Updating the core LocusWalker tools to include the Sample infrastructure that I added last month. This commit touches a lot of files, but only significantly changes a few: LocusIteratorByState and ReadBackedPileup and associated classes.

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@4711 348d0f76-0448-11de-a6fe-93d51630548a

java/src/org/broadinstitute/sting/gatk/AbstractGenomeAnalysisEngine.java

@@ -320,6 +320,8 @@ public abstract class AbstractGenomeAnalysisEngine {
         validateSuppliedReads();
         readsDataSource = createReadsDataSource(genomeLocParser);
 
+        sampleDataSource = new SampleDataSource(getSAMFileHeader(), argCollection.sampleFiles);
+
         for (SamRecordFilter filter : filters)
             if (filter instanceof SamRecordHeaderFilter)
                 ((SamRecordHeaderFilter)filter).setHeader(this.getSAMFileHeader());
@@ -744,6 +746,10 @@ public abstract class AbstractGenomeAnalysisEngine {
         return unpackedReads;
     }
 
+    public SampleDataSource getSampleMetadata() {
+        return this.sampleDataSource;
+    }
+
     /**
      * Get a sample by its ID
      * If an alias is passed in, return the main sample object
@@ -802,6 +808,14 @@ public abstract class AbstractGenomeAnalysisEngine {
         return sampleDataSource.getChildren(sample);
     }
 
+    /**
+     * Gets all the samples
+     * @return
+     */
+    public Collection<Sample> getSamples() {
+        return sampleDataSource.getSamples();
+    }
+
     /**
      * Takes a list of sample names and returns their corresponding sample objects
      *

java/src/org/broadinstitute/sting/gatk/contexts/StratifiedAlignmentContext.java

@@ -25,6 +25,8 @@
 
 package org.broadinstitute.sting.gatk.contexts;
 
+import org.broadinstitute.sting.gatk.GenomeAnalysisEngine;
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.utils.exceptions.ReviewedStingException;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import org.broadinstitute.sting.utils.exceptions.UserException;
@@ -84,7 +86,7 @@ public class StratifiedAlignmentContext<RBP extends ReadBackedPileup> {
      * @return a Map of sample name to StratifiedAlignmentContext
      *
      **/
-    public static <RBP extends ReadBackedPileup,PE extends PileupElement> Map<String, StratifiedAlignmentContext> splitContextBySample(RBP pileup) {
+    public static <RBP extends ReadBackedPileup,PE extends PileupElement> Map<Sample, StratifiedAlignmentContext> splitContextBySample(RBP pileup) {
         return splitContextBySample(pileup, null);
     }
 
@@ -97,20 +99,20 @@ public class StratifiedAlignmentContext<RBP extends ReadBackedPileup> {
      * @return a Map of sample name to StratifiedAlignmentContext
      *
      **/
-    public static <RBP extends ReadBackedPileup> Map<String, StratifiedAlignmentContext> splitContextBySample(RBP pileup, String assumedSingleSample) {
+    public static <RBP extends ReadBackedPileup> Map<Sample, StratifiedAlignmentContext> splitContextBySample(RBP pileup, Sample assumedSingleSample) {
 
         GenomeLoc loc = pileup.getLocation();
-        HashMap<String, StratifiedAlignmentContext> contexts = new HashMap<String, StratifiedAlignmentContext>();
+        HashMap<Sample, StratifiedAlignmentContext> contexts = new HashMap<Sample, StratifiedAlignmentContext>();
 
-        for(String sampleName: pileup.getSamples()) {
+        for(Sample sample: pileup.getSamples()) {
-            RBP pileupBySample = (RBP)pileup.getPileupForSample(sampleName);
+            RBP pileupBySample = (RBP)pileup.getPileupForSample(sample);
 
             // Don't add empty pileups to the split context.
             if(pileupBySample.size() == 0)
                 continue;
 
-            if(sampleName != null)
+            if(sample != null)
-                contexts.put(sampleName,new StratifiedAlignmentContext<RBP>(loc,pileupBySample));
+                contexts.put(sample,new StratifiedAlignmentContext<RBP>(loc,pileupBySample));
             else {
                 if(assumedSingleSample == null) {
                     throw new UserException.ReadMissingReadGroup(pileupBySample.iterator().next().getRead());
@@ -122,7 +124,48 @@ public class StratifiedAlignmentContext<RBP extends ReadBackedPileup> {
         return contexts;
     }
 
-     /**
+
+
+    public static <RBP extends ReadBackedPileup,PE extends PileupElement> Map<String, StratifiedAlignmentContext> splitContextBySampleName(RBP pileup) {
+        return splitContextBySampleName(pileup, null);
+    }
+
+    /**
+     * Splits the given AlignmentContext into a StratifiedAlignmentContext per sample, but referencd by sample name instead
+     * of sample object.
+     *
+     * @param pileup                the original pileup
+     *
+     * @return a Map of sample name to StratifiedAlignmentContext
+     *
+     **/
+    public static <RBP extends ReadBackedPileup> Map<String, StratifiedAlignmentContext> splitContextBySampleName(RBP pileup, String assumedSingleSample) {
+
+        GenomeLoc loc = pileup.getLocation();
+        HashMap<String, StratifiedAlignmentContext> contexts = new HashMap<String, StratifiedAlignmentContext>();
+
+        for(String sample: pileup.getSampleNames()) {
+            RBP pileupBySample = (RBP)pileup.getPileupForSampleName(sample);
+
+            // Don't add empty pileups to the split context.
+            if(pileupBySample.size() == 0)
+                continue;
+
+            if(sample != null)
+                contexts.put(sample,new StratifiedAlignmentContext<RBP>(loc,pileupBySample));
+            else {
+                if(assumedSingleSample == null) {
+                    throw new UserException.ReadMissingReadGroup(pileupBySample.iterator().next().getRead());
+                }
+                contexts.put(assumedSingleSample,new StratifiedAlignmentContext<RBP>(loc,pileupBySample));
+            }
+        }
+
+        return contexts;
+    }
+
+
+    /**
      * Splits the given AlignmentContext into a StratifiedAlignmentContext per read group.
      *
      * @param pileup                the original pileup
@@ -131,14 +174,14 @@ public class StratifiedAlignmentContext<RBP extends ReadBackedPileup> {
      *
      **/
     public static <RBP extends ReadBackedPileup> Map<String,StratifiedAlignmentContext<RBP>> splitContextByReadGroup(RBP pileup) {
-         HashMap<String,StratifiedAlignmentContext<RBP>> contexts = new HashMap<String,StratifiedAlignmentContext<RBP>>();
+        HashMap<String,StratifiedAlignmentContext<RBP>> contexts = new HashMap<String,StratifiedAlignmentContext<RBP>>();
-         for(String readGroupId: pileup.getReadGroups())
+        for(String readGroupId: pileup.getReadGroups())
-             contexts.put(readGroupId,new StratifiedAlignmentContext<RBP>(pileup.getLocation(),(RBP)pileup.getPileupForReadGroup(readGroupId)));
+            contexts.put(readGroupId,new StratifiedAlignmentContext<RBP>(pileup.getLocation(),(RBP)pileup.getPileupForReadGroup(readGroupId)));
-         return contexts;
+        return contexts;
     }
 
     public static AlignmentContext joinContexts(Collection<StratifiedAlignmentContext> contexts) {
-         
+
         // validation
         GenomeLoc loc = contexts.iterator().next().getLocation();
         boolean isExtended = contexts.iterator().next().basePileup instanceof ReadBackedExtendedEventPileup;

java/src/org/broadinstitute/sting/gatk/datasources/sample/Sample.java

@@ -194,6 +194,6 @@ public class Sample implements java.io.Serializable {
 
     @Override
     public int hashCode() {
-        return id.hashCode();
+        return id != null ? id.hashCode() : "".hashCode();
     }
 }

java/src/org/broadinstitute/sting/gatk/datasources/sample/SampleDataSource.java

@@ -37,13 +37,13 @@ public class SampleDataSource {
     /**
      * SAMFileHeader that has been created for this analysis.
      */
-    private final SAMFileHeader header;
+    private SAMFileHeader header;
 
