import jalview.analysis.Dna;
import jalview.api.AlignViewControllerGuiI;
import jalview.bin.Cache;
-import jalview.datamodel.AlignmentI;
import jalview.datamodel.DBRefEntry;
import jalview.datamodel.GeneLociI;
import jalview.datamodel.Mapping;
import jalview.datamodel.features.FeatureSource;
import jalview.datamodel.features.FeatureSources;
import jalview.ext.ensembl.EnsemblMap;
+import jalview.ext.htsjdk.HtsContigDb;
import jalview.ext.htsjdk.VCFReader;
import jalview.io.gff.Gff3Helper;
import jalview.io.gff.SequenceOntologyI;
import jalview.util.MappingUtils;
import jalview.util.MessageManager;
+import java.io.File;
import java.io.IOException;
import java.util.ArrayList;
import java.util.HashMap;
private static final String EXCL = "!";
/*
- * the alignment we are associating VCF data with
+ * the VCF file we are processing
*/
- private AlignmentI al;
+ protected String vcfFilePath;
/*
* mappings between VCF and sequence reference assembly regions, as
*/
private Map<String, Map<int[], int[]>> assemblyMappings;
+ private VCFReader reader;
+
/*
* holds details of the VCF header lines (metadata)
*/
Map<Integer, String> vepFieldsOfInterest;
/**
- * Constructor given an alignment context
+ * Constructor given a VCF file
*
* @param alignment
*/
- public VCFLoader(AlignmentI alignment)
+ public VCFLoader(String vcfFile)
{
- al = alignment;
+ try
+ {
+ initialise(vcfFile);
+ } catch (IOException e)
+ {
+ System.err.println("Error opening VCF file: " + e.getMessage());
+ }
// map of species!chromosome!fromAssembly!toAssembly to {fromRange, toRange}
assemblyMappings = new HashMap<>();
}
/**
- * Starts a new thread to query and load VCF variant data on to the alignment
+ * Starts a new thread to query and load VCF variant data on to the given
+ * sequences
* <p>
* This method is not thread safe - concurrent threads should use separate
* instances of this class.
*
- * @param filePath
+ * @param seqs
* @param gui
*/
- public void loadVCF(final String filePath,
- final AlignViewControllerGuiI gui)
+ public void loadVCF(SequenceI[] seqs, final AlignViewControllerGuiI gui)
{
if (gui != null)
{
new Thread()
{
-
@Override
public void run()
{
- VCFLoader.this.doLoad(filePath, gui);
+ VCFLoader.this.doLoad(seqs, gui);
}
-
}.start();
}
/**
- * Loads VCF on to an alignment - provided it can be related to one or more
- * sequence's chromosomal coordinates
+ * Reads the specified contig sequence and adds its VCF variants to it
*
- * @param filePath
- * @param gui
- * optional callback handler for messages
+ * @param contig
+ * the id of a single sequence (contig) to load
+ * @return
*/
- protected void doLoad(String filePath, AlignViewControllerGuiI gui)
+ public SequenceI loadVCFContig(String contig)
{
- VCFReader reader = null;
- try
+ String ref = header.getOtherHeaderLine(VCFHeader.REFERENCE_KEY)
+ .getValue();
+ if (ref.startsWith("file://"))
{
- // long start = System.currentTimeMillis();
- reader = new VCFReader(filePath);
-
- header = reader.getFileHeader();
-
- try
- {
- dictionary = header.getSequenceDictionary();
- } catch (SAMException e)
- {
- // ignore - thrown if any contig line lacks length info
- }
+ ref = ref.substring(7);
+ }
- sourceId = filePath;
+ SequenceI seq = null;
+ File dbFile = new File(ref);
- saveMetadata(sourceId);
+ if (dbFile.exists())
+ {
+ HtsContigDb db = new HtsContigDb("", dbFile);
+ seq = db.getSequenceProxy(contig);
+ loadSequenceVCF(seq, ref);
+ db.close();
+ }
+ else
+ {
+ System.err.println("VCF reference not found: " + ref);
+ }
- /*
- * get offset of CSQ ALLELE_NUM and Feature if declared
- */
- parseCsqHeader();
+ return seq;
+ }
+ /**
+ * Loads VCF on to one or more sequences
+ *
+ * @param seqs
+ * @param gui
+ * optional callback handler for messages
+ */
+ protected void doLoad(SequenceI[] seqs, AlignViewControllerGuiI gui)
+ {
+ try
+ {
VCFHeaderLine ref = header
.getOtherHeaderLine(VCFHeader.REFERENCE_KEY);
String vcfAssembly = ref.getValue();
/*
* query for VCF overlapping each sequence in turn
*/
- for (SequenceI seq : al.getSequences())
+ for (SequenceI seq : seqs)
{
- int added = loadSequenceVCF(seq, reader, vcfAssembly);
+ int added = loadSequenceVCF(seq, vcfAssembly);
if (added > 0)
