Renaming quantitative trait to OtherPhenotype which is now a String not a double

-- we can now use PED file to represent population data or other arbitrary phenotype data, not just doubles

public/java/src/org/broadinstitute/sting/gatk/samples/Affection.java

@@ -28,8 +28,8 @@ package org.broadinstitute.sting.gatk.samples;
  * Categorical sample trait for association and analysis
  *
  * Samples can have unknown status, be affected or unaffected by the
- * categorical trait, or they can be marked as actually having a
+ * categorical trait, or they can be marked as actually having an
- * quantitative trait value (stored in an associated value in the Sample class)
+ * other trait value (stored in an associated value in the Sample class)
  *
  * @author Mark DePristo
  * @since Sept. 2011
@@ -41,6 +41,6 @@ public enum Affection {
     AFFECTED,
     /** Unaffected by the disease */
     UNAFFECTED,
-    /** A quantitative trait: value of the trait is stored elsewhere */
+    /** An "other" trait: value of the trait is stored elsewhere and is an arbitrary string */
-    QUANTITATIVE
+    OTHER
 }

public/java/src/org/broadinstitute/sting/gatk/samples/PedReader.java

@@ -63,6 +63,8 @@ import java.util.*;
  *  A PED file must have 1 and only 1 phenotype in the sixth column. The phenotype can be either a
  *  quantitative trait or an affection status column: PLINK will automatically detect which type
  *  (i.e. based on whether a value other than 0, 1, 2 or the missing genotype code is observed).
+ *  Note that the GATK actually supports arbitrary values for quantitative trait -- not just doubles --
+ *  and are actually representing these values as strings instead of doubles
  *
  *  NOTE Quantitative traits with decimal points must be coded with a period/full-stop character and
  *  not a comma, i.e. 2.394 not 2,394
@@ -212,7 +214,7 @@ public class PedReader {
             splits.add(parts);
             lineNo++;
         }
-        logger.info("Trait is quantitative? " + isQT);
+        logger.info("Phenotype is other? " + isQT);
 
         // now go through and parse each record
         lineNo = 1;
@@ -220,7 +222,7 @@ public class PedReader {
         for ( final String[] parts : splits ) {
             String familyID = null, individualID, paternalID = null, maternalID = null;
             Gender sex = Gender.UNKNOWN;
-            double quantitativePhenotype = Sample.UNSET_QT;
+            String quantitativePhenotype = Sample.UNSET_QT;
             Affection affection = Affection.UNKNOWN;
 
             if ( familyPos != -1 ) familyID = maybeMissing(parts[familyPos]);
@@ -239,8 +241,8 @@ public class PedReader {
                     if ( parts[phenotypePos].equals(MISSING_VALUE1) )
                         affection = Affection.UNKNOWN;
                     else {
-                        affection = Affection.QUANTITATIVE;
+                        affection = Affection.OTHER;
-                        quantitativePhenotype = Double.valueOf(parts[phenotypePos]);
+                        quantitativePhenotype = parts[phenotypePos];
                     }
                 } else {
                     if ( parts[phenotypePos].equals(MISSING_VALUE1) ) affection = Affection.UNKNOWN;

public/java/src/org/broadinstitute/sting/gatk/samples/Sample.java

@@ -12,22 +12,22 @@ import java.util.Map;
 public class Sample implements Comparable<Sample> { // implements java.io.Serializable {
     final private String familyID, paternalID, maternalID;
     final private Gender gender;
-    final private double quantitativePhenotype;
+    final private String otherPhenotype;
     final private Affection affection;
     final private String ID;
     final private SampleDB infoDB;
     final private Map<String, Object> properties = new HashMap<String, Object>();
 
-    public final static double UNSET_QT = Double.NaN;
+    public final static String UNSET_QT = null;
 
     public Sample(final String ID, final SampleDB infoDB,
                   final String familyID, final String paternalID, final String maternalID,
-                  final Gender gender, final Affection affection, final double quantitativePhenotype) {
+                  final Gender gender, final Affection affection, final String otherPhenotype) {
         this.familyID = familyID;
         this.paternalID = paternalID;
         this.maternalID = maternalID;
         this.gender = gender;
-        this.quantitativePhenotype = quantitativePhenotype;
+        this.otherPhenotype = otherPhenotype;
         this.affection = affection;
         this.ID = ID;
         this.infoDB = infoDB;
@@ -35,8 +35,8 @@ public class Sample implements Comparable<Sample> { // implements java.io.Serial
 
     protected Sample(final String ID,
                      final String familyID, final String paternalID, final String maternalID,
-                     final Gender gender, final Affection affection, final double quantitativePhenotype) {
+                     final Gender gender, final Affection affection, final String otherPhenotype) {
-        this(ID, null, familyID, paternalID, maternalID, gender, affection, quantitativePhenotype);
+        this(ID, null, familyID, paternalID, maternalID, gender, affection, otherPhenotype);
     }
 
     protected Sample(final String ID,
@@ -51,8 +51,8 @@ public class Sample implements Comparable<Sample> { // implements java.io.Serial
         this(ID, infoDB, familyID, paternalID, maternalID, gender, Affection.UNKNOWN, UNSET_QT);
     }
 
