diff --git a/python/firehose_out_email.py b/python/firehose_out_email.py
new file mode 100755
index 000000000..b2c6700f6
--- /dev/null
+++ b/python/firehose_out_email.py
@@ -0,0 +1,35 @@
+#this script produces the output to go in emails 
+import os
+import re
+import sample_lister
+
+class SampSet(sample_lister.SampleSet)
+    def __init__(self, sampset, pathname):
+        self.sampset=sampset
+        self.pathname=pathname
+        sample_lister.SampleSet.__init__(self, sampset, pathname)
+    def evalout(self):
+        '''This produced the output that needs to go in the emails'''
+        filename = "/humgen/gsa-firehose/firehose/firehose_output/trunk/Sample_Set/" + self.sampset +"/UnifiedGenotyper/"+ self.sampset+".filtered.eval"
+        evalfile = open(filename, "r").read()
+        annotations= ["all", "novel", "known", "snp_at_known_non_snps", "filtered"]
+        variant=dict(zip(annotations, ('','','','','')))
+        ratio=dict(zip(annotations, ('','','','','')))
+        bpre=re.compile("all,summary,variant_counts +n bases covered +(\d+)")
+        size=repr(bpre.search(evalfile).group(1))
+        for a in annotations:
+            anregexv  = re.compile(a + ",summary,variant_counts +variants +(\d+)")
+            variant[a] = repr(anregexv.search(evalfile).group(1))
+            anregexr = re.compile(a + ",summary,transitions_transversions +ratio +(\d+.\d+)")
+            print (anregexr.search(evalfile))
+            ratio[a] = repr(anregexr.search(evalfile).group(1))
+            print (a)
+            print (ratio[a])
+        print("Samples processed:\n\n Target size: \t" +size+"  bp \n\n\t\t\t\t\t Variants \t\t Ti/TV \n (true positives)\t All \t\t " +variant["all"]+ " \t\t " + ratio["all"] +" \n \t\t\t Known \t\t " +variant["known"]+ " \t\t " + ratio['known']+" \n \t\t\t Novel \t\t " +variant["novel"]+" \t\t " + ratio['novel']+  " \n*************************************************************************\n (false  \tSNPS at known indels \t " +variant["snp_at_known_non_snps"]+"\t\t\t " + ratio['snp_at_known_non_snps']+ " \n positives) \t\t filtered \t " +variant["filtered"]+" \t\t " + ratio['filtered'] )
+
+
+#EOMI=SampSet("EOMI_Kathiresan_NHGRI", "test")
+#EOMI.evalout() <-this and the line above are examples
+
+#TODO: make this send the email when run
+#TODO: make this find the list of bams, bed files, and annotated vcfs.