Skip to content

Commit

Permalink
Generating JSON file from VEP variant effects
Browse files Browse the repository at this point in the history
  • Loading branch information
Cristina Yenyxe Gonzalez Garcia committed May 7, 2014
1 parent bd67bed commit 3b92d0c
Show file tree
Hide file tree
Showing 6 changed files with 453 additions and 0 deletions.
4 changes: 4 additions & 0 deletions cellbase-build/pom.xml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,10 @@
<groupId>org.opencb.commons</groupId>
<artifactId>bioformats</artifactId>
</dependency>
<dependency>
<groupId>org.opencb.biodata</groupId>
<artifactId>models</artifactId>
</dependency>

<dependency>
<groupId>psidev.psi.mi</groupId>
Expand Down
260 changes: 260 additions & 0 deletions cellbase-build/src/main/java/org/opencb/cellbase/build/transform/VariantEffectParser.java
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,260 @@
package org.opencb.cellbase.build.transform;

import java.io.BufferedReader;
import java.io.FileInputStream;
import java.io.IOException;
import java.io.InputStreamReader;
import java.nio.file.Path;
import java.util.logging.Level;
import java.util.logging.Logger;
import org.apache.commons.lang.StringUtils;
import org.opencb.biodata.models.variant.effect.ConsequenceType;
import org.opencb.biodata.models.variant.effect.ConsequenceTypeMappings;
import org.opencb.biodata.models.variant.effect.ProteinSubstitutionScores;
import org.opencb.biodata.models.variant.effect.VariantEffect;
import org.opencb.cellbase.build.transform.serializers.json.JsonSerializer;


/**
*
* @author Cristina Yenyxe Gonzalez Garcia <cyenyxe@ebi.ac.uk>
*/
public class VariantEffectParser {

private JsonSerializer serializer;

public VariantEffectParser(JsonSerializer serializer) {
this.serializer = serializer;
}

public int parse(Path file) throws IOException {
BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream(file.toFile())));
String line = null;
VariantEffect currentEffect = null;
String currentAllele = null;

int numEffectsWritten = 0;

while((line = reader.readLine()) != null) {
if (line.startsWith("#")) {
continue; // Header will just be ignored
}

String[] fields = line.split("\t");
String[] positionFields = fields[0].split("[\\_\\/]");

if (positionFields.length < 4) {
// Only entries chr_pos_ref/alt will be parsed, ie, 1_909238_G/C or 3_361464_A/-
// Entries like 5_121187650_duplication will be ignored
continue;
}

if (isNewVariant(positionFields[0], Integer.parseInt(positionFields[1]), positionFields[2], positionFields[3], currentEffect, currentAllele)) {
if (currentEffect != null && serializer != null) {
if (serializer.write(currentEffect)) {
numEffectsWritten++;
}
}

currentEffect = new VariantEffect(positionFields[0], Integer.parseInt(positionFields[1]), Integer.parseInt(positionFields[1]), positionFields[2]);
currentAllele = positionFields[3];
} else if (isNewAllele(positionFields[0], Integer.parseInt(positionFields[1]), positionFields[2], positionFields[3], currentEffect, currentAllele)) {
currentAllele = positionFields[3];
}

parseLine(fields, currentEffect, currentAllele);
}

// Don't forget to serialize the last effect read!
if (currentEffect != null && serializer != null) {
if (serializer.write(currentEffect)) {
numEffectsWritten++;
}
}

return numEffectsWritten;
}

private boolean isNewVariant(String chromosome, int start, String referenceAllele, String alternateAllele,
VariantEffect current, String currentAllele) {
if (current == null) {
return true;
}

return !chromosome.equals(current.getChromosome())
|| start != current.getStart()
|| !referenceAllele.equals(current.getReferenceAllele())
// || !alternateAllele.equals(currentAllele)
;
}

private boolean isNewAllele(String chromosome, int start, String referenceAllele, String alternateAllele,
VariantEffect current, String currentAllele) {
if (current == null) {
return true;
}

return chromosome.equals(current.getChromosome())
&& start == current.getStart()
&& referenceAllele.equals(current.getReferenceAllele())
&& !alternateAllele.equals(currentAllele);
}

private void parseLine(String[] fields, VariantEffect effect, String alternateAllele) {
ConsequenceType ct = new ConsequenceType(alternateAllele);
effect.addConsequenceType(alternateAllele, ct);

// Gene and feature information
ct.setGeneId(fields[3]);
ct.setFeatureId(fields[4]);
ct.setFeatureType(fields[5]);

