public void testGetGenomicRangesFromFeatures()
 129   {
 130     EnsemblCdna testee = new EnsemblCdna();
 131     SequenceI genomic = new SequenceDummy("chr7");
 132     genomic.setStart(10000);
 133     genomic.setEnd(50000);
 134     String transcriptId = "ABC123";
 135
 136     // exon at (start+10000) length 501
 137     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
 138             null);
 139     sf.setValue("Parent", transcriptId);
 140     sf.setStrand("+");
 141     genomic.addSequenceFeature(sf);
 142
 143     // exon (sub-type) at (start + exon_variant) length 101
 144     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
 145     sf.setValue("Parent", transcriptId);
 146     sf.setStrand("+");
 147     genomic.addSequenceFeature(sf);
 148
 149     // exon belonging to a different transcript doesn't count
 150     sf = new SequenceFeature("exon", "", 11500, 12600, 0f, null);
 151     sf.setValue("Parent", "anotherOne");
 152     genomic.addSequenceFeature(sf);
 153
 154     // transcript feature doesn't count
 155     sf = new SequenceFeature("transcript", "", 10000, 50000, 0f, null);
 156     sf.setStrand("-"); // weird but ignored
 157     genomic.addSequenceFeature(sf);
 158
 159     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 160             transcriptId, 23);
 161     List<int[]> fromRanges = ranges.getFromRanges();
 162     assertEquals(2, fromRanges.size());
 163     // from ranges should be sorted by start order
 164     assertEquals(10500, fromRanges.get(0)[0]);
 165     assertEquals(10600, fromRanges.get(0)[1]);
 166     assertEquals(20000, fromRanges.get(1)[0]);
 167     assertEquals(20500, fromRanges.get(1)[1]);
 168     // to range should start from given start numbering
 169     List<int[]> toRanges = ranges.getToRanges();
 170     assertEquals(1, toRanges.size());
 171     assertEquals(23, toRanges.get(0)[0]);
 172     assertEquals(624, toRanges.get(0)[1]);
 173   }
 174
 175   /**
 176    * The method under test should give up and return null if both forward and
 177    * reverse strands are present in the features of interest
 178    */
 179   @Test(groups = "Functional")
 180   public void testGetGenomicRangesFromFeatures_mixedStrand()
 181   {
 182     EnsemblCdna testee = new EnsemblCdna();
 183     SequenceI genomic = new SequenceDummy("chr7");
 184     genomic.setStart(10000);
 185     genomic.setEnd(50000);
 186     String transcriptId = "ABC123";
 187
 188     SequenceFeature sf = new SequenceFeature("exon", "", 20000, 20500, 0f,
 189             null);
 190     sf.setValue("Parent", "transcript:" + transcriptId);
 191     sf.setStrand("-");
 192     genomic.addSequenceFeature(sf);
 193
 194     sf = new SequenceFeature("coding_exon", "", 10500, 10600, 0f, null);
 195     sf.setValue("Parent", "transcript:" + transcriptId);
 196     sf.setStrand("+");
 197     genomic.addSequenceFeature(sf);
 198
 199     MapList ranges = testee.getGenomicRangesFromFeatures(genomic,
 200             transcriptId, 23);
 201     assertNull(ranges);
 202   }
 203
 204   /**
 205    * Test the method that retains features except for 'transcript' (or
 206    * subtypes), or features with parent other than the given id
 207    */
 208   @Test(groups = "Functional")
 209   public void testRetainFeature()
 210   {
 211     String accId = "ABC123";
 212     EnsemblCdna testee = new EnsemblCdna();
 213
 214     SequenceFeature sf = new SequenceFeature("transcript", "", 20000,
 215             20500, 0f, null);
 216     assertFalse(testee.retainFeature(sf, accId));
 217