     /**
      * This is where Sample objects are stored. Samples are usually accessed by their ID, which is unique, so
      * this is stored as a HashMap.
      */
-    private HashMap<String, Sample> samples = new HashMap<String, Sample>();
+    private final HashMap<String, Sample> samples = new HashMap<String, Sample>();
 
     /**
      * Samples can have "aliases", because sometimes the same sample is referenced by different IDs in different
@@ -65,6 +65,7 @@ public class SampleDataSource {
      * @param sampleFiles Sample files that were included on the command line
      */
     public SampleDataSource(SAMFileHeader header, List<File> sampleFiles) {
+        this();
         this.header = header;
         // create empty sample object for each sample referenced in the SAM header
         for (String sampleName : SampleUtils.getSAMFileSamples(header)) {
@@ -83,10 +84,14 @@ public class SampleDataSource {
         }
     }
 
+    public SampleDataSource() {
+        samples.put(null, new Sample(null));
+    }
+
     /**
      * Hallucinates sample objects for all the samples in the SAM file and stores them
      */
-    private void getSamplesFromSAMFile() {
+    public void addSamplesFromSAMHeader(SAMFileHeader header) {
         for (String sampleName : SampleUtils.getSAMFileSamples(header)) {
             if (!hasSample(sampleName)) {
                 Sample newSample = new Sample(sampleName);
@@ -100,7 +105,7 @@ public class SampleDataSource {
      * Parse one sample file and integrate it with samples that are already there
      * Fail quickly if we find any errors in the file
      */
-    private void addFile(File sampleFile) {
+    public void addFile(File sampleFile) {
 
         BufferedReader reader;
         try {
@@ -466,6 +471,12 @@ public class SampleDataSource {
         return children;
     }
 
+    public Set<Sample> getSamples() {
+        HashSet<Sample> set = new HashSet<Sample>();
+        set.addAll(samples.values());
+        return set;
+    }
+
     /**
      * Takes a collection of sample names and returns their corresponding sample objects
      * Note that, since a set is returned, if you pass in a list with duplicates names there will not be any duplicates in the returned set
@@ -572,4 +583,9 @@ public class SampleDataSource {
         return context.subContextFromGenotypes(genotypes.values());
     }
 
+    public static SampleDataSource createEmptyDataSource() {
+        SAMFileHeader header = new SAMFileHeader();
+        return new SampleDataSource(header, null);
+    }
+
 }

java/src/org/broadinstitute/sting/gatk/executive/LinearMicroScheduler.java

@@ -53,7 +53,7 @@ public class LinearMicroScheduler extends MicroScheduler {
             // New experimental code for managing locus intervals.
             if(shard.getShardType() == Shard.ShardType.LOCUS) {
                 LocusWalker lWalker = (LocusWalker)walker;
-                WindowMaker windowMaker = new WindowMaker(shard, engine.getGenomeLocParser(), getReadIterator(shard), shard.getGenomeLocs(), lWalker.getDiscards());
+                WindowMaker windowMaker = new WindowMaker(shard, engine.getGenomeLocParser(), getReadIterator(shard), shard.getGenomeLocs(), lWalker.getDiscards(), engine.getSampleMetadata());
                 for(WindowMaker.WindowMakerIterator iterator: windowMaker) {
                     ShardDataProvider dataProvider = new LocusShardDataProvider(shard,iterator.getSourceInfo(),engine.getGenomeLocParser(),iterator.getLocus(),iterator,reference,rods);
                     Object result = traversalEngine.traverse(walker, dataProvider, accumulator.getReduceInit());

java/src/org/broadinstitute/sting/gatk/executive/ShardTraverser.java

@@ -61,7 +61,7 @@ public class ShardTraverser implements Callable {
 
             Object accumulator = walker.reduceInit();
             LocusWalker lWalker = (LocusWalker)walker;
-            WindowMaker windowMaker = new WindowMaker(shard,microScheduler.getEngine().getGenomeLocParser(),microScheduler.getReadIterator(shard),shard.getGenomeLocs(),lWalker.getDiscards());
+            WindowMaker windowMaker = new WindowMaker(shard,microScheduler.getEngine().getGenomeLocParser(),microScheduler.getReadIterator(shard),shard.getGenomeLocs(),lWalker.getDiscards(), microScheduler.engine.getSampleMetadata()); // todo: microScheduler.engine is protected - is it okay to user it here?
             ShardDataProvider dataProvider = null;
 
             for(WindowMaker.WindowMakerIterator iterator: windowMaker) {

java/src/org/broadinstitute/sting/gatk/executive/WindowMaker.java

@@ -1,5 +1,8 @@
 package org.broadinstitute.sting.gatk.executive;
 
+import org.broadinstitute.sting.gatk.GenomeAnalysisEngine;
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
+import org.broadinstitute.sting.gatk.datasources.sample.SampleDataSource;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import org.broadinstitute.sting.gatk.iterators.*;
 import org.broadinstitute.sting.gatk.ReadProperties;
@@ -50,13 +53,16 @@ public class WindowMaker implements Iterable<WindowMaker.WindowMakerIterator>, I
      * the given intervals.
      * @param iterator The data source for this window.
      * @param intervals The set of intervals over which to traverse.
+     * @param sampleData SampleDataSource that we can reference reads with
      * @param discards a filter at that indicates read position relative to some locus?
      */
-    public WindowMaker(Shard shard, GenomeLocParser genomeLocParser, StingSAMIterator iterator, List<GenomeLoc> intervals, List<LocusIteratorFilter> discards ) {
+
+    public WindowMaker(Shard shard, GenomeLocParser genomeLocParser, StingSAMIterator iterator, List<GenomeLoc> intervals, List<LocusIteratorFilter> discards, SampleDataSource sampleData ) {
         this.sourceInfo = shard.getReadProperties();
         this.readIterator = iterator;
 
-        LocusIterator locusIterator = new LocusIteratorByState(iterator,sourceInfo,genomeLocParser,discards);
+        LocusIterator locusIterator = new LocusIteratorByState(iterator,sourceInfo,genomeLocParser,sampleData,discards);
+
 
         this.sourceIterator = new PeekableIterator<AlignmentContext>(locusIterator);
         this.intervalIterator = intervals.size()>0 ? new PeekableIterator<GenomeLoc>(intervals.iterator()) : null;

java/src/org/broadinstitute/sting/gatk/iterators/LocusIteratorByState.java

@@ -33,6 +33,8 @@ import org.broadinstitute.sting.gatk.GenomeAnalysisEngine;
 import org.broadinstitute.sting.gatk.DownsamplingMethod;
 import org.broadinstitute.sting.gatk.DownsampleType;
 import org.broadinstitute.sting.gatk.contexts.AlignmentContext;
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
+import org.broadinstitute.sting.gatk.datasources.sample.SampleDataSource;
 import org.broadinstitute.sting.utils.*;
 import org.broadinstitute.sting.utils.exceptions.UserException;
 import org.broadinstitute.sting.utils.pileup.*;
@@ -65,8 +67,11 @@ public class LocusIteratorByState extends LocusIterator {
      * Used to create new GenomeLocs.
      */
     private final GenomeLocParser genomeLocParser;
-
+    
-    private final Collection<String> sampleNames = new ArrayList<String>();
+    private final SampleDataSource sampleData;
+    
+    private final ArrayList<Sample> samples;
+    
     private final ReadStateManager readStates;
 
