@Override
public List<QueryResult> getAllByRegionList(List<Region> regionList, QueryOptions options) {
// db.regulatory_region.find({"chunkIds": {$in:["1_200", "1_300"]}, "start": 601156})
QueryBuilder builder = new QueryBuilder();
List<Object> featureType = options.getList("featureType", null);
List<Object> featureClass = options.getList("featureClass", null);
// options = addExcludeReturnFields("chunkIds", options);
List<DBObject> queries = new ArrayList<>();
for (Region region : regionList) {
int firstChunkId = getChunkId(region.getStart(), regulatoryRegionChunkSize);
int lastChunkId = getChunkId(region.getEnd(), regulatoryRegionChunkSize);
BasicDBList chunksId = new BasicDBList();
for (int j = firstChunkId; j <= lastChunkId; j++) {
String chunkId =
region.getChromosome() + "_" + j + "_" + regulatoryRegionChunkSize / 1000 + "k";
chunksId.add(chunkId);
}
// logger.info(chunksId.toString());
builder =
builder
.start("_chunkIds")
.in(chunksId)
.and("start")
.lessThanEquals(region.getEnd())
.and("end")
.greaterThanEquals(region.getStart());
if (featureType != null && featureType.size() > 0) {
BasicDBList featureTypeDBList = new BasicDBList();
featureTypeDBList.addAll(featureType);
builder = builder.and("featureType").in(featureTypeDBList);
}
if (featureClass != null && featureClass.size() > 0) {
BasicDBList featureClassDBList = new BasicDBList();
featureClassDBList.addAll(featureClass);
builder = builder.and("featureClass").in(featureClassDBList);
}
queries.add(builder.get());
}
// System.out.println(">>"+regionList);
// System.out.println(">>"+builder.get().toString());
return executeQueryList2(regionList, queries, options);
}
public QueryResult getSimpleVariantsByRegion(
Region region, String sourceId, QueryOptions options) {
Long start, end, dbstart, dbend;
start = System.currentTimeMillis();
boolean includeStats;
boolean includeEffects;
if (!options.containsKey("stats") && !options.containsKey("effects")) {
includeStats = true;
includeEffects = true;
} else {
includeStats = options.containsKey("stats") && options.getBoolean("stats");
includeEffects = options.containsKey("effects") && options.getBoolean("effects");
}
QueryResult<Variant> queryResult =
new QueryResult<>(
String.format("%s:%d-%d", region.getChromosome(), region.getStart(), region.getEnd()));
List<Variant> results = new ArrayList<>();
String startRow = buildRowkey(region.getChromosome(), Long.toString(region.getStart()));
String stopRow = buildRowkey(region.getChromosome(), Long.toString(region.getEnd()));
BasicDBObject query =
new BasicDBObject("position", new BasicDBObject("$gte", startRow).append("$lte", stopRow))
.append("sources.sourceId", sourceId);
DBCollection collection = db.getCollection("variants");
dbstart = System.currentTimeMillis();
DBCursor variantInStudies = collection.find(query);
dbend = System.currentTimeMillis();
queryResult.setDbTime(dbend - dbstart);
for (DBObject result : variantInStudies) {
String[] rowkeyParts = result.get("position").toString().split("_");
String chromosome = rowkeyParts[0].replaceFirst("^0+(?!$)", "");
int position = Integer.parseInt(rowkeyParts[1]);
BasicDBList studies = (BasicDBList) result.get("sources");
BasicDBObject st = (BasicDBObject) studies.get(0);
String ref = (String) st.get("ref");
String alt = StringUtils.join((ArrayList<String>) st.get("alt"), ",");
// TODO Needs rework
Variant variant = new Variant(chromosome, position, position, ref, alt);
// Set stats informations
if (includeStats) {
VariantStats stats = new VariantStats();
BasicDBObject mongoStats = (BasicDBObject) st.get("stats");
stats.setMaf((float) (double) mongoStats.get("maf"));
stats.setMafAllele((String) mongoStats.get("alleleMaf"));
stats.setMissingGenotypes((int) mongoStats.get("missing"));
