From 781c0c33a42a4df39624e33da7eb300a580c61b3 Mon Sep 17 00:00:00 2001
From: Ryan Poplin <rpoplin@broadinstitute.org>
Date: Sat, 2 Jul 2011 17:55:10 -0400
Subject: [PATCH] Use the worst X% of calls in addition to the bad training
 sites list. Don't include the already added calls in the calculation of X%

---
 .../walkers/variantrecalibration/VariantDataManager.java     | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/public/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/VariantDataManager.java b/public/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/VariantDataManager.java
index cd739f9fc..309d1b8fa 100755
--- a/public/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/VariantDataManager.java
+++ b/public/java/src/org/broadinstitute/sting/gatk/walkers/variantrecalibration/VariantDataManager.java
@@ -151,13 +151,13 @@ public class VariantDataManager {
         int index = 0, numAdded = 0;
         while( numAdded < numToAdd ) {
             final VariantDatum datum = data.get(index++);
-            if( !datum.failingSTDThreshold && !Double.isInfinite(datum.lod) ) {
+            if( !datum.atAntiTrainingSite && !datum.failingSTDThreshold && !Double.isInfinite(datum.lod) ) {
                 datum.atAntiTrainingSite = true;
                 trainingData.add( datum );
                 numAdded++;
             }
         }
-        logger.info( "Additionally training with worst " + (float) bottomPercentage * 100.0f + "% of passing data --> " + (trainingData.size() - numBadSitesAdded) + " variants with LOD <= " + String.format("%.4f", data.get(index).lod) + "." );
+        logger.info( "Additionally training with worst " + String.format("%0.3f", (float) bottomPercentage * 100.0f) + "% of passing data --> " + (trainingData.size() - numBadSitesAdded) + " variants with LOD <= " + String.format("%.4f", data.get(index).lod) + "." );
         return trainingData;
     }
 
@@ -243,6 +243,7 @@ public class VariantDataManager {
         datum.isKnown = false;
         datum.atTruthSite = false;
         datum.atTrainingSite = false;
+        datum.atAntiTrainingSite = false;
         datum.prior = 2.0;
         datum.consensusCount = 0;