     static private class SAMRecordState {
@@ -273,19 +278,23 @@ public class LocusIteratorByState extends LocusIterator {
     // constructors and other basic operations
     //
     // -----------------------------------------------------------------------------------------------------------------
-    public LocusIteratorByState(final Iterator<SAMRecord> samIterator, ReadProperties readInformation, GenomeLocParser genomeLocParser ) {
+
-        this(samIterator, readInformation, genomeLocParser, NO_FILTERS);
+    public LocusIteratorByState(final Iterator<SAMRecord> samIterator, ReadProperties readInformation, GenomeLocParser genomeLocParser, SampleDataSource sampleData ) {
+        this(samIterator, readInformation, genomeLocParser, sampleData, NO_FILTERS);
     }
-    
+
-    public LocusIteratorByState(final Iterator<SAMRecord> samIterator, ReadProperties readInformation, GenomeLocParser genomeLocParser, List<LocusIteratorFilter> filters ) {
+    public LocusIteratorByState(final Iterator<SAMRecord> samIterator, ReadProperties readInformation, GenomeLocParser genomeLocParser, SampleDataSource sampleData, List<LocusIteratorFilter> filters ) {
         this.readInfo = readInformation;
         this.genomeLocParser = genomeLocParser;
         this.filters = filters;
-        // Aggregate all sample names.
+        
-        sampleNames.addAll(SampleUtils.getSAMFileSamples(readInfo.getHeader()));
+        // set up the sample data
-        // Add a null sample name as a catch-all for reads without samples
+        this.sampleData = sampleData;
-        if(!sampleNames.contains(null)) sampleNames.add(null);        
+        this.samples = new ArrayList<Sample>();
-        readStates = new ReadStateManager(samIterator,readInformation.getDownsamplingMethod(),sampleNames);
+        this.samples.addAll(sampleData.getSamples());
+        
+        readStates = new ReadStateManager(samIterator,readInformation.getDownsamplingMethod(),sampleData);
+    
     }
 
     public Iterator<AlignmentContext> iterator() {
@@ -303,8 +312,8 @@ public class LocusIteratorByState extends LocusIterator {
     }
 
     public void printState() {
-        for(String sampleName: sampleNames) {
+        for(Sample sample: samples) {
-            Iterator<SAMRecordState> iterator = readStates.iteratorForSample(sampleName);
+            Iterator<SAMRecordState> iterator = readStates.iteratorForSample(sample);
             while(iterator.hasNext()) {
                 SAMRecordState state = iterator.next();
                 logger.debug(String.format("printState():"));
@@ -354,8 +363,8 @@ public class LocusIteratorByState extends LocusIterator {
             // In this case, the subsequent call to next() will emit the normal pileup at the current base
             // and shift the position.
             if (readInfo.generateExtendedEvents() && hasExtendedEvents) {
-                Map<String,ReadBackedExtendedEventPileupImpl> fullExtendedEventPileup =
+                Map<Sample,ReadBackedExtendedEventPileupImpl> fullExtendedEventPileup =
-                        new HashMap<String,ReadBackedExtendedEventPileupImpl>();
+                        new HashMap<Sample,ReadBackedExtendedEventPileupImpl>();
 
                 SAMRecordState our1stState = readStates.getFirst();
                 // get current location on the reference and decrement it by 1: the indels we just stepped over
@@ -363,10 +372,10 @@ public class LocusIteratorByState extends LocusIterator {
                 GenomeLoc loc = genomeLocParser.incPos(our1stState.getLocation(genomeLocParser),-1);
 
                 boolean hasBeenSampled = false;
-                for(String sampleName: sampleNames) {
+                for(Sample sample: samples) {
-                    Iterator<SAMRecordState> iterator = readStates.iteratorForSample(sampleName);
+                    Iterator<SAMRecordState> iterator = readStates.iteratorForSample(sample);
                     ArrayList<ExtendedEventPileupElement> indelPile = new ArrayList<ExtendedEventPileupElement>(readStates.size());
-                    hasBeenSampled |= loc.getStart() <= readStates.readStatesBySample.get(sampleName).getDownsamplingExtent();
+                    hasBeenSampled |= loc.getStart() <= readStates.readStatesBySample.get(sample).getDownsamplingExtent();
 
                     size = 0;
                     nDeletions = 0;
@@ -419,7 +428,7 @@ public class LocusIteratorByState extends LocusIterator {
                             nMQ0Reads++;
                         }
                     }
-                    if( indelPile.size() != 0 ) fullExtendedEventPileup.put(sampleName,new ReadBackedExtendedEventPileupImpl(loc,indelPile,size,maxDeletionLength,nInsertions,nDeletions,nMQ0Reads));
+                    if( indelPile.size() != 0 ) fullExtendedEventPileup.put(sample,new ReadBackedExtendedEventPileupImpl(loc,indelPile,size,maxDeletionLength,nInsertions,nDeletions,nMQ0Reads));
                 }
                 hasExtendedEvents = false; // we are done with extended events prior to current ref base
 //                System.out.println("Indel(s) at "+loc);
@@ -427,14 +436,14 @@ public class LocusIteratorByState extends LocusIterator {
                 nextAlignmentContext = new AlignmentContext(loc, new ReadBackedExtendedEventPileupImpl(loc, fullExtendedEventPileup), hasBeenSampled);
             }  else {
                 GenomeLoc location = getLocation();
-                Map<String,ReadBackedPileupImpl> fullPileup = new HashMap<String,ReadBackedPileupImpl>();
+                Map<Sample,ReadBackedPileupImpl> fullPileup = new HashMap<Sample,ReadBackedPileupImpl>();
 
                 // todo -- performance problem -- should be lazy, really
                 boolean hasBeenSampled = false;
-                for(String sampleName: sampleNames) {
+                for(Sample sample: samples) {
-                    Iterator<SAMRecordState> iterator = readStates.iteratorForSample(sampleName);
+                    Iterator<SAMRecordState> iterator = readStates.iteratorForSample(sample);
                     ArrayList<PileupElement> pile = new ArrayList<PileupElement>(readStates.size());
-                    hasBeenSampled |= location.getStart() <= readStates.readStatesBySample.get(sampleName).getDownsamplingExtent();
+                    hasBeenSampled |= location.getStart() <= readStates.readStatesBySample.get(sample).getDownsamplingExtent();
 
                     size = 0;
                     nDeletions = 0;
@@ -464,7 +473,7 @@ public class LocusIteratorByState extends LocusIterator {
                             nMQ0Reads++;
                         }
                     }
-                    if( pile.size() != 0 ) fullPileup.put(sampleName,new ReadBackedPileupImpl(location,pile,size,nDeletions,nMQ0Reads));
+                    if( pile.size() != 0 ) fullPileup.put(sample,new ReadBackedPileupImpl(location,pile,size,nDeletions,nMQ0Reads));
                 }
 
                 updateReadStates(); // critical - must be called after we get the current state offsets and location
@@ -507,8 +516,8 @@ public class LocusIteratorByState extends LocusIterator {
 //    }
 
     private void updateReadStates() {
-        for(String sampleName: sampleNames) {
+        for(Sample sample: samples) {
-            Iterator<SAMRecordState> it = readStates.iteratorForSample(sampleName);
+            Iterator<SAMRecordState> it = readStates.iteratorForSample(sample);
             while ( it.hasNext() ) {
                 SAMRecordState state = it.next();
                 CigarOperator op = state.stepForwardOnGenome();
@@ -536,13 +545,13 @@ public class LocusIteratorByState extends LocusIterator {
 
         private final SamplePartitioner samplePartitioner;
 
-        private final Map<String,PerSampleReadStateManager> readStatesBySample = new HashMap<String,PerSampleReadStateManager>();
+        private final Map<Sample,PerSampleReadStateManager> readStatesBySample = new HashMap<Sample,PerSampleReadStateManager>();
 
         private final int targetCoverage;
 
         private int totalReadStates = 0;
 