 218     sf = new SequenceFeature("aberrant_processed_transcript", "", 20000,
 219             20500, 0f, null);
 220     assertFalse(testee.retainFeature(sf, accId));
 221
 222     sf = new SequenceFeature("NMD_transcript_variant", "", 20000, 20500,
 223             0f, null);
 224     assertFalse(testee.retainFeature(sf, accId));
 225
 226     // other feature with no parent is retained
 227     sf = new SequenceFeature("sequence_variant", "", 20000, 20500, 0f, null);
 228     assertTrue(testee.retainFeature(sf, accId));
 229
 230     // other feature with desired parent is retained
 231     sf.setValue("Parent", accId);
 232     assertTrue(testee.retainFeature(sf, accId));
 233
 234     // test is not case-sensitive
 235     assertTrue(testee.retainFeature(sf, accId.toLowerCase(Locale.ROOT)));
 236
 237     // feature with wrong parent is not retained
 238     sf.setValue("Parent", "XYZ");
 239     assertFalse(testee.retainFeature(sf, accId));
 240   }
 241
 242   /**
 243    * Test the method that picks out 'exon' (or subtype) features with the
 244    * accession id as parent
 245    */
 246   @Test(groups = "Functional")
 247   public void testGetIdentifyingFeatures()
 248   {
 249     String accId = "ABC123";
 250     SequenceI seq = new Sequence(accId, "MKLNFRQIE");
 251
 252     // exon with no parent: not valid
 253     SequenceFeature sf1 = new SequenceFeature("exon", "", 1, 2, 0f, null);
 254     seq.addSequenceFeature(sf1);
 255
 256     // exon with wrong parent: not valid
 257     SequenceFeature sf2 = new SequenceFeature("exon", "", 1, 2, 0f, null);
 258     sf2.setValue("Parent", "XYZ");
 259     seq.addSequenceFeature(sf2);
 260
 261     // exon with right parent is valid
 262     SequenceFeature sf3 = new SequenceFeature("exon", "", 1, 2, 0f, null);
 263     sf3.setValue("Parent", accId);
 264     seq.addSequenceFeature(sf3);
 265
 266     // exon sub-type with right parent is valid
 267     SequenceFeature sf4 = new SequenceFeature("coding_exon", "", 1, 2, 0f,
 268             null);
 269     sf4.setValue("Parent", accId);
 270     seq.addSequenceFeature(sf4);
 271
 272     // transcript not valid:
 273     SequenceFeature sf5 = new SequenceFeature("transcript", "", 1, 2, 0f,
 274             null);
 275     sf5.setValue("Parent", accId);
 276     seq.addSequenceFeature(sf5);
 277
 278     // CDS not valid:
 279     SequenceFeature sf6 = new SequenceFeature("transcript", "", 1, 2, 0f,
 280             null);
 281     sf6.setValue("Parent", accId);
 282     seq.addSequenceFeature(sf6);
 283
 284     List<SequenceFeature> sfs = new EnsemblCdna()
 285             .getIdentifyingFeatures(seq, accId);
 286     assertFalse(sfs.contains(sf1));
 287     assertFalse(sfs.contains(sf2));
 288     assertTrue(sfs.contains(sf3));
 289     assertTrue(sfs.contains(sf4));
 290     assertFalse(sfs.contains(sf5));
 291     assertFalse(sfs.contains(sf6));
 292   }
 293
 294   @Test(groups = "Functional")
 295   public void testIsValidReference() throws Exception
 296   {
 297     EnsemblSequenceFetcher esq = new EnsemblCdna();
 298     Assert.assertTrue(esq.isValidReference("CCDS5863.1"));
 299     Assert.assertTrue(esq.isValidReference("ENST00000288602"));
 300     Assert.assertTrue(esq.isValidReference("ENSG00000288602"));
 301     Assert.assertFalse(esq.isValidReference("ENSP00000288602"));
 302     Assert.assertFalse(esq.isValidReference("ENST0000288602"));
 303     // non-human species having a 3 character identifier included:
 304     Assert.assertTrue(esq.isValidReference("ENSMUSG00000099398"));
 305   }
 306 }