{
seqCount++;
}
if (gui != null)
{
- // long elapsed = System.currentTimeMillis() - start;
String msg = MessageManager.formatMessage("label.added_vcf",
varCount, seqCount);
gui.setStatus(msg);
}
/**
+ * Opens the VCF file and parses header data
+ *
+ * @param filePath
+ * @throws IOException
+ */
+ private void initialise(String filePath) throws IOException
+ {
+ vcfFilePath = filePath;
+
+ reader = new VCFReader(filePath);
+
+ header = reader.getFileHeader();
+
+ try
+ {
+ dictionary = header.getSequenceDictionary();
+ } catch (SAMException e)
+ {
+ // ignore - thrown if any contig line lacks length info
+ }
+
+ sourceId = filePath;
+
+ saveMetadata(sourceId);
+
+ /*
+ * get offset of CSQ ALLELE_NUM and Feature if declared
+ */
+ parseCsqHeader();
+ }
+
+ /**
* Reads metadata (such as INFO field descriptions and datatypes) and saves
* them for future reference
*
}
/**
- * Tries to add overlapping variants read from a VCF file to the given
- * sequence, and returns the number of variant features added. Note that this
- * requires the sequence to hold information as to its species, chromosomal
- * positions and reference assembly, in order to be able to map the VCF
- * variants to the sequence (or not)
+ * Tries to add overlapping variants read from a VCF file to the given sequence,
+ * and returns the number of variant features added
*
* @param seq
- * @param reader
* @param vcfAssembly
* @return
*/
- protected int loadSequenceVCF(SequenceI seq, VCFReader reader,
- String vcfAssembly)
+ protected int loadSequenceVCF(SequenceI seq, String vcfAssembly)
{
VCFMap vcfMap = getVcfMap(seq, vcfAssembly);
if (vcfMap == null)
{
dss = seq;
}
- return addVcfVariants(dss, reader, vcfMap, vcfAssembly);
+ return addVcfVariants(dss, vcfMap);
}
/**
* overlapping variants found.
*
* @param seq
- * @param reader
* @param map
* mapping from sequence to VCF coordinates
- * @param vcfAssembly
- * the '##reference' identifier for the VCF reference assembly
* @return
*/
- protected int addVcfVariants(SequenceI seq, VCFReader reader,
- VCFMap map, String vcfAssembly)
+ protected int addVcfVariants(SequenceI seq, VCFMap map)
{
boolean forwardStrand = map.map.isToForwardStrand();
String type = SequenceOntologyI.SEQUENCE_VARIANT;
if (consequence != null)
{
- type = getOntologyTerm(seq, variant, altAlleleIndex,
- consequence);
+ type = getOntologyTerm(consequence);
}
float score = getAlleleFrequency(variant, altAlleleIndex);
sf.setValue(Gff3Helper.ALLELES, alleles);
- addAlleleProperties(variant, seq, sf, altAlleleIndex, consequence);
+ addAlleleProperties(variant, sf, altAlleleIndex, consequence);
seq.addSequenceFeature(sf);
* <li>sequence id can be matched to VEP Feature (or SnpEff Feature_ID)</li>
* </ul>
*
- * @param seq
- * @param variant
- * @param altAlleleIndex
* @param consequence
* @return
* @see http://www.sequenceontology.org/browser/current_svn/term/SO:0001060
*/
- String getOntologyTerm(SequenceI seq, VariantContext variant,
- int altAlleleIndex, String consequence)
+ String getOntologyTerm(String consequence)
{
String type = SequenceOntologyI.SEQUENCE_VARIANT;
* Add any allele-specific VCF key-value data to the sequence feature
*
* @param variant
- * @param seq
* @param sf
* @param altAlelleIndex
* (0, 1..)
* if not null, the consequence specific to this sequence (transcript
* feature) and allele
*/
- protected void addAlleleProperties(VariantContext variant, SequenceI seq,
+ protected void addAlleleProperties(VariantContext variant,
SequenceFeature sf, final int altAlelleIndex, String consequence)
{
Map<String, Object> atts = variant.getAttributes();
*/
if (CSQ_FIELD.equals(key))
{
- addConsequences(variant, seq, sf, consequence);
+ addConsequences(variant, sf, consequence);
continue;
}
* transcript (sequence) being processed)
*
* @param variant
- * @param seq
* @param sf
* @param myConsequence
*/
- protected void addConsequences(VariantContext variant, SequenceI seq,
- SequenceFeature sf, String myConsequence)
+ protected void addConsequences(VariantContext variant, SequenceFeature sf,
+ String myConsequence)
{
Object value = variant.getAttribute(CSQ_FIELD);
// TODO if CSQ not present, try ANN (for SnpEff consequence data)?
