diff --git a/java/src/org/broadinstitute/sting/gatk/refdata/AnnotatorROD.java b/java/src/org/broadinstitute/sting/gatk/refdata/AnnotatorROD.java
new file mode 100755
index 000000000..cdb75fade
--- /dev/null
+++ b/java/src/org/broadinstitute/sting/gatk/refdata/AnnotatorROD.java
@@ -0,0 +1,134 @@
+package org.broadinstitute.sting.gatk.refdata;
+
+import java.io.File;
+import java.io.FileNotFoundException;
+import java.io.IOException;
+import java.util.ArrayList;
+import java.util.Arrays;
+
+import org.apache.log4j.Logger;
+import org.broadinstitute.sting.utils.Utils;
+import org.broadinstitute.sting.utils.xReadLines;
+
+/**
+ * <pre>
+ * This slightly modified TabularROD format is used as input to the
+ * GenomicAnnotator.
+ * The main differences from TabularROD are:
+ * - the delimiter is \t instead of \s+
+ * - incomplete records are allowed (eg. where some values are left blank)
+ * - the header column is just the first non-comment, non-empty row in the file.
+ *    It no longer has to start with the "HEADER" keyword.
+ * </pre>
+ *
+ * More details can be found here:  http://www.broadinstitute.org/gsa/wiki/index.php/GenomicAnnotator
+ */
+public class AnnotatorROD extends TabularROD {
+
+    private static Logger logger = Logger.getLogger(AnnotatorROD.class);
+
+    /** Special column names */
+    //public static final String CHRPOS_COLUMN = "chrpos";
+    public static final String HAPLOTYPE_REFERENCE_COLUMN = "haplotypeReference";
+    public static final String HAPLOTYPE_ALTERNATE_COLUMN = "haplotypeAlternate";
+    public static final String HAPLOTYPE_STRAND_COLUMN = "haplotypeStrand";
+
+
+    private static int parsedRecords = 0;
+
+   /**
+    * Constructor.
+    *
+    * @param name The binding name provided as the first in the list of -B args.
+    */
+    public AnnotatorROD(String name) {
+        super(name, new ArrayList<String>());
+
+        setDelimiter("\t", "\\t");
+    }
+
+
+    /**
+     * Walks through the source files looking for the header line, which it
+     * returns as a list of strings.
+     *
+     * @param source
+     * @return
+     */
+    public Object initialize(final File source) throws FileNotFoundException {
+        ArrayList<String> header = AnnotatorROD.readHeader(source);
+        return header;
+    }
+
+
+
+
+    /**
+     * Finds and parses the header string in the file.
+     * @param source The source file.
+     * @return A List of the column names parsed out of the header.
+     * @throws FileNotFoundException
+     */
+    public static ArrayList<String> readHeader(final File source) throws FileNotFoundException {
+        ArrayList<String> header = null;
+        xReadLines reader = new xReadLines(source);
+
+        //find the 1st line that's non-empty and not a comment
+        for ( String line : reader ) {
+            line = line.trim();
+            if ( line.isEmpty() || line.startsWith("#") ) {
+                continue;
+            }
+
+            header = new ArrayList<String>(Arrays.asList(line.split(DELIMITER_REGEX)));
+            break;
+        }
+
+        // check that we found the header
+        if ( header == null ) {
+            throw new RuntimeException("No header in " + source + ". All lines are either comments or empty.");
+        }
+
+        logger.info(String.format("Found header line containing %d columns:\n[%s]", header.size(), Utils.join("\t", header)));
+
+        try {
+            reader.close();
+        } catch ( IOException e ) {
+            throw new RuntimeException(e);
+        }
+
+        return header;
+    }
+
+
+    /**
+     * Used by ROD management system to set the data in this ROD associated with a line in a rod
+     *
+     * @param headerObj
+     * @param parts
+     * @return
+     * @throws IOException
+     */
+    public boolean parseLine(final Object headerObj, final String[] parts) throws IOException {
+        ArrayList<String> header = (ArrayList<String>)(headerObj);
+
+        //save the header in the super-class
+        getHeader().addAll(header);
+
+        if ( parts.length == 0 || parts[0].startsWith("#") || header.get(0).equals(parts[0]) /* Skip the header line */ )
+            return false;
+
+
+        if ( header.size() < parts.length) {
+            throw new IOException(String.format("Encountered line with more columns than have names in the header. Header has %d columns, this line has %d columns.", header.size(), parts.length));
+        }
+
+        for ( int i = 0; i < parts.length; i++ ) {
+            put(header.get(i), parts[i]);
+        }
+
+        if ( printRecordsParsed ) System.out.printf("Parsed %d records %s%n", ++parsedRecords, this);
+
+        return true;
+    }
+}
diff --git a/java/src/org/broadinstitute/sting/gatk/refdata/tracks/builders/RODTrackBuilder.java b/java/src/org/broadinstitute/sting/gatk/refdata/tracks/builders/RODTrackBuilder.java
index 4534cf6d6..28a382582 100644
--- a/java/src/org/broadinstitute/sting/gatk/refdata/tracks/builders/RODTrackBuilder.java
+++ b/java/src/org/broadinstitute/sting/gatk/refdata/tracks/builders/RODTrackBuilder.java
@@ -61,6 +61,7 @@ public class RODTrackBuilder implements RMDTrackBuilder {
         Types.put("GELI", rodGELI.class);
         Types.put("RefSeq", rodRefSeq.class);
         Types.put("Table", TabularROD.class);
+        Types.put("AnnotatorInputTable", AnnotatorROD.class);
         Types.put("PooledEM", PooledEMSNPROD.class);
         Types.put("CleanedOutSNP", CleanedOutSNPROD.class);
         Types.put("SangerSNP", SangerSNPROD.class);