List<Genotype> genotypeCount = new ArrayList<>();
for (BasicDBObject s : (List<BasicDBObject>) mongoStats.get("genotypeCount")) {
for (Map.Entry<String, Object> entry : s.entrySet()) {
Genotype genotype = new Genotype(entry.getKey());
genotype.setCount((Integer) entry.getValue());
genotypeCount.add(genotype);
}
}
stats.setGenotypes(genotypeCount);
variant.setStats(stats);
}
// TODO Set consequence type names
if (includeEffects) {
BasicDBList mongoEffects = (BasicDBList) st.get("effects");
if (mongoEffects != null) {
for (Object e : mongoEffects) {
String effectObo = e.toString();
VariantEffect effect = new VariantEffect();
effect.setConsequenceTypeObo(effectObo);
variant.addEffect(effect);
}
}
}
results.add(variant);
}
queryResult.setResult(results);
queryResult.setNumResults(results.size());
end = System.currentTimeMillis();
queryResult.setTime(end - start);
return queryResult;
}
@Override
public QueryResult<Variant> getAllVariantsByRegionAndStudy(
Region region, String sourceId, QueryOptions options) {
Long start, end, dbstart, dbend;
start = System.currentTimeMillis();
QueryResult<Variant> queryResult =
new QueryResult<>(
String.format("%s:%d-%d", region.getChromosome(), region.getStart(), region.getEnd()));
List<Variant> results = new LinkedList<>();
boolean includeSamples;
boolean includeStats;
boolean includeEffects;
if (!options.containsKey("samples")
&& !options.containsKey("stats")
&& !options.containsKey("effects")) {
includeSamples = true;
includeStats = true;
includeEffects = true;
} else {
includeSamples = options.containsKey("samples") && options.getBoolean("samples");
includeStats = options.containsKey("stats") && options.getBoolean("stats");
includeEffects = options.containsKey("effects") && options.getBoolean("effects");
}
try {
String startRow = buildRowkey(region.getChromosome(), Long.toString(region.getStart()));
String stopRow = buildRowkey(region.getChromosome(), Long.toString(region.getEnd()));
HTable table = new HTable(admin.getConfiguration(), tableName);
dbstart = System.currentTimeMillis();
Scan regionScan = new Scan(startRow.getBytes(), stopRow.getBytes());
ResultScanner scanres = table.getScanner(regionScan);
dbend = System.currentTimeMillis();
queryResult.setDbTime(dbend - dbstart);
// Iterate over results and, optionally, their samples and statistics
for (Result result : scanres) {
String[] rowkeyParts = new String(result.getRow(), CHARSET_UTF_8).split("_");
String chromosome = rowkeyParts[0].replaceFirst("^0+(?!$)", "");
int position = Integer.parseInt(rowkeyParts[1]);
// Get basic result fields from Protocol Buffers message
NavigableMap<byte[], byte[]> infoMap = result.getFamilyMap("i".getBytes());
byte[] byteInfo = infoMap.get((sourceId + "_data").getBytes());
VariantFieldsProtos.VariantInfo protoInfo =
VariantFieldsProtos.VariantInfo.parseFrom(byteInfo);
String reference = protoInfo.getReference();
String alternate = StringUtils.join(protoInfo.getAlternateList(), ",");
String format = StringUtils.join(protoInfo.getFormatList(), ":");
Variant variant = new Variant(chromosome, position, position, reference, alternate);
// Set samples if requested
if (includeSamples) {
NavigableMap<byte[], byte[]> sampleMap = result.getFamilyMap("d".getBytes());
Map<String, Map<String, String>> resultSampleMap = new HashMap<>();
// Set samples
for (byte[] s : sampleMap.keySet()) {
String sampleName = (new String(s, CHARSET_UTF_8)).replaceAll(sourceId + "_", "");
VariantFieldsProtos.VariantSample sample =
VariantFieldsProtos.VariantSample.parseFrom(sampleMap.get(s));
String sample1 = sample.getSample();
String[] values = sample1.split(":");
String[] fields = format.split(":");
Map<String, String> singleSampleMap = new HashMap<>();
for (int i = 0; i < fields.length; i++) {
