import jalview.datamodel.SequenceFeature;
import jalview.datamodel.SequenceGroup;
import jalview.datamodel.SequenceI;
+import jalview.io.gff.SequenceOntology;
import jalview.schemes.ResidueProperties;
import jalview.util.DBRefUtils;
import jalview.util.MapList;
/*
* transfer any features on dna that overlap the CDS
*/
- transferFeatures(dnaSeq, cds, dnaToCds, "CDS" /* SequenceOntology.CDS */);
+ transferFeatures(dnaSeq, cds, dnaToCds, null, "CDS" /* SequenceOntology.CDS */);
}
return cdsSequences;
}
/**
- * Transfers any co-located features on 'fromSeq' to 'toSeq', adjusting the
+ * Transfers co-located features on 'fromSeq' to 'toSeq', adjusting the
* feature start/end ranges, optionally omitting specified feature types.
+ * Returns the number of features copied.
*
* @param fromSeq
* @param toSeq
+ * @param select
+ * if not null, only features of this type are copied (including
+ * subtypes in the Sequence Ontology)
* @param mapping
* the mapping from 'fromSeq' to 'toSeq'
* @param omitting
*/
- protected static void transferFeatures(SequenceI fromSeq,
- SequenceI toSeq, MapList mapping, String... omitting)
+ public static int transferFeatures(SequenceI fromSeq, SequenceI toSeq,
+ MapList mapping, String select, String... omitting)
{
SequenceI copyTo = toSeq;
while (copyTo.getDatasetSequence() != null)
copyTo = copyTo.getDatasetSequence();
}
+ SequenceOntology so = SequenceOntology.getInstance();
+ int count = 0;
SequenceFeature[] sfs = fromSeq.getSequenceFeatures();
if (sfs != null)
{
for (SequenceFeature sf : sfs)
{
String type = sf.getType();
+ if (select != null && !so.isA(type, select))
+ {
+ continue;
+ }
boolean omit = false;
for (String toOmit : omitting)
{
* locate the mapped range - null if either start or end is
* not mapped (no partial overlaps are calculated)
*/
- int[] mappedTo = mapping.locateInTo(sf.getBegin(), sf.getEnd());
+ int start = sf.getBegin();
+ int end = sf.getEnd();
+ int[] mappedTo = mapping.locateInTo(start, end);
+ /*
+ * if whole exon range doesn't map, try interpreting it
+ * as 5' or 3' exon overlapping the CDS range
+ */
+ if (mappedTo == null)
+ {
+ mappedTo = mapping.locateInTo(end, end);
+ if (mappedTo != null)
+ {
+ /*
+ * end of exon is in CDS range - 5' overlap
+ * to a range from the start of the peptide
+ */
+ mappedTo[0] = 1;
+ }
+ }
+ if (mappedTo == null)
+ {
+ mappedTo = mapping.locateInTo(start, start);
+ if (mappedTo != null)
+ {
+ /*
+ * start of exon is in CDS range - 3' overlap
+ * to a range up to the end of the peptide
+ */
+ mappedTo[1] = toSeq.getLength();
+ }
+ }
if (mappedTo != null)
{
SequenceFeature copy = new SequenceFeature(sf);
copy.setBegin(Math.min(mappedTo[0], mappedTo[1]));
copy.setEnd(Math.max(mappedTo[0], mappedTo[1]));
copyTo.addSequenceFeature(copy);
+ count++;
}
}
}
+ return count;
}
/**
package jalview.ext.ensembl;
+import jalview.analysis.AlignmentUtils;
import jalview.datamodel.Alignment;
import jalview.datamodel.AlignmentI;
import jalview.datamodel.DBRefEntry;
peptide = peptide.getDatasetSequence();
}
- mapExonFeaturesToProtein(dnaSeq, peptide, dnaToProtein);
+ AlignmentUtils.transferFeatures(dnaSeq, peptide, dnaToProtein,
+ SequenceOntology.EXON);
LinkedHashMap<Integer, String[][]> variants = buildDnaVariantsMap(
dnaSeq, dnaToProtein);
}
/**
- * Transfers exon features to the corresponding mapped regions of the protein
- * sequence. This is useful because it allows visualisation of exon boundaries
- * on the peptide (using 'colour by label' for the exon name). Returns the
- * number of features written.
