Allow variant jobs to run on specific interval list

Author: bbimber

SequenceAnalysis/api-src/org/labkey/api/sequenceanalysis/pipeline/ReferenceGenome.java

@@ -15,6 +15,7 @@
  */
 package org.labkey.api.sequenceanalysis.pipeline;
 
+import htsjdk.samtools.SAMSequenceDictionary;
 import org.jetbrains.annotations.NotNull;
 
 import java.io.File;
@@ -87,6 +88,8 @@ public interface ReferenceGenome extends Serializable
      */
     File getSequenceDictionary();
 
+    SAMSequenceDictionary extractDictionary();
+
     /**
      * @return True if this is a genome not defined in the main database, such as a job using an ad hoc FASTA file or genome based on querying the NT records
      */

SequenceAnalysis/api-src/org/labkey/api/sequenceanalysis/pipeline/VariantProcessingStep.java

@@ -51,7 +51,8 @@ enum ScatterGatherMethod
         none(false),
         contig(false),
         chunked(true),
-        fixedJobs(false);
+        fixedJobs(false),
+        specificInternals(false);
 
         private final boolean _mayRequireSort;
 

SequenceAnalysis/resources/web/SequenceAnalysis/panel/VariantScatterGatherPanel.js

@@ -87,6 +87,17 @@ Ext4.define('SequenceAnalysis.panel.VariantScatterGatherPanel', {
                                 value: 10
                             });
                         }
+                        else if (val === 'specificIntervals') {
+                            toAdd.push({
+                                xtype: 'sequenceanalysis-intervalfield',
+                                labelWidth: this.labelWidth,
+                                name: 'scatterGather.specificIntervals',
+                                label: 'Intervals to Process',
+                                helpPopup: 'The intervals to process. They should be in the form: chr01:102-20394',
+                                allowBlank: false,
+                                defaultValue: null
+                            });
+                        }
 
                         if (toAdd.length) {
                             panel.add(toAdd);

SequenceAnalysis/src/org/labkey/sequenceanalysis/analysis/GenotypeGVCFHandler.java

@@ -36,7 +36,6 @@
 import org.labkey.api.writer.PrintWriters;
 import org.labkey.sequenceanalysis.ScatterGatherUtils;
 import org.labkey.sequenceanalysis.SequenceAnalysisModule;
-import org.labkey.sequenceanalysis.pipeline.JobContextImpl;
 import org.labkey.sequenceanalysis.pipeline.ProcessVariantsHandler;
 import org.labkey.sequenceanalysis.pipeline.VariantProcessingJob;
 import org.labkey.sequenceanalysis.run.util.AbstractGenomicsDBImportHandler;
@@ -50,7 +49,6 @@
 import java.io.IOException;
 import java.io.PrintWriter;
 import java.util.ArrayList;
-import java.util.Arrays;
 import java.util.Collection;
 import java.util.Date;
 import java.util.HashSet;

SequenceAnalysis/src/org/labkey/sequenceanalysis/pipeline/ReferenceGenomeImpl.java

@@ -1,6 +1,8 @@
 package org.labkey.sequenceanalysis.pipeline;
 
 import com.fasterxml.jackson.annotation.JsonIgnore;
+import htsjdk.samtools.SAMSequenceDictionary;
+import htsjdk.variant.utils.SAMSequenceDictionaryExtractor;
 import org.apache.logging.log4j.Logger;
 import org.jetbrains.annotations.NotNull;
 import org.jetbrains.annotations.Nullable;
@@ -110,6 +112,13 @@ public File getSequenceDictionary()
         return getWorkingFastaFile() == null ? null : new File(FileUtil.getBaseName(getWorkingFastaFile().getPath()) + ".dict");
     }
 
+    @JsonIgnore
+    @Override
+    public SAMSequenceDictionary extractDictionary()
+    {
+        return getSequenceDictionary() == null ? null : SAMSequenceDictionaryExtractor.extractDictionary(getSequenceDictionary().toPath());
+    }
+
     @Override
     public String getName()
     {

SequenceAnalysis/src/org/labkey/sequenceanalysis/pipeline/VariantProcessingJob.java