-        public ReadStateManager(Iterator<SAMRecord> source, DownsamplingMethod downsamplingMethod, Collection<String> sampleNames) {
+        public ReadStateManager(Iterator<SAMRecord> source, DownsamplingMethod downsamplingMethod, SampleDataSource sampleData) {
             this.iterator = new PeekableIterator<SAMRecord>(source);
             this.downsamplingMethod = downsamplingMethod.type != null ? downsamplingMethod : DownsamplingMethod.NONE;
             switch(this.downsamplingMethod.type) {
@@ -556,17 +565,17 @@ public class LocusIteratorByState extends LocusIterator {
             }
 
             Map<String,ReadSelector> readSelectors = new HashMap<String,ReadSelector>();
-            for(String sampleName: sampleNames) {
+            for(Sample sample: samples) {
-                readStatesBySample.put(sampleName,new PerSampleReadStateManager());
+                readStatesBySample.put(sample,new PerSampleReadStateManager());
-                readSelectors.put(sampleName,downsamplingMethod.type == DownsampleType.BY_SAMPLE ? new NRandomReadSelector(null,targetCoverage) : new AllReadsSelector());
+                readSelectors.put(sample.getId(),downsamplingMethod.type == DownsampleType.BY_SAMPLE ? new NRandomReadSelector(null,targetCoverage) : new AllReadsSelector());
             }
 
             samplePartitioner = new SamplePartitioner(readSelectors);
         }
 
-        public Iterator<SAMRecordState> iteratorForSample(final String sampleName) {
+        public Iterator<SAMRecordState> iteratorForSample(final Sample sample) {
             return new Iterator<SAMRecordState>() {
-                private Iterator<SAMRecordState> wrappedIterator = readStatesBySample.get(sampleName).iterator();
+                private Iterator<SAMRecordState> wrappedIterator = readStatesBySample.get(sample).iterator();
 
                 public boolean hasNext() {
                     return wrappedIterator.hasNext();
@@ -592,8 +601,8 @@ public class LocusIteratorByState extends LocusIterator {
         }
 
         public SAMRecordState getFirst() {
-            for(String sampleName: sampleNames) {
+            for(Sample sample: samples) {
-                PerSampleReadStateManager reads = readStatesBySample.get(sampleName);
+                PerSampleReadStateManager reads = readStatesBySample.get(sample);
                 if(!reads.isEmpty())
                     return reads.peek();
             }
@@ -626,12 +635,12 @@ public class LocusIteratorByState extends LocusIterator {
             }
             samplePartitioner.complete();
 
-            for(String sampleName: sampleNames) {
+            for(Sample sample: samples) {
-                ReadSelector aggregator = samplePartitioner.getSelectedReads(sampleName);
+                ReadSelector aggregator = samplePartitioner.getSelectedReads(sample.getId());
 
                 Collection<SAMRecord> newReads = new ArrayList<SAMRecord>(aggregator.getSelectedReads());
 
-                PerSampleReadStateManager statesBySample = readStatesBySample.get(sampleName);
+                PerSampleReadStateManager statesBySample = readStatesBySample.get(sample);
                 int numReads = statesBySample.size();
                 int downsamplingExtent = aggregator.getDownsamplingExtent();
 
@@ -998,6 +1007,9 @@ class NRandomReadSelector implements ReadSelector {
     }
 }
 
+/**
+ * Note: stores reads by sample ID string, not by sample object
+ */
 class SamplePartitioner implements ReadSelector {
     private final Map<String,ReadSelector> readsBySample;
     private long readsSeen = 0;

java/src/org/broadinstitute/sting/gatk/walkers/annotator/VariantAnnotator.java

@@ -211,9 +211,9 @@ public class VariantAnnotator extends RodWalker<Integer, Integer> {
         Map<String, StratifiedAlignmentContext> stratifiedContexts;
         if ( BaseUtils.simpleBaseToBaseIndex(ref.getBase()) != -1 ) {
             if ( ! context.hasExtendedEventPileup() ) {
-                stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(context.getBasePileup(), ASSUME_SINGLE_SAMPLE);
+                stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(context.getBasePileup(), ASSUME_SINGLE_SAMPLE);
             } else {
-                stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(context.getExtendedEventPileup(), ASSUME_SINGLE_SAMPLE);
+                stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(context.getExtendedEventPileup(), ASSUME_SINGLE_SAMPLE);
             }
             if ( stratifiedContexts != null ) {
                 annotatedVCs = engine.annotateContext(tracker, ref, stratifiedContexts, vc);

java/src/org/broadinstitute/sting/gatk/walkers/annotator/genomicannotator/GenomicAnnotator.java

@@ -249,7 +249,7 @@ public class GenomicAnnotator extends RodWalker<Integer, Integer> implements Tre
                     (vc.isVariant() && !vc.isBiallelic()) ) {
                 results.add(vc);
             } else {
-                Map<String, StratifiedAlignmentContext> stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(context.getBasePileup());
+                Map<String, StratifiedAlignmentContext> stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(context.getBasePileup());
                 if ( stratifiedContexts != null )
                     results.addAll(engine.annotateContext(tracker, ref, stratifiedContexts, vc));
                 else

java/src/org/broadinstitute/sting/gatk/walkers/genotyper/UnifiedGenotyperEngine.java

@@ -150,7 +150,7 @@ public class UnifiedGenotyperEngine {
                 return null;
 
             // stratify the AlignmentContext and cut by sample
-            Map<String, StratifiedAlignmentContext> stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(pileup, UAC.ASSUME_SINGLE_SAMPLE);
+            Map<String, StratifiedAlignmentContext> stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(pileup, UAC.ASSUME_SINGLE_SAMPLE);
             if ( stratifiedContexts == null )
                 return null;
 
@@ -177,7 +177,7 @@ public class UnifiedGenotyperEngine {
                 return null;
 
             // stratify the AlignmentContext and cut by sample
-            Map<String, StratifiedAlignmentContext> stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(pileup, UAC.ASSUME_SINGLE_SAMPLE);
+            Map<String, StratifiedAlignmentContext> stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(pileup, UAC.ASSUME_SINGLE_SAMPLE);
             if ( stratifiedContexts == null )
                 return null;
 
@@ -187,7 +187,7 @@ public class UnifiedGenotyperEngine {
             // annotate the call, if possible
             if ( call != null && call.vc != null && annotationEngine != null ) {
                 // first off, we want to use the *unfiltered* context for the annotations
-                stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(rawContext.getBasePileup(), UAC.ASSUME_SINGLE_SAMPLE);
+                stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(rawContext.getBasePileup(), UAC.ASSUME_SINGLE_SAMPLE);
 
                 Collection<VariantContext> variantContexts = annotationEngine.annotateContext(tracker, refContext, stratifiedContexts, call.vc);
                 call.vc = variantContexts.iterator().next(); //We know the collection will always have exactly 1 element.

java/src/org/broadinstitute/sting/gatk/walkers/phasing/ReadBackedPhasingWalker.java

@@ -32,6 +32,7 @@ import org.broadinstitute.sting.commandline.Hidden;
 import org.broadinstitute.sting.gatk.contexts.AlignmentContext;
 import org.broadinstitute.sting.gatk.contexts.ReferenceContext;
 import org.broadinstitute.sting.gatk.contexts.variantcontext.VariantContextUtils;
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.gatk.filters.ZeroMappingQualityReadFilter;
 import org.broadinstitute.sting.gatk.refdata.RefMetaDataTracker;
 import org.broadinstitute.sting.gatk.refdata.ReferenceOrderedDatum;
@@ -1025,14 +1026,14 @@ public class ReadBackedPhasingWalker extends RodWalker<PhasingStatsAndOutput, Ph
                     // filter the read-base pileup based on min base and mapping qualities:
                     pileup = pileup.getBaseAndMappingFilteredPileup(MIN_BASE_QUALITY_SCORE, MIN_MAPPING_QUALITY_SCORE);
                     if (pileup != null) {
-                        for (String samp : pileup.getSamples()) {
+                        for (Sample sample : pileup.getSamples()) {
-                            ReadBackedPileup samplePileup = pileup.getPileupForSample(samp);
+                            ReadBackedPileup samplePileup = pileup.getPileupForSample(sample);
                             ReadBasesAtPosition readBases = new ReadBasesAtPosition();
                             for (PileupElement p : samplePileup) {
                                 if (!p.isDeletion()) // IGNORE deletions for now
                                     readBases.putReadBase(p);
                             }
-                            sampleReadBases.put(samp, readBases);
+                            sampleReadBases.put(sample.getId(), readBases);
                         }
                     }
                 }

java/src/org/broadinstitute/sting/oneoffprojects/walkers/AlleleBalanceHistogramWalker.java