- *
- * @param dnaSeq
- * @param peptide
- * @param dnaToProtein
- */
- static int mapExonFeaturesToProtein(SequenceI dnaSeq, SequenceI peptide,
- MapList dnaToProtein)
- {
- SequenceFeature[] sfs = dnaSeq.getSequenceFeatures();
- if (sfs == null)
- {
- return 0;
- }
-
- SequenceOntology so = SequenceOntology.getInstance();
- int count = 0;
-
- for (SequenceFeature sf : sfs)
- {
- if (so.isA(sf.getType(), SequenceOntology.EXON))
- {
- int start = sf.getBegin();
- int end = sf.getEnd();
- int[] mapsTo = dnaToProtein.locateInTo(start, end);
- if (mapsTo != null)
- {
- SequenceFeature copy = new SequenceFeature(SequenceOntology.EXON,
- sf.getDescription(), mapsTo[0], mapsTo[1], 0f, null);
- peptide.addSequenceFeature(copy);
- count++;
- }
- }
- }
- return count;
- }
-
- /**
* Builds a map whose key is position in the protein sequence, and value is an
* array of all variants for the coding codon positions
*
"AAA---CCCTTT---");
}
+ /**
+ * Tests for transferring features between mapped sequences
+ */
@Test(groups = { "Functional" })
public void testTransferFeatures()
{
new int[] { 1, 6 }, 1, 1);
/*
- * behaviour of transferFeatures depends on MapList.locateInTo()
- * if start and end positions are mapped, returns the mapped region
- * if either is not mapped, does _not_ search for overlapped region
+ * transferFeatures() will build 'partial overlap' for regions
+ * that partially overlap 5' or 3' (start or end) of target sequence
*/
- AlignmentUtils.transferFeatures(dna, cds, map);
+ AlignmentUtils.transferFeatures(dna, cds, map, null);
SequenceFeature[] sfs = cds.getSequenceFeatures();
- assertEquals(4, sfs.length);
+ assertEquals(6, sfs.length);
SequenceFeature sf = sfs[0];
+ assertEquals("type2", sf.getType());
+ assertEquals("desc2", sf.getDescription());
+ assertEquals(2f, sf.getScore());
+ assertEquals(1, sf.getBegin());
+ assertEquals(1, sf.getEnd());
+
+ sf = sfs[1];
assertEquals("type3", sf.getType());
assertEquals("desc3", sf.getDescription());
assertEquals(3f, sf.getScore());
assertEquals(1, sf.getBegin());
assertEquals(3, sf.getEnd());
- sf = sfs[1];
+ sf = sfs[2];
assertEquals("type4", sf.getType());
assertEquals(2, sf.getBegin());
assertEquals(5, sf.getEnd());
- sf = sfs[2];
+ sf = sfs[3];
assertEquals("type5", sf.getType());
assertEquals(1, sf.getBegin());
assertEquals(6, sf.getEnd());
- sf = sfs[3];
+ sf = sfs[4];
assertEquals("type8", sf.getType());
assertEquals(6, sf.getBegin());
assertEquals(6, sf.getEnd());
+
+ sf = sfs[5];
+ assertEquals("type9", sf.getType());
+ assertEquals(6, sf.getBegin());
+ assertEquals(6, sf.getEnd());
+ }
+
+ /**
+ * Tests for transferring features between mapped sequences
+ */
+ @Test(groups = { "Functional" })
+ public void testTransferFeatures_withOmit()
+ {
+ SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
+ SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
+
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 6 }, 1, 1);
+
+ // [5, 11] maps to [2, 5]
+ dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
+ null));
+ // [4, 12] maps to [1, 6]
+ dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
+ null));
+ // [12, 12] maps to [6, 6]
+ dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
+ 8f, null));
+
+ // desc4 and desc8 are the 'omit these' varargs
+ AlignmentUtils.transferFeatures(dna, cds, map, null, "type4", "type8");
+ SequenceFeature[] sfs = cds.getSequenceFeatures();
+ assertEquals(1, sfs.length);
+
+ SequenceFeature sf = sfs[0];
+ assertEquals("type5", sf.getType());
+ assertEquals(1, sf.getBegin());
+ assertEquals(6, sf.getEnd());
+ }
+
+ /**
+ * Tests for transferring features between mapped sequences
+ */
+ @Test(groups = { "Functional" })
+ public void testTransferFeatures_withSelect()
+ {
+ SequenceI dna = new Sequence("dna/20-34", "acgTAGcaaGCCcgt");
+ SequenceI cds = new Sequence("cds/10-15", "TAGGCC");
+
+ MapList map = new MapList(new int[] { 4, 6, 10, 12 },
+ new int[] { 1, 6 }, 1, 1);
+
+ // [5, 11] maps to [2, 5]
+ dna.addSequenceFeature(new SequenceFeature("type4", "desc4", 5, 11, 4f,
+ null));
+ // [4, 12] maps to [1, 6]
+ dna.addSequenceFeature(new SequenceFeature("type5", "desc5", 4, 12, 5f,
+ null));
+ // [12, 12] maps to [6, 6]
+ dna.addSequenceFeature(new SequenceFeature("type8", "desc8", 12, 12,
+ 8f, null));
+
+ // "type5" is the 'select this type' argument
+ AlignmentUtils.transferFeatures(dna, cds, map, "type5");
+ SequenceFeature[] sfs = cds.getSequenceFeatures();
+ assertEquals(1, sfs.length);
+
+ SequenceFeature sf = sfs[0];
+ assertEquals("type5", sf.getType());
+ assertEquals(1, sf.getBegin());
+ assertEquals(6, sf.getEnd());
}
}
git://source.jalview.org / jalview.git / commitdiff