-    public Sample(final String ID, final SampleDB infoDB, final Affection affection, final double quantitativePhenotype) {
+    public Sample(final String ID, final SampleDB infoDB, final Affection affection, final String otherPhenotype) {
-        this(ID, infoDB, null, null, null, Gender.UNKNOWN, affection, quantitativePhenotype);
+        this(ID, infoDB, null, null, null, Gender.UNKNOWN, affection, otherPhenotype);
     }
 
     public Sample(String id, SampleDB infoDB) {
@@ -86,12 +86,12 @@ public class Sample implements Comparable<Sample> { // implements java.io.Serial
         return affection;
     }
 
-    public boolean hasQuantitativeTrait() {
+    public boolean hasOtherPhenotype() {
-        return affection == Affection.QUANTITATIVE;
+        return affection == Affection.OTHER;
     }
 
-    public double getQuantitativePhenotype() {
+    public String getOtherPhenotype() {
-        return quantitativePhenotype;
+        return otherPhenotype;
     }
 
     /**
@@ -127,7 +127,7 @@ public class Sample implements Comparable<Sample> { // implements java.io.Serial
     public String toString() {
         return String.format("Sample %s fam=%s dad=%s mom=%s gender=%s affection=%s qt=%s props=%s",
                 getID(), getFamilyID(), getPaternalID(), getMaternalID(), getGender(), getAffection(),
-                getQuantitativePhenotype(), properties);
+                getOtherPhenotype(), properties);
     }
 
 //    // -------------------------------------------------------------------------------------
@@ -174,7 +174,7 @@ public class Sample implements Comparable<Sample> { // implements java.io.Serial
                     equalOrNull(paternalID, otherSample.paternalID) &&
                     equalOrNull(maternalID, otherSample.maternalID) &&
                     equalOrNull(gender, otherSample.gender) &&
-                    equalOrNull(quantitativePhenotype, otherSample.quantitativePhenotype) &&
+                    equalOrNull(otherPhenotype, otherSample.otherPhenotype) &&
                     equalOrNull(affection, otherSample.affection) &&
                     equalOrNull(properties, otherSample.properties);
         }
@@ -215,7 +215,7 @@ public class Sample implements Comparable<Sample> { // implements java.io.Serial
                     mergeValues(prev.getID(), "Material_ID", prev.getMaternalID(), next.getMaternalID(), null),
                     mergeValues(prev.getID(), "Gender", prev.getGender(), next.getGender(), Gender.UNKNOWN),
                     mergeValues(prev.getID(), "Affection", prev.getAffection(), next.getAffection(), Affection.UNKNOWN),
-                    mergeValues(prev.getID(), "QuantitativeTrait", prev.getQuantitativePhenotype(), next.getQuantitativePhenotype(), UNSET_QT));
+                    mergeValues(prev.getID(), "OtherPhenotype", prev.getOtherPhenotype(), next.getOtherPhenotype(), UNSET_QT));
                     //mergeValues(prev.getID(), "ExtraProperties", prev.getExtraProperties(), next.getExtraProperties(), Collections.emptyMap()));
         }
     }

public/java/src/org/broadinstitute/sting/gatk/samples/SampleDB.java

@@ -8,17 +8,6 @@ import org.broadinstitute.sting.utils.variantcontext.Genotype;
 import java.util.*;
 
 /**
- * Created by IntelliJ IDEA.
- * User: brett
- * Date: Jul 26, 2010
- * Time: 3:30:09 PM
- *
- * This class stores and manages sample metadata. This data is encoded in a sample file, which can be included
- * in the GATK by the "--samples" argument. This class reads and parses those files.
- *
- * Although there are a set of public methods for accessing sample data, they aren't used by walkers - they are really
- * only used by GenomeAnalysisEngine. An instance of GenomeAnalysisEngine has one SampleDataSource. When a walker
- * wants to access sample data, it asks GenomeAnalysis to fetch this data from its SampleDataSource.
  *
  */
 public class SampleDB {

public/java/test/org/broadinstitute/sting/gatk/samples/PedReaderUnitTest.java

@@ -156,26 +156,26 @@ public class PedReaderUnitTest extends BaseTest {
                         "fam1 mom granddad2 grandma2 2 2"));
 