@@ -70,7 +70,7 @@ public class AlleleBalanceHistogramWalker extends LocusWalker<Map<String,Double>
     }
 
     private HashMap<String,Double> getAlleleBalanceBySample(VariantContext vc, ReferenceContext ref, AlignmentContext context) {
-        Map<String, StratifiedAlignmentContext> sampleContext = StratifiedAlignmentContext.splitContextBySample(context.getBasePileup(),null);
+        Map<String, StratifiedAlignmentContext> sampleContext = StratifiedAlignmentContext.splitContextBySampleName(context.getBasePileup(),null);
         HashMap<String,Double> balances = new HashMap<String,Double>();
         System.out.println("----- "+ref.getLocus()+" -----");
         int returnedBalances = 0;

java/src/org/broadinstitute/sting/oneoffprojects/walkers/MendelianViolationClassifier.java

@@ -467,7 +467,7 @@ public class MendelianViolationClassifier extends LocusWalker<MendelianViolation
                 throw new ReviewedStingException("Parental bases have length zero at "+trio.toString());
             }
 
-            Map<String,StratifiedAlignmentContext> splitContext = StratifiedAlignmentContext.splitContextBySample(context.getBasePileup());
+            Map<String,StratifiedAlignmentContext> splitContext = StratifiedAlignmentContext.splitContextBySampleName(context.getBasePileup());
             Double proportion = getAlleleProportion(parental,splitContext.get(trioStructure.child));
             if ( proportion != null ) {
                 violation.addAttribute(MendelianInfoKey.ProportionOfParentAllele.getKey(), proportion);
@@ -501,7 +501,7 @@ public class MendelianViolationClassifier extends LocusWalker<MendelianViolation
         // look for tri-allelic sites mis-called as hom -- as a speedup we do this only at non-filtered, non genotype error sites
 
         if ( ! trio.isFiltered()  ) {
-            Map<String,StratifiedAlignmentContext> splitCon = StratifiedAlignmentContext.splitContextBySample(context.getBasePileup());
+            Map<String,StratifiedAlignmentContext> splitCon = StratifiedAlignmentContext.splitContextBySampleName(context.getBasePileup());
             Pair<Allele,Integer> triAl = getTriAllelicQuality(tracker, ref, trio, splitCon);
             if ( triAl != null ) {
                 violation.addAttribute(MendelianInfoKey.TriAllelicBase.getKey(),triAl.first.toString());

java/src/org/broadinstitute/sting/playground/gatk/walkers/genotyper/UnifiedGenotyperEngine.java

@@ -300,7 +300,7 @@ public class UnifiedGenotyperEngine {
             else if (rawContext.hasBasePileup())
                 pileup = rawContext.getBasePileup();
 
-            stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(pileup, UAC.ASSUME_SINGLE_SAMPLE);
+            stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(pileup, UAC.ASSUME_SINGLE_SAMPLE);
 
             Collection<VariantContext> variantContexts = annotationEngine.annotateContext(tracker, refContext, stratifiedContexts, vc);
             vc = variantContexts.iterator().next(); //We know the collection will always have exactly 1 element.
@@ -330,7 +330,7 @@ public class UnifiedGenotyperEngine {
                 return null;
 
             // stratify the AlignmentContext and cut by sample
-            stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(pileup, UAC.ASSUME_SINGLE_SAMPLE);
+            stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(pileup, UAC.ASSUME_SINGLE_SAMPLE);
 
         } else if ( UAC.GLmodel == GenotypeLikelihoodsCalculationModel.Model.SNP && !rawContext.hasExtendedEventPileup() ) {
 
@@ -339,7 +339,7 @@ public class UnifiedGenotyperEngine {
                 return null;
 
             // stratify the AlignmentContext and cut by sample
-            stratifiedContexts = StratifiedAlignmentContext.splitContextBySample(rawContext.getBasePileup(), UAC.ASSUME_SINGLE_SAMPLE);
+            stratifiedContexts = StratifiedAlignmentContext.splitContextBySampleName(rawContext.getBasePileup(), UAC.ASSUME_SINGLE_SAMPLE);
 
             // filter the reads (and test for bad pileups)
             if ( !filterPileup(stratifiedContexts, badBaseFilter) )

java/src/org/broadinstitute/sting/utils/pileup/AbstractReadBackedPileup.java

@@ -24,6 +24,7 @@
 
 package org.broadinstitute.sting.utils.pileup;
 
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.utils.exceptions.ReviewedStingException;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import org.broadinstitute.sting.utils.BaseUtils;
@@ -32,6 +33,7 @@ import org.broadinstitute.sting.gatk.iterators.IterableIterator;
 import java.util.*;
 
 import net.sf.samtools.SAMRecord;
+import org.broadinstitute.sting.utils.exceptions.StingException;
 
 /**
  * A generic implementation of read-backed pileups.
@@ -113,10 +115,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
         calculateCachedData();
     }
 
-    protected AbstractReadBackedPileup(GenomeLoc loc, Map<String,? extends AbstractReadBackedPileup<RBP,PE>> pileupsBySample) {
+    protected AbstractReadBackedPileup(GenomeLoc loc, Map<Sample,? extends AbstractReadBackedPileup<RBP,PE>> pileupsBySample) {
         this.loc = loc;
         PerSamplePileupElementTracker<PE> tracker = new PerSamplePileupElementTracker<PE>();
-        for(Map.Entry<String,? extends AbstractReadBackedPileup<RBP,PE>> pileupEntry: pileupsBySample.entrySet()) {
+        for(Map.Entry<Sample,? extends AbstractReadBackedPileup<RBP,PE>> pileupEntry: pileupsBySample.entrySet()) {
             tracker.addElements(pileupEntry.getKey(),pileupEntry.getValue().pileupElementTracker);
             addPileupToCumulativeStats(pileupEntry.getValue());
         }
@@ -212,10 +214,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
                 PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
                 PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-                for(String sampleName: tracker.getSamples()) {
+                for(Sample sample: tracker.getSamples()) {
-                    PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                    PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                     AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getPileupWithoutDeletions();
-                    filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                    filteredTracker.addElements(sample,pileup.pileupElementTracker);
                 }
                 return (RBP)createNewPileup(loc,filteredTracker);
 
@@ -250,10 +252,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                 AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getOverlappingFragmentFilteredPileup();
-                filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                filteredTracker.addElements(sample,pileup.pileupElementTracker);
             }
             return (RBP)createNewPileup(loc,filteredTracker);
         }
@@ -304,10 +306,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
                 PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
                 PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-                for(String sampleName: tracker.getSamples()) {
+                for(Sample sample: tracker.getSamples()) {
-                    PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                    PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                     AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getPileupWithoutMappingQualityZeroReads();
-                    filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                    filteredTracker.addElements(sample,pileup.pileupElementTracker);
                 }
                 return (RBP)createNewPileup(loc,filteredTracker);
 
