import static org.testng.AssertJUnit.assertEquals;
import static org.testng.AssertJUnit.assertFalse;
import static org.testng.AssertJUnit.assertNotNull;
+import static org.testng.AssertJUnit.assertNull;
import static org.testng.AssertJUnit.assertSame;
import static org.testng.AssertJUnit.assertTrue;
import static org.testng.internal.junit.ArrayAsserts.assertArrayEquals;
public class FeaturesFileTest
{
+ private static final String TAB = "\t";
private static String simpleGffFile = "examples/testdata/simpleGff3.gff";
@AfterClass(alwaysRun = true)
+ "ENDGROUP\tgrp2\n";
assertEquals(expected, exported);
}
+
+ /**
+ * Test parsing a features file with Jalview format features, including
+ * STARTMOTIFS/ENDMOTIFS lines with motifs to be matched to create features
+ *
+ * @throws Exception
+ */
+ @Test(groups = { "Functional" })
+ public void testParse_jalviewFeaturesWithMotifs() throws IOException
+ {
+ File f = new File("examples/uniref50.fa");
+ AlignmentI al = readAlignmentFile(f);
+ AlignFrame af = new AlignFrame(al, 500, 500);
+ Map<String, FeatureColourI> colours = af.getFeatureRenderer()
+ .getFeatureColours();
+
+ /*
+ * hide columns with YKV motif; these should not get
+ * matched by the Finder
+ */
+ al.getHiddenColumns().hideColumns(62, 64);
+
+ // @formatter:off
+ String featureData =
+ /*
+ * features in the null grup
+ */
+ "HELIX" + TAB + "blue\n" +
+ "MOTIF1" + TAB + "green\n" +
+ "MOTIF2" + TAB + "250,200,150|100,50,0|-3.9|4.5|above|-2.0\n" +
+ "adescription" + TAB + "FER_CAPAN" + TAB + "-1" + TAB + "42" + TAB + "45" + TAB + "HELIX\n" +
+ "STARTMOTIFS\n" +
+ "FLP" + TAB + "MOTIF1" + TAB + "flxMotifP\n" +
+ "F[LR]N" + TAB + "MOTIF1" + TAB + "flxMotifN\n" +
+ "fld" + TAB + "MOTIF1" + TAB + "flxMotifD\n" +
+ "YKV" + TAB + "MOTIF1" + TAB + "ykvMotif\n" +
+ "ENDMOTIFS\n" +
+ /*
+ * features in group uniprot
+ */
+ "STARTGROUP" + TAB + "uniprot\n" +
+ "bdescription" + TAB + "FER_CAPAN" + TAB + "-1" + TAB + "47" + TAB + "48" + TAB + "HELIX\n" +
+ "STARTMOTIFS\n" +
+ "FLG" + TAB + "MOTIF1" + TAB + "flxMotifG\n" +
+ "VTT" + TAB + "MOTIF2" + TAB + "vxtMotifT" + TAB + "-3.21\n" +
+ "VRT" + TAB + "MOTIF2" + TAB + "vxtMotifR\n" +
+ "ENDMOTIFS\n" +
+ "ENDGROUP";
+ // @formatter:on
+ FeaturesFile featuresFile = new FeaturesFile(featureData,
+ DataSourceType.PASTE);
+ assertTrue("Failed to parse features file",
+ featuresFile.parse(al, colours, true));
+
+ // verify HELIX features were parsed as normal
+ List<SequenceFeature> sfs = al.getSequenceAt(1).findFeatures(0, 999,
+ "HELIX");
+ assertEquals(2, sfs.size());
+ SequenceFeature sf = sfs.get(0);
+ assertNull(sf.getFeatureGroup());
+ assertEquals(42, sf.getBegin());
+ assertEquals(45, sf.getEnd());
+ assertEquals("adescription", sf.getDescription());
+ sf = sfs.get(1);
+ assertEquals("uniprot", sf.getFeatureGroup());
+ assertEquals(47, sf.getBegin());
+ assertEquals(48, sf.getEnd());
+ assertEquals("bdescription", sf.getDescription());
+
+ /*
+ * feature type MOTIF1
+ * FLP motif should match FER1_SOLLC/13-15 and Q93XJ9_SOLTU/13-15
+ * F[LR]N should match O80429_MAIZE/107-109
+ * fld should match nothing (as case sensitive)
+ * feature group should be null for the above
+ * FLG should match FER1_PEA/36-38, feature group uniprot
+ * YKV should match nothing as entirely within hidden columns
+ */
+ for (SequenceI seq : al.getSequences())
+ {
+ List<SequenceFeature> features = seq.findFeatures(0, 9999, "MOTIF1");
+ String name = seq.getName();
+ if (name.equals("FER1_SOLLC") || name.equals("Q93XJ9_SOLTU"))
+ {
+ assertEquals(1, features.size());
+ sf = features.get(0);
+ assertNull(sf.getFeatureGroup());
+ assertEquals(13, sf.getBegin());
+ assertEquals(15, sf.getEnd());
+ assertEquals("flxMotifP", sf.getDescription());
+ }
+ else if (name.equals("O80429_MAIZE"))
+ {
+ assertEquals(1, features.size());
+ sf = features.get(0);
+ assertNull(sf.getFeatureGroup());
+ assertEquals(107, sf.getBegin());
+ assertEquals(109, sf.getEnd());
+ assertEquals("flxMotifN", sf.getDescription());
+ }
+ else if (name.equals("FER1_PEA"))
+ {
+ assertEquals(1, features.size());
+ sf = features.get(0);
+ assertEquals("uniprot", sf.getFeatureGroup());
+ assertEquals(36, sf.getBegin());
+ assertEquals(38, sf.getEnd());
+ assertEquals("flxMotifG", sf.getDescription());
+ }
+ else
+ {
+ assertTrue("MOTIF1 features found for " + name, features.isEmpty());
+ }
+ }
+
+ /*
+ * feature type MOTIF2
+ * VTT motif should match FER1_PEA/26-28
+ * VRT should match nothing
+ */
+ for (SequenceI seq : al.getSequences())
+ {
+ List<SequenceFeature> features = seq.findFeatures(0, 9999, "MOTIF2");
+ String name = seq.getName();
+ if (name.equals("FER1_PEA"))
+ {
+ assertEquals(1, features.size());
+ sf = features.get(0);
+ assertEquals("uniprot", sf.getFeatureGroup());
+ assertEquals(26, sf.getBegin());
+ assertEquals(28, sf.getEnd());
+ assertEquals("vxtMotifT", sf.getDescription());
+ assertEquals(-3.21f, sf.getScore());
+ }
+ else
+ {
+ assertTrue("MOTIF2 features found for " + name, features.isEmpty());
+ assertTrue(features.isEmpty());
+ }
+ }
+ }
}