         // Quantitative trait
-        new PedReaderTest("QuantitativeTrait",
+        new PedReaderTest("OtherPhenotype",
                 Arrays.asList(
-                        new Sample("s1", "fam1", null, null, Gender.MALE,   Affection.QUANTITATIVE, 1.0),
+                        new Sample("s1", "fam1", null, null, Gender.MALE,   Affection.OTHER, "1.0"),
-                        new Sample("s2", "fam2", null, null, Gender.FEMALE, Affection.QUANTITATIVE, 10.0)),
+                        new Sample("s2", "fam2", null, null, Gender.FEMALE, Affection.OTHER, "10.0")),
                 String.format("%s%n%s",
                         "fam1 s1 0 0 1 1",
                         "fam2 s2 0 0 2 10.0"));
 
-        new PedReaderTest("QuantitativeTraitWithMissing",
+        new PedReaderTest("OtherPhenotypeWithMissing",
                 Arrays.asList(
                         new Sample("s1", "fam1", null, null, Gender.MALE,   Affection.UNKNOWN, Sample.UNSET_QT),
-                        new Sample("s2", "fam2", null, null, Gender.FEMALE, Affection.QUANTITATIVE, 10.0)),
+                        new Sample("s2", "fam2", null, null, Gender.FEMALE, Affection.OTHER, "10.0")),
                 String.format("%s%n%s",
                         "fam1 s1 0 0 1 -9",
                         "fam2 s2 0 0 2 10.0"));
 
-        new PedReaderTest("QuantitativeTraitOnlyInts",
+        new PedReaderTest("OtherPhenotypeOnlyInts",
                 Arrays.asList(
-                        new Sample("s1", "fam1", null, null, Gender.MALE,   Affection.QUANTITATIVE, 1.0),
+                        new Sample("s1", "fam1", null, null, Gender.MALE,   Affection.OTHER, "1"),
-                        new Sample("s2", "fam2", null, null, Gender.FEMALE, Affection.QUANTITATIVE, 10.0)),
+                        new Sample("s2", "fam2", null, null, Gender.FEMALE, Affection.OTHER, "10")),
                 String.format("%s%n%s",
                         "fam1 s1 0 0 1 1",
                         "fam2 s2 0 0 2 10"));

public/java/test/org/broadinstitute/sting/gatk/samples/SampleUnitTest.java

@@ -6,16 +6,13 @@ import org.testng.annotations.BeforeClass;
 import org.testng.annotations.Test;
 
 /**
- * Created by IntelliJ IDEA.
+ *
- * User: brett
- * Date: Sep 9, 2010
- * Time: 8:21:00 AM
  */
 public class SampleUnitTest extends BaseTest {
     SampleDB db;
     static Sample fam1A, fam1B, fam1C;
     static Sample s1, s2;
-    static Sample trait1, trait2, trait3, trait4;
+    static Sample trait1, trait2, trait3, trait4, trait5;
 
     @BeforeClass
     public void init() {
@@ -31,7 +28,8 @@ public class SampleUnitTest extends BaseTest {
         trait1 = new Sample("t1", db, Affection.AFFECTED, Sample.UNSET_QT);
         trait2 = new Sample("t2", db, Affection.UNAFFECTED, Sample.UNSET_QT);
         trait3 = new Sample("t3", db, Affection.UNKNOWN, Sample.UNSET_QT);
-        trait4 = new Sample("t4", db, Affection.QUANTITATIVE, 1.0);
+        trait4 = new Sample("t4", db, Affection.OTHER, "1.0");
+        trait5 = new Sample("t4", db, Affection.OTHER, "CEU");
     }
 
     /**
@@ -47,13 +45,14 @@ public class SampleUnitTest extends BaseTest {
         Assert.assertEquals(null, fam1B.getMaternalID());
 
         Assert.assertEquals(Affection.AFFECTED, trait1.getAffection());
-        Assert.assertEquals(Sample.UNSET_QT, trait1.getQuantitativePhenotype());
+        Assert.assertEquals(Sample.UNSET_QT, trait1.getOtherPhenotype());
         Assert.assertEquals(Affection.UNAFFECTED, trait2.getAffection());
-        Assert.assertEquals(Sample.UNSET_QT, trait2.getQuantitativePhenotype());
+        Assert.assertEquals(Sample.UNSET_QT, trait2.getOtherPhenotype());
         Assert.assertEquals(Affection.UNKNOWN, trait3.getAffection());
-        Assert.assertEquals(Sample.UNSET_QT, trait3.getQuantitativePhenotype());
+        Assert.assertEquals(Sample.UNSET_QT, trait3.getOtherPhenotype());
-        Assert.assertEquals(Affection.QUANTITATIVE, trait4.getAffection());
+        Assert.assertEquals(Affection.OTHER, trait4.getAffection());
-        Assert.assertEquals(1.0, trait4.getQuantitativePhenotype());
+        Assert.assertEquals("1.0", trait4.getOtherPhenotype());
+        Assert.assertEquals("CEU", trait5.getOtherPhenotype());
     }
 
     @Test()