@@ -333,10 +335,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                 AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getPositiveStrandPileup();
-                filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                filteredTracker.addElements(sample,pileup.pileupElementTracker);
             }
             return (RBP)createNewPileup(loc,filteredTracker);
         }
@@ -362,10 +364,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                 AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getNegativeStrandPileup();
-                filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                filteredTracker.addElements(sample,pileup.pileupElementTracker);
             }
             return (RBP)createNewPileup(loc,filteredTracker);
         }
@@ -392,10 +394,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                 AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getFilteredPileup(filter);
-                filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                filteredTracker.addElements(sample,pileup.pileupElementTracker);
             }
 
             return (RBP)createNewPileup(loc,filteredTracker);
@@ -424,10 +426,10 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                 AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getBaseAndMappingFilteredPileup(minBaseQ,minMapQ);
-                filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                filteredTracker.addElements(sample,pileup.pileupElementTracker);
             }
 
             return (RBP)createNewPileup(loc,filteredTracker);
@@ -491,11 +493,11 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             PerSamplePileupElementTracker<PE> filteredTracker = new PerSamplePileupElementTracker<PE>();
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
                 AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements).getPileupForReadGroup(targetReadGroupId);
                 if(pileup != null)
-                    filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                    filteredTracker.addElements(sample,pileup.pileupElementTracker);
             }
             return (RBP)createNewPileup(loc,filteredTracker);
 
@@ -517,10 +519,14 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
         }
     }
 
-    public Collection<String> getSamples() {
+    public Collection<String> getSampleNames() {
         if(pileupElementTracker instanceof PerSamplePileupElementTracker) {
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
-            return tracker.getSamples();
+            Collection<String> sampleNames = new HashSet<String>();
+            for (Sample sample : tracker.getSamples()) {
+                sampleNames.add(sample.getId());
+            }
+            return sampleNames;
         }
         else {
             Collection<String> sampleNames = new HashSet<String>();
@@ -533,6 +539,16 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
         }
     }
 
+    @Override
+    public Collection<Sample> getSamples() {
+        if(!(pileupElementTracker instanceof PerSamplePileupElementTracker)) {
+            throw new StingException("Must be an instance of PerSampleElementTracker");
+        }
+        PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
+        return tracker.getSamples();
+    }
+
+
     /**
      * Returns a pileup randomly downsampled to the desiredCoverage.
      *
@@ -558,8 +574,8 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
 
             int current = 0;
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                PileupElementTracker<PE> perSampleElements = tracker.getElements(sampleName);
+                PileupElementTracker<PE> perSampleElements = tracker.getElements(sample);
 
                 List<PileupElement> filteredPileup = new ArrayList<PileupElement>();
                 for(PileupElement p: perSampleElements) {
@@ -569,7 +585,7 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
 
                 if(!filteredPileup.isEmpty()) {
                     AbstractReadBackedPileup<RBP,PE> pileup = createNewPileup(loc,perSampleElements);
-                    filteredTracker.addElements(sampleName,pileup.pileupElementTracker);
+                    filteredTracker.addElements(sample,pileup.pileupElementTracker);
                 }
 
                 current++;
@@ -593,7 +609,7 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
     }
 
     @Override
-    public RBP getPileupForSample(String sampleName) {
+    public RBP getPileupForSampleName(String sampleName) {
         if(pileupElementTracker instanceof PerSamplePileupElementTracker) {
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             return (RBP)createNewPileup(loc,tracker.getElements(sampleName));
@@ -615,6 +631,29 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
         }
     }
 
+    @Override
+    public RBP getPileupForSample(Sample sample) {
+        if(pileupElementTracker instanceof PerSamplePileupElementTracker) {
+            PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
+            return (RBP)createNewPileup(loc,tracker.getElements(sample));
+        }
+        else {
+            UnifiedPileupElementTracker<PE> filteredTracker = new UnifiedPileupElementTracker<PE>();
+            for(PE p: pileupElementTracker) {
+                SAMRecord read = p.getRead();
+                if(sample != null) {
+                    if(read.getReadGroup() != null && sample.getId().equals(read.getReadGroup().getSample()))
+                        filteredTracker.add(p);
+                }
+                else {
+                    if(read.getReadGroup() == null || read.getReadGroup().getSample() == null)
+                        filteredTracker.add(p);
+                }
+            }
+            return filteredTracker.size()>0 ? (RBP)createNewPileup(loc,filteredTracker) : null;
+        }
+    }
+
     // --------------------------------------------------------
     //
     // iterators
@@ -693,8 +732,8 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
             boolean hasSecondaryBases = false;
 
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                hasSecondaryBases |= createNewPileup(loc,tracker.getElements(sampleName)).hasSecondaryBases();
+                hasSecondaryBases |= createNewPileup(loc,tracker.getElements(sample)).hasSecondaryBases();
             }
 
             return hasSecondaryBases;
@@ -722,8 +761,8 @@ public abstract class AbstractReadBackedPileup<RBP extends AbstractReadBackedPil
 
         if(pileupElementTracker instanceof PerSamplePileupElementTracker) {
             PerSamplePileupElementTracker<PE> tracker = (PerSamplePileupElementTracker<PE>)pileupElementTracker;
-            for(String sampleName: tracker.getSamples()) {
+            for(Sample sample: tracker.getSamples()) {
-                int[] countsBySample = createNewPileup(loc,tracker.getElements(sampleName)).getBaseCounts();
+                int[] countsBySample = createNewPileup(loc,tracker.getElements(sample)).getBaseCounts();
                 for(int i = 0; i < counts.length; i++)
                     counts[i] += countsBySample[i];
             }

java/src/org/broadinstitute/sting/utils/pileup/MergingPileupElementIterator.java

@@ -25,6 +25,7 @@
 package org.broadinstitute.sting.utils.pileup;
 
 import net.sf.picard.util.PeekableIterator;
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 
 import java.util.PriorityQueue;
 import java.util.Comparator;
@@ -41,8 +42,8 @@ class MergingPileupElementIterator<PE extends PileupElement> implements Iterator
 
     public MergingPileupElementIterator(PerSamplePileupElementTracker<PE> tracker) {
         perSampleIterators = new PriorityQueue<PeekableIterator<PE>>(tracker.getSamples().size(),new PileupElementIteratorComparator());
-        for(String sampleName: tracker.getSamples()) {
+        for(Sample sample: tracker.getSamples()) {
-            PileupElementTracker<PE> trackerPerSample = tracker.getElements(sampleName);
+            PileupElementTracker<PE> trackerPerSample = tracker.getElements(sample);
             if(trackerPerSample.size() != 0)
                 perSampleIterators.add(new PeekableIterator<PE>(trackerPerSample.iterator()));
         }

java/src/org/broadinstitute/sting/utils/pileup/PileupElementTracker.java

@@ -24,6 +24,8 @@
 
 package org.broadinstitute.sting.utils.pileup;
 
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
+
 import java.util.*;
 
 /**
@@ -58,19 +60,21 @@ class UnifiedPileupElementTracker<PE extends PileupElement> extends PileupElemen
 }
 
 class PerSamplePileupElementTracker<PE extends PileupElement> extends PileupElementTracker<PE> {
-    private final Map<String,PileupElementTracker<PE>> pileup;
+    private final Map<Sample,PileupElementTracker<PE>> pileup;
+    private final Map<String, Sample> sampleNames = new HashMap<String, Sample>();
     private int size = 0;
 
     public PerSamplePileupElementTracker() {
-        pileup = new HashMap<String,PileupElementTracker<PE>>();
+        pileup = new HashMap<Sample,PileupElementTracker<PE>>();
     }
 
-    public PerSamplePileupElementTracker(Map<String,AbstractReadBackedPileup<?,PE>> pileupsBySample) {
+    public PerSamplePileupElementTracker(Map<Sample,AbstractReadBackedPileup<?,PE>> pileupsBySample) {
-        pileup = new HashMap<String,PileupElementTracker<PE>>();
+        pileup = new HashMap<Sample,PileupElementTracker<PE>>();
-        for(Map.Entry<String,AbstractReadBackedPileup<?,PE>> entry: pileupsBySample.entrySet()) {
+        for(Map.Entry<Sample,AbstractReadBackedPileup<?,PE>> entry: pileupsBySample.entrySet()) {
-            String sampleName = entry.getKey();
+            Sample sample = entry.getKey();
             AbstractReadBackedPileup<?,PE> pileupBySample = entry.getValue();
-            pileup.put(sampleName,pileupBySample.pileupElementTracker);
+            pileup.put(sample,pileupBySample.pileupElementTracker);
+            sampleNames.put(sample.getId(), sample);
         }
     }
 