// List of consequence types as SO codes
String[] consequencesName = fields[6].split(",");
int[] consequencesSo = new int[consequencesName.length];
for (int i = 0; i < consequencesName.length; i++) {
Integer so = ConsequenceTypeMappings.termToAccession.get(consequencesName[i]);
if (so != null) {
consequencesSo[i] = so;
} else {
Logger.getLogger(JsonSerializer.class.getName()).log(Level.WARNING, "{0} is not a valid consequence type", consequencesName[i]);
}
}
ct.setConsequenceTypes(consequencesSo);

// Fields related to position can be empty (marked with "-")
if (!"-".equals(fields[7]) && StringUtils.isNumeric(fields[7])) {
ct.setcDnaPosition(Integer.parseInt(fields[7]));
}
if (!"-".equals(fields[8]) && StringUtils.isNumeric(fields[8])) {
ct.setCdsPosition(Integer.parseInt(fields[8]));
}
if (!"-".equals(fields[9]) && StringUtils.isNumeric(fields[9])) {
ct.setProteinPosition(Integer.parseInt(fields[9]));
}

// Fields related to AA and codon changes can also be empty (marked with "-")
if (!"-".equals(fields[10])) {
ct.setAminoacidChange(fields[10]);
}
if (!"-".equals(fields[11])) {
ct.setCodonChange(fields[11]);
}

// Variant ID
if (!"-".equals(fields[12])) {
ct.setVariationId(fields[12]);
}

parseExtraFields(fields[13], effect, ct);
}

private void parseExtraFields(String extra, VariantEffect effect, ConsequenceType ct) {
for (String field : extra.split(";")) {
String[] keyValue = field.split("=");

switch (keyValue[0].toLowerCase()) {
case "aa_maf":
effect.getFrequencies().setMafNhlbiEspAfricanAmerican(Float.parseFloat(keyValue[1]));
break;
case "afr_maf":
effect.getFrequencies().setMaf1000GAfrican(Float.parseFloat(keyValue[1]));
break;
case "amr_maf":
effect.getFrequencies().setMaf1000GAmerican(Float.parseFloat(keyValue[1]));
break;
case "asn_maf":
effect.getFrequencies().setMaf1000GAsian(Float.parseFloat(keyValue[1]));
break;
case "biotype":
ct.setFeatureBiotype(keyValue[1]);
break;
case "canonical":
ct.setCanonical(keyValue[1].equalsIgnoreCase("YES") || keyValue[1].equalsIgnoreCase("Y"));
break;
case "ccds":
ct.setCcdsId(keyValue[1]);
break;
case "cell_type":
effect.getRegulatoryEffect().setCellType(keyValue[1]);
break;
case "clin_sig":
ct.setClinicalSignificance(keyValue[1]);
break;
case "distance":
ct.setVariantToTranscriptDistance(Integer.parseInt(keyValue[1]));
break;
case "domains":
ct.setProteinDomains(keyValue[1]);
break;
case "ea_maf":
effect.getFrequencies().setMafNhlbiEspEuropeanAmerican(Float.parseFloat(keyValue[1]));
break;
case "ensp":
ct.setProteinId(keyValue[1]);
break;
case "eur_maf":
effect.getFrequencies().setMaf1000GEuropean(Float.parseFloat(keyValue[1]));
break;
case "exon":
ct.setExonNumber(keyValue[1]);
break;
case "gmaf": // Format is GMAF=G:0.2640
String[] gmafFields = keyValue[1].split(":");
effect.getFrequencies().setAllele1000g(gmafFields[0]);
effect.getFrequencies().setMaf1000G(Float.parseFloat(gmafFields[1]));
break;
case "hgvsc":
ct.setHgvsc(keyValue[1]);
break;
case "hgvsp":
ct.setHgvsp(keyValue[1]);
break;
case "high_inf_pos":
effect.getRegulatoryEffect().setHighInformationPosition(keyValue[1].equalsIgnoreCase("YES") || keyValue[1].equalsIgnoreCase("Y"));
break;
case "intron":
ct.setIntronNumber(keyValue[1]);
break;
case "motif_name":
effect.getRegulatoryEffect().setMotifName(keyValue[1]);
break;
case "motif_pos":
effect.getRegulatoryEffect().setMotifPosition(Integer.parseInt(keyValue[1]));
break;
case "motif_score_change":
effect.getRegulatoryEffect().setMotifScoreChange(Float.parseFloat(keyValue[1]));
break;
case "polyphen": // Format is PolyPhen=possibly_damaging(0.859)
String[] polyphenFields = keyValue[1].split("[\\(\\)]");
effect.getProteinSubstitutionScores().setPolyphenEffect(ProteinSubstitutionScores.PolyphenEffect.valueOf(polyphenFields[0].toUpperCase()));
effect.getProteinSubstitutionScores().setPolyphenScore(Float.parseFloat(polyphenFields[1]));
break;
case "pubmed":
ct.setPubmed(keyValue[1].split(","));
break;
case "sift": // Format is SIFT=tolerated(0.07)
String[] siftFields = keyValue[1].split("[\\(\\)]");
effect.getProteinSubstitutionScores().setSiftEffect(ProteinSubstitutionScores.SiftEffect.valueOf(siftFields[0].toUpperCase()));
effect.getProteinSubstitutionScores().setSiftScore(Float.parseFloat(siftFields[1]));
break;
case "strand":
ct.setFeatureStrand(keyValue[1]);
break;
case "sv":
ct.setStructuralVariantsId(keyValue[1].split(","));
break;
case "symbol":
ct.setGeneName(keyValue[1]);
break;
case "symbol_source":
ct.setGeneNameSource(keyValue[1]);
break;
default:
// ALLELE_NUM, FREQS, IND, ZYG
break;
}
}