singleSampleMap.put(fields[i], values[i]);
}
// TODO
// variant.addSampleData(sampleName, singleSampleMap);
}
}
// Set stats if requested
if (includeStats) {
byte[] byteStats = infoMap.get((sourceId + "_stats").getBytes());
VariantFieldsProtos.VariantStats protoStats =
VariantFieldsProtos.VariantStats.parseFrom(byteStats);
VariantStats variantStats =
new VariantStats(
chromosome,
position,
reference,
alternate,
protoStats.getMaf(),
protoStats.getMgf(),
protoStats.getMafAllele(),
protoStats.getMgfGenotype(),
protoStats.getMissingAlleles(),
protoStats.getMissingGenotypes(),
protoStats.getMendelianErrors(),
protoStats.getIsIndel(),
protoStats.getCasesPercentDominant(),
protoStats.getControlsPercentDominant(),
protoStats.getCasesPercentRecessive(),
protoStats.getControlsPercentRecessive());
variant.setStats(variantStats);
}
if (includeEffects) {
QueryResult<VariantEffect> queryEffects = getEffectsByVariant(variant, options);
variant.setEffect(queryEffects.getResult());
}
results.add(variant);
}
} catch (IOException e) {
System.err.println(e.getClass().getName() + ": " + e.getMessage());
}
queryResult.setResult(results);
queryResult.setNumResults(results.size());
end = System.currentTimeMillis();
queryResult.setTime(end - start);
return queryResult;
}
@Override
public QueryResult getVariantsHistogramByRegion(
Region region, String sourceId, boolean histogramLogarithm, int histogramMax) {
QueryResult<ObjectMap> queryResult =
new QueryResult<>(
String.format("%s:%d-%d", region.getChromosome(), region.getStart(), region.getEnd()));
List<ObjectMap> data = new ArrayList<>();
String startRow = buildRowkey(region.getChromosome(), Long.toString(region.getStart()));
String stopRow = buildRowkey(region.getChromosome(), Long.toString(region.getEnd()));
long startTime = System.currentTimeMillis();
long startDbTime = System.currentTimeMillis();
BasicDBObject query =
new BasicDBObject("position", new BasicDBObject("$gte", startRow).append("$lte", stopRow))
.append("studies.studyId", sourceId);
DBCollection collection = db.getCollection("variants");
DBCursor queryResults = collection.find(query);
queryResult.setDbTime(System.currentTimeMillis() - startDbTime);
int resultSize = queryResults.size();
if (resultSize > histogramMax) { // Need to group results to fit maximum size of the histogram
int sumChunkSize = resultSize / histogramMax;
int i = 0, j = 0;
int featuresCount = 0;
ObjectMap item = null;
for (DBObject result : queryResults) {
// featuresCount += result.getInt("features_count");
// if (i == 0) {
// item = new ObjectMap("chromosome", result.getString("chromosome"));
// item.put("chunkId", result.getInt("chunk_id"));
// item.put("start", result.getInt("start"));
// } else if (i == sumChunkSize - 1 || j == resultSize - 1) {
// if (histogramLogarithm) {
// item.put("featuresCount", (featuresCount > 0) ?
// Math.log(featuresCount) : 0);
// } else {
// item.put("featuresCount", featuresCount);
// }
// item.put("end", result.getInt("end"));
// data.add(item);
// i = -1;
// featuresCount = 0;
// }
// j++;
// i++;
}
} else {
for (DBObject result : queryResults) {
// ObjectMap item = new ObjectMap("chromosome",
// result.getString("chromosome"));
// item.put("chunkId", result.getInt("chunk_id"));
// item.put("start", result.getInt("start"));
// if (histogramLogarithm) {
// int features_count = result.getInt("features_count");
// result.put("featuresCount", (features_count > 0) ?
// Math.log(features_count) : 0);
// } else {
// item.put("featuresCount", result.getInt("features_count"));
// }
// item.put("end", result.getInt("end"));
// data.add(item);
}
}
queryResult.setResult(data);
queryResult.setNumResults(data.size());
queryResult.setTime(System.currentTimeMillis() - startTime);
return queryResult;
}