@@ -78,16 +82,21 @@ class PerSamplePileupElementTracker<PE extends PileupElement> extends PileupElem
      * Gets a list of all the samples stored in this pileup.
      * @return List of samples in this pileup.
      */
-    public Collection<String> getSamples() {
+    public Collection<Sample> getSamples() {
         return pileup.keySet();
     }
 
-    public PileupElementTracker<PE> getElements(final String sample) {
+    public PileupElementTracker<PE> getElements(final Sample sample) {
         return pileup.get(sample);
     }
 
-    public void addElements(final String sample, PileupElementTracker<PE> elements) {
+    public PileupElementTracker<PE> getElements(final String sampleName) {
+        return pileup.get(sampleNames.get(sampleName));
+    }
+
+    public void addElements(final Sample sample, PileupElementTracker<PE> elements) {
         pileup.put(sample,elements);
+        sampleNames.put(sample.getId(), sample);
         size += elements.size();
     }
 

java/src/org/broadinstitute/sting/utils/pileup/ReadBackedExtendedEventPileup.java

@@ -24,6 +24,7 @@
 
 package org.broadinstitute.sting.utils.pileup;
 
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import org.broadinstitute.sting.utils.collections.Pair;
 
@@ -122,14 +123,20 @@ public interface ReadBackedExtendedEventPileup extends ReadBackedPileup {
      * Gets a list of all the samples stored in this pileup.
      * @return List of samples in this pileup.
      */
-    public Collection<String> getSamples();
+    public Collection<String> getSampleNames();
+
+        /**
+     * Gets a list of all the samples stored in this pileup.
+     * @return List of samples in this pileup.
+     */
+    public Collection<Sample> getSamples();
 
     /**
      * Gets the particular subset of this pileup with the given sample name.
-     * @param sampleName Name of the sample to use.
+     * @param sample Name of the sample to use.
      * @return A subset of this pileup containing only reads with the given sample.
      */
-    public ReadBackedExtendedEventPileup getPileupForSample(String sampleName);
+    public ReadBackedExtendedEventPileup getPileupForSample(Sample sample);
 
     public Iterable<ExtendedEventPileupElement> toExtendedIterable();
 
@@ -212,5 +219,5 @@ public interface ReadBackedExtendedEventPileup extends ReadBackedPileup {
      * Get an array of the mapping qualities
      * @return
      */
-    public byte[] getMappingQuals();    
+    public byte[] getMappingQuals();
 }

java/src/org/broadinstitute/sting/utils/pileup/ReadBackedExtendedEventPileupImpl.java

@@ -23,6 +23,7 @@
  */
 package org.broadinstitute.sting.utils.pileup;
 
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.utils.exceptions.ReviewedStingException;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import org.broadinstitute.sting.utils.collections.Pair;
@@ -55,7 +56,8 @@ public class ReadBackedExtendedEventPileupImpl extends AbstractReadBackedPileup<
         this.nInsertions = nInsertions;
     }
 
-    public ReadBackedExtendedEventPileupImpl(GenomeLoc loc, Map<String,? extends ReadBackedExtendedEventPileupImpl> pileupElementsBySample) {
+    // this is the good new one
+    public ReadBackedExtendedEventPileupImpl(GenomeLoc loc, Map<Sample,? extends ReadBackedExtendedEventPileupImpl> pileupElementsBySample) {
         super(loc,pileupElementsBySample);
     }
 
@@ -221,5 +223,5 @@ public class ReadBackedExtendedEventPileupImpl extends AbstractReadBackedPileup<
         } else {
             return "-"+new String(refBases,1,length).toUpperCase();
         }
-    }    
+    }
 }

java/src/org/broadinstitute/sting/utils/pileup/ReadBackedPileup.java

@@ -24,6 +24,7 @@
 
 package org.broadinstitute.sting.utils.pileup;
 
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import org.broadinstitute.sting.gatk.iterators.IterableIterator;
 import net.sf.samtools.SAMRecord;
@@ -128,18 +129,31 @@ public interface ReadBackedPileup extends Iterable<PileupElement> {
      */
     public ReadBackedPileup getPileupForReadGroup(String readGroupId);
 
-    /**
+     /**
      * Gets a collection of all the samples stored in this pileup.
      * @return Collection of samples in this pileup.
      */
-    public Collection<String> getSamples();
+    public Collection<Sample> getSamples();
+
+    /**
+     * Gets a collection of *names* of all the samples stored in this pileup.
+     * @return Collection of names
+     */
+    public Collection<String> getSampleNames();
 
     /**
      * Gets the particular subset of this pileup with the given sample name.
      * @param sampleName Name of the sample to use.
      * @return A subset of this pileup containing only reads with the given sample.
      */
-    public ReadBackedPileup getPileupForSample(String sampleName);
+    public ReadBackedPileup getPileupForSampleName(String sampleName);
+
+    /**
+     * Gets the particular subset of this pileup with the given sample.
+     * @param sample Sample to use.
+     * @return A subset of this pileup containing only reads with the given sample.
+     */
+    public ReadBackedPileup getPileupForSample(Sample sample);
     
     /**
      * Simple useful routine to count the number of deletion bases in this pileup

java/src/org/broadinstitute/sting/utils/pileup/ReadBackedPileupImpl.java

@@ -23,6 +23,7 @@
  */
 package org.broadinstitute.sting.utils.pileup;
 
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.utils.GenomeLoc;
 import net.sf.samtools.SAMRecord;
 
@@ -47,7 +48,7 @@ public class ReadBackedPileupImpl extends AbstractReadBackedPileup<ReadBackedPil
         super(loc,pileupElements);
     }
 
-    public ReadBackedPileupImpl(GenomeLoc loc, Map<String,ReadBackedPileupImpl> pileupElementsBySample) {
+    public ReadBackedPileupImpl(GenomeLoc loc, Map<Sample,ReadBackedPileupImpl> pileupElementsBySample) {
         super(loc,pileupElementsBySample);
     }
 

java/test/org/broadinstitute/sting/gatk/datasources/providers/LocusViewTemplate.java

@@ -7,6 +7,7 @@ import net.sf.samtools.*;
 import org.broadinstitute.sting.BaseTest;
 import org.broadinstitute.sting.gatk.ReadProperties;
 import org.broadinstitute.sting.gatk.executive.WindowMaker;
+import org.broadinstitute.sting.gatk.datasources.sample.SampleDataSource;
 import org.broadinstitute.sting.gatk.datasources.shards.LocusShard;
 import org.broadinstitute.sting.gatk.datasources.shards.Shard;
 import org.broadinstitute.sting.gatk.datasources.shards.MockLocusShard;
@@ -52,7 +53,7 @@ public abstract class LocusViewTemplate extends BaseTest {
 
         GenomeLoc shardBounds = genomeLocParser.createGenomeLoc("chr1", 1, 5);
         Shard shard = new LocusShard(new SAMDataSource(Collections.<SAMReaderID>emptyList(),genomeLocParser),Collections.singletonList(shardBounds),Collections.<SAMReaderID,SAMFileSpan>emptyMap());
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, null, genomeLocParser, window.getLocus(), window, null, null);
 
@@ -68,7 +69,7 @@ public abstract class LocusViewTemplate extends BaseTest {
 
         GenomeLoc shardBounds = genomeLocParser.createGenomeLoc("chr1", 1, 5);
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(shardBounds));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
 
