diff --git a/Scripts/filterPausing.py b/Scripts/filterPausing.py
index aca0705..04dcb60 100644
--- a/Scripts/filterPausing.py
+++ b/Scripts/filterPausing.py
@@ -2,6 +2,7 @@
 # -*- coding: utf-8 -*-
 import os
 import pandas as pd
+import statsmodels.stats.multitest as mtest
 
 dfs=[]
 for fn in snakemake.input:
@@ -10,5 +11,7 @@
     df=pd.read_csv(fn,delimiter='\t',header=None)
     dfs.append(df)
 df=pd.concat(dfs,sort=[0,1],ignore_index=True)
+df[4]=1.-df[4]/100
+df[4]=mtest.fdrcorrection(df[4])[1]
 df.to_csv(snakemake.output[1],sep='\t',index=False,header=False)
-df[df[4]>=snakemake.params[0]].to_csv(snakemake.output[0],sep='\t',index=False,header=False)
+df[df[4]<snakemake.params[0]].to_csv(snakemake.output[0],sep='\t',index=False,header=False)
diff --git a/Snakefile b/Snakefile
index 050a73c..074f07a 100644
--- a/Snakefile
+++ b/Snakefile
@@ -77,7 +77,7 @@ rule FilterPeaks:
         "Results/Significant_{prefix}.bed",
         "Results/All_{prefix}.bed"
     params:
-        config["percentile"]
+        config["padj"]
     script:
         "Scripts/filterPausing.py"
 
diff --git a/config.yaml b/config.yaml
index a781e9a..1511578 100644
--- a/config.yaml
+++ b/config.yaml
@@ -26,10 +26,12 @@ window_size: 200
 #minimal number of reads to call a peak
 minimum_intensity: 5
 
-#
+#should reads with no supporting density in the window be included
 include_spikes: FALSE
 
+#number of redistributions of reads done to obtain the distribution of
+#the expected maximum
 bootstrap_N: 10000
 
-#
-percentile: 95
+#adjusted pvalue required to report as significant
+padj: 0.05