}
}
113 changes: 113 additions & 0 deletions ...ld/src/main/java/org/opencb/cellbase/build/transform/serializers/json/JsonSerializer.java
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,113 @@
package org.opencb.cellbase.build.transform.serializers.json;

import com.fasterxml.jackson.core.JsonFactory;
import com.fasterxml.jackson.core.JsonGenerator;
import com.fasterxml.jackson.databind.ObjectMapper;
import java.io.FileOutputStream;
import java.io.IOException;
import java.io.OutputStream;
import java.nio.file.Path;
import java.nio.file.Paths;
import java.util.List;
import java.util.logging.Level;
import java.util.logging.Logger;
import java.util.zip.GZIPOutputStream;
import org.opencb.commons.io.DataWriter;

/**
*
* @author Cristina Yenyxe Gonzalez Garcia <cyenyxe@ebi.ac.uk>
*/
public class JsonSerializer<T> implements DataWriter<T> {

private Path outdir;
private Path file;

protected JsonFactory factory;
protected ObjectMapper jsonObjectMapper;
protected JsonGenerator generator;
private OutputStream stream;

public JsonSerializer(Path outdir, Path file) {
this.outdir = outdir;
this.file = file;
this.factory = new JsonFactory();
this.jsonObjectMapper = new ObjectMapper(this.factory);
}

@Override
public boolean open() {
try {
stream = new GZIPOutputStream(new FileOutputStream(
Paths.get(outdir.toString(), file.getFileName().toString()).toAbsolutePath().toString() + ".json.gz"));
} catch (IOException ex) {
Logger.getLogger(JsonSerializer.class.getName()).log(Level.SEVERE, null, ex);
return false;
}
return true;
}

@Override
public boolean pre() {
try {
generator = factory.createGenerator(stream);
} catch (IOException ex) {
Logger.getLogger(JsonSerializer.class.getName()).log(Level.SEVERE, null, ex);
return false;
}

return true;
}

@Override
public boolean write(T elem) {
try {
generator.writeObject(elem);
generator.writeRaw('\n');
} catch (IOException ex) {
Logger.getLogger(JsonSerializer.class.getName()).log(Level.SEVERE, elem.toString(), ex);
return false;
}

return true;
}

@Override
public boolean write(List<T> batch) {
for (T elem : batch) {
try {
generator.writeObject(elem);
generator.writeRaw('\n');
} catch (IOException ex) {
Logger.getLogger(JsonSerializer.class.getName()).log(Level.SEVERE, elem.toString(), ex);
return false;
}
}

return true;
}

@Override
public boolean post() {
try {
stream.flush();
generator.flush();
} catch (IOException ex) {
Logger.getLogger(JsonSerializer.class.getName()).log(Level.SEVERE, null, ex);
return false;
}
return true;
}

@Override
public boolean close() {
try {
generator.close();
} catch (IOException ex) {
Logger.getLogger(JsonSerializer.class.getName()).log(Level.SEVERE, null, ex);
return false;
}
return true;
}

}
Loading

0 comments on commit 3b92d0c

Please sign in to comment.