@@ -83,7 +84,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -97,7 +98,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -111,7 +112,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -125,7 +126,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 6, 15)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -139,7 +140,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -154,7 +155,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read1, read2);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -173,7 +174,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read1, read2, read3, read4);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -192,7 +193,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read1, read2, read3, read4);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -213,7 +214,7 @@ public abstract class LocusViewTemplate extends BaseTest {
         SAMRecordIterator iterator = new SAMRecordIterator(read1, read2, read3, read4, read5, read6);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 1, 10)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(), LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(), LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);
@@ -241,7 +242,7 @@ public abstract class LocusViewTemplate extends BaseTest {
                                                            read07, read08, read09, read10, read11, read12);
 
         Shard shard = new MockLocusShard(genomeLocParser,Collections.singletonList(genomeLocParser.createGenomeLoc("chr1", 6, 15)));
-        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS);
+        WindowMaker windowMaker = new WindowMaker(shard,genomeLocParser,iterator,shard.getGenomeLocs(),LocusIteratorByState.NO_FILTERS, new SampleDataSource());
         WindowMaker.WindowMakerIterator window = windowMaker.next();
         LocusShardDataProvider dataProvider = new LocusShardDataProvider(shard, window.getSourceInfo(), genomeLocParser, window.getLocus(), window, null, null);
         LocusView view = createView(dataProvider);

java/test/org/broadinstitute/sting/gatk/datasources/sample/SampleDataSourceUnitTest.java

@@ -39,7 +39,7 @@ public class SampleDataSourceUnitTest extends BaseTest {
     public void basicLoadSampleFileTest() {
         File sampleFile = new File(sampleFilesDir + "basicSampleFile.yaml");
         SampleDataSource s = new SampleDataSource(header, makeFileList(sampleFile));
-        Assert.assertTrue(s.sampleCount() == 4);
+        Assert.assertTrue(s.sampleCount() == 5);
         Sample sampleA = s.getSampleById("sampleA");
         Sample sampleB = s.getSampleById("sampleB");
         Assert.assertTrue(sampleB.getMother() == sampleA);
@@ -78,7 +78,7 @@ public class SampleDataSourceUnitTest extends BaseTest {
         File sampleFile = new File(sampleFilesDir + "basicSampleFileWithAlias.yaml");
         SampleDataSource s = new SampleDataSource(header, makeFileList(sampleFile));
         // this file has two samples, but one has an alias. let's make sure that checks out...
-        Assert.assertTrue(s.sampleCount() == 2);
+        Assert.assertTrue(s.sampleCount() == 3);
         Assert.assertTrue(s.getSampleById("sampleA") == s.getSampleById("sampleC"));
     }
 
@@ -129,20 +129,21 @@ public class SampleDataSourceUnitTest extends BaseTest {
         files.add(new File(sampleFilesDir + "basicSampleFileExt.yaml"));
         files.add(new File(sampleFilesDir + "basicSampleFileExt2.yaml"));
         SampleDataSource s = new SampleDataSource(header, files);
-        Assert.assertTrue(s.sampleCount() == 5);
+        Assert.assertTrue(s.sampleCount() == 6);
         Assert.assertTrue(s.getSampleById("sampleE").getProperty("propC").equals("valC"));
         Assert.assertTrue(s.getSampleById("sampleA").getProperty("propA").equals("valA"));
     }
 
     /**
      * testing getSamplesWithProperty
-     * in this file there are 4 samples - 2 with population "CEU", 1 with population "ABC", 1 with no population
+     * in this file there are 5 samples - 2 with population "CEU", 1 with population "ABC", 1 with no population,
+     * and then the default null sample
      */
     @Test()
     public void getSamplesWithPropertyTest() {
         File sampleFile = new File(sampleFilesDir + "sampleFileWithProperties.yaml");
         SampleDataSource s = new SampleDataSource(header, makeFileList(sampleFile));
-        Assert.assertTrue(s.sampleCount() == 4);
+        Assert.assertTrue(s.sampleCount() == 5);
         Set<Sample> ceuSamples = s.getSamplesWithProperty("population", "CEU");
         Assert.assertTrue(ceuSamples.size() == 2);
 
@@ -209,7 +210,7 @@ public class SampleDataSourceUnitTest extends BaseTest {
 
         File sampleFile = new File(sampleFilesDir + "sampleFileWithProperties.yaml");
         SampleDataSource s = new SampleDataSource(header, makeFileList(sampleFile));
-        Assert.assertTrue(s.sampleCount() == 4);
+        Assert.assertTrue(s.sampleCount() == 5);
 
         List<Allele> alleleCollection = new ArrayList<Allele>();
         Allele a1 = Allele.create("A", true);

java/test/org/broadinstitute/sting/gatk/iterators/LocusIteratorByStateUnitTest.java

@@ -9,6 +9,7 @@ import org.testng.Assert;
 import org.broadinstitute.sting.BaseTest;
 import org.broadinstitute.sting.gatk.ReadProperties;
 import org.broadinstitute.sting.gatk.arguments.ValidationExclusion;
+import org.broadinstitute.sting.gatk.datasources.sample.SampleDataSource;
 import org.broadinstitute.sting.gatk.datasources.simpleDataSources.SAMReaderID;
 import org.broadinstitute.sting.gatk.contexts.AlignmentContext;
 import org.broadinstitute.sting.utils.GenomeLocParser;
@@ -63,7 +64,7 @@ public class LocusIteratorByStateUnitTest extends BaseTest {
         List<SAMRecord> reads = Arrays.asList(before,during,after);
 
         // create the iterator by state with the fake reads and fake records
-        li = new LocusIteratorByState(new FakeCloseableIterator<SAMRecord>(reads.iterator()),readAttributes,genomeLocParser);
+        li = new LocusIteratorByState(new FakeCloseableIterator<SAMRecord>(reads.iterator()),readAttributes,genomeLocParser, new SampleDataSource());
 
         boolean foundExtendedEventPileup = false;
         while (li.hasNext()) {
@@ -115,7 +116,7 @@ public class LocusIteratorByStateUnitTest extends BaseTest {
         List<SAMRecord> reads = Arrays.asList(before,during,after);
 
         // create the iterator by state with the fake reads and fake records
-        li = new LocusIteratorByState(new FakeCloseableIterator<SAMRecord>(reads.iterator()),readAttributes,genomeLocParser);
+        li = new LocusIteratorByState(new FakeCloseableIterator<SAMRecord>(reads.iterator()),readAttributes,genomeLocParser, new SampleDataSource());
 
         boolean foundExtendedEventPileup = false;
         while (li.hasNext()) {

java/test/org/broadinstitute/sting/utils/pileup/ReadBackedPileupUnitTest.java

@@ -28,6 +28,7 @@ import net.sf.samtools.SAMFileHeader;
 import net.sf.samtools.SAMReadGroupRecord;
 import net.sf.samtools.SAMRecord;
 import org.testng.Assert;
+import org.broadinstitute.sting.gatk.datasources.sample.Sample;
 import org.broadinstitute.sting.utils.sam.ArtificialSAMUtils;
 
 import org.testng.annotations.Test;
@@ -143,9 +144,9 @@ public class ReadBackedPileupUnitTest {
         ReadBackedPileupImpl sample2Pileup = new ReadBackedPileupImpl(null,
                                                                       Arrays.asList(read2,read4),
                                                                       Arrays.asList(1,1));
-        Map<String,ReadBackedPileupImpl> sampleToPileupMap = new HashMap<String,ReadBackedPileupImpl>();
+        Map<Sample,ReadBackedPileupImpl> sampleToPileupMap = new HashMap<Sample,ReadBackedPileupImpl>();
-        sampleToPileupMap.put(readGroupOne.getSample(),sample1Pileup);
+        sampleToPileupMap.put(new Sample(readGroupOne.getSample()),sample1Pileup);
-        sampleToPileupMap.put(readGroupTwo.getSample(),sample2Pileup);
+        sampleToPileupMap.put(new Sample(readGroupTwo.getSample()),sample2Pileup);
 
         ReadBackedPileup compositePileup = new ReadBackedPileupImpl(null,sampleToPileupMap);