Cache.loadProperties("test/jalview/io/testProps.jvprops");
Cache.setProperty("VCF_FIELDS", ".*");
Cache.setProperty("VEP_FIELDS", ".*");
+ Cache.initLogger();
}
@Test(groups = "Functional")
public void testDoLoad() throws IOException
{
AlignmentI al = buildAlignment();
- VCFLoader loader = new VCFLoader(al);
File f = makeVcf();
+ VCFLoader loader = new VCFLoader(f.getPath());
- loader.doLoad(f.getPath(), null);
+ loader.doLoad(al.getSequencesArray(), null);
/*
* verify variant feature(s) added to gene
{
AlignmentI al = buildAlignment();
- VCFLoader loader = new VCFLoader(al);
-
File f = makeVcf();
- loader.doLoad(f.getPath(), null);
+ VCFLoader loader = new VCFLoader(f.getPath());
+
+ loader.doLoad(al.getSequencesArray(), null);
/*
* verify variant feature(s) added to gene2
{
AlignmentI al = buildAlignment();
- VCFLoader loader = new VCFLoader(al);
+ VCFLoader loader = new VCFLoader("test/jalview/io/vcf/testVcf.vcf");
/*
* VCF data file with variants at gene3 positions
* 13 C/G, C/T
* 17 A/AC (insertion), A/G
*/
- loader.doLoad("test/jalview/io/vcf/testVcf.dat", null);
+ loader.doLoad(al.getSequencesArray(), null);
/*
* verify variant feature(s) added to gene3
assertEquals(map.size(), 9);
assertEquals(sf.getValueAsString("CSQ", "Feature"), "transcript4");
}
-}
+
+ /**
+ * A test that demonstrates loading a contig sequence from an indexed sequence
+ * database which is the reference for a VCF file
+ *
+ * @throws IOException
+ */
+ @Test(groups = "Functional")
+ public void testLoadVCFContig() throws IOException
+ {
+ VCFLoader loader = new VCFLoader(
+ "test/jalview/io/vcf/testVcf2.vcf");
+
+ SequenceI seq = loader.loadVCFContig("contig123");
+ assertEquals(seq.getLength(), 15);
+ assertEquals(seq.getSequenceAsString(), "AAAAACCCCCGGGGG");
+ List<SequenceFeature> features = seq.getSequenceFeatures();
+ SequenceFeatures.sortFeatures(features, true);
+ assertEquals(features.size(), 2);
+ SequenceFeature sf = features.get(0);
+ assertEquals(sf.getBegin(), 8);
+ assertEquals(sf.getEnd(), 8);
+ assertEquals(sf.getDescription(), "C,A");
+ sf = features.get(1);
+ assertEquals(sf.getBegin(), 12);
+ assertEquals(sf.getEnd(), 12);
+ assertEquals(sf.getDescription(), "G,T");
+
+ seq = loader.loadVCFContig("contig789");
+ assertEquals(seq.getLength(), 25);
+ assertEquals(seq.getSequenceAsString(), "GGGGGTTTTTAAAAACCCCCGGGGG");
+ features = seq.getSequenceFeatures();
+ SequenceFeatures.sortFeatures(features, true);
+ assertEquals(features.size(), 2);
+ sf = features.get(0);
+ assertEquals(sf.getBegin(), 2);
+ assertEquals(sf.getEnd(), 2);
+ assertEquals(sf.getDescription(), "G,T");
+ sf = features.get(1);
+ assertEquals(sf.getBegin(), 21);
+ assertEquals(sf.getEnd(), 21);
+ assertEquals(sf.getDescription(), "G,A");
+
+ seq = loader.loadVCFContig("contig456");
+ assertEquals(seq.getLength(), 20);
+ assertEquals(seq.getSequenceAsString(), "CCCCCGGGGGTTTTTAAAAA");
+ features = seq.getSequenceFeatures();
+ SequenceFeatures.sortFeatures(features, true);
+ assertEquals(features.size(), 1);
+ sf = features.get(0);
+ assertEquals(sf.getBegin(), 15);
+ assertEquals(sf.getEnd(), 15);
+ assertEquals(sf.getDescription(), "T,C");
+ }
+}
\ No newline at end of file
git://source.jalview.org / jalview.git / commitdiff