@@ -37,6 +37,7 @@
 import org.labkey.api.sequenceanalysis.pipeline.VariantProcessingStep;
 import org.labkey.api.writer.PrintWriters;
 import org.labkey.sequenceanalysis.util.ScatterGatherUtils;
+import org.labkey.sequenceanalysis.util.SequenceUtil;
 
 import java.io.File;
 import java.io.IOException;
@@ -164,6 +165,22 @@ else if (_scatterGatherMethod == VariantProcessingStep.ScatterGatherMethod.fixed
             getLogger().info("Creating " + numJobs + " jobs with approximate size: " + jobSize + " bp.");
             ret = ScatterGatherUtils.divideGenome(dict, jobSize, true, -1, false);
         }
+        else if (_scatterGatherMethod == VariantProcessingStep.ScatterGatherMethod.specificInternals)
+        {
+            try
+            {
+                String intervalsRaw = StringUtils.trimToNull(getParameterJson().getString("scatterGather.specificIntervals"));
+                String[] intervals = intervalsRaw.split(";");
+                List<Interval> values = SequenceUtil.validateAndParseIntervals(intervals, dict);
+
+                ret = new LinkedHashMap<>();
+                ret.put("Job1", values);
+            }
+            catch (PipelineJobException e)
+            {
+                throw new IllegalArgumentException(e);
+            }
+        }
         else
         {
             throw new IllegalArgumentException("Unknown scatter type: " + _scatterGatherMethod.name());

SequenceAnalysis/src/org/labkey/sequenceanalysis/run/variant/DepthOfCoverageHandler.java

@@ -116,7 +116,7 @@ public void processFilesRemote(List<SequenceOutputFile> inputFiles, JobContext c
             if (intervalString != null)
             {
                 String[] intervals = intervalString.split(";");
-                validateIntervals(intervals);
+                SequenceUtil.validateAndParseIntervals(intervals, rg.extractDictionary());
                 for (String i : intervals)
                 {
                     extraArgs.add("-L");
@@ -225,25 +225,4 @@ public void processFilesRemote(List<SequenceOutputFile> inputFiles, JobContext c
             ctx.getFileManager().addSequenceOutput(so);
         }
     }
-
-    public static void validateIntervals(String[] intervals) throws PipelineJobException
-    {
-        for (String i : intervals)
-        {
-            //NOTE: the contig name can contain hyphen..
-            String[] tokens = i.split(":");
-            if (tokens.length > 2)
-            {
-                throw new PipelineJobException("Invalid interval: " + i);
-            }
-            else if (tokens.length == 2)
-            {
-                String[] coords = tokens[1].split("-");
-                if (coords.length != 2)
-                {
-                    throw new PipelineJobException("Invalid interval: " + i);
-                }
-            }
-        }
-    }
 }

SequenceAnalysis/src/org/labkey/sequenceanalysis/util/SequenceUtil.java

@@ -5,6 +5,8 @@
 import htsjdk.samtools.SAMReadGroupRecord;
 import htsjdk.samtools.SAMRecord;
 import htsjdk.samtools.SAMRecordIterator;
+import htsjdk.samtools.SAMSequenceDictionary;
+import htsjdk.samtools.SAMSequenceRecord;
 import htsjdk.samtools.SamReader;
 import htsjdk.samtools.SamReaderFactory;
 import htsjdk.samtools.ValidationStringency;
@@ -661,4 +663,45 @@ public static String getLegalReadGroupName(String rsName)
     {
         return rsName.replaceAll(" ", "_");
     }
+
+    public static List<Interval> validateAndParseIntervals(String[] intervals, SAMSequenceDictionary dict) throws PipelineJobException
+    {
+        List<Interval> ret = new ArrayList<>();
+        for (String i : intervals)
+        {
+            String contig;
+            //NOTE: the contig name can contain hyphen..
+            String[] tokens = i.split(":");
+            if (tokens.length > 2)
+            {
+                throw new PipelineJobException("Invalid interval: " + i);
+            }
+
+            if (tokens.length == 2)
+            {
+                String[] coords = tokens[1].split("-");
+                if (coords.length != 2)
+                {
+                    throw new PipelineJobException("Invalid interval: " + i);
+                }
+
+                int start = Integer.parseInt(coords[0]);
+                int end = Integer.parseInt(coords[1]);
+
+                ret.add(new Interval(tokens[0], start, end));
+            }
+            else
+            {
+                SAMSequenceRecord rec = dict.getSequence(tokens[0]);
+                if (rec == null)
+                {
+                    throw new PipelineJobException("Unable to find sequence: " + tokens[0]);
+                }
+
+                ret.add(new Interval(tokens[0], 1, rec.getSequenceLength()));
+            }
+        }
+
+        return ret;
+    }
 }