diff --git a/script_in_snakemake/split_cpg_groups.py b/script_in_snakemake/split_cpg_groups.py
index 4c02c08..c78d630 100644
--- a/script_in_snakemake/split_cpg_groups.py
+++ b/script_in_snakemake/split_cpg_groups.py
@@ -21,13 +21,10 @@
 	-keeping strandedness and readID information
 """
 
-ch=[]
-st=[]
-pos=[]
-log=[]
-readID=[]
-target ="CG"
 
+target ="CG"
+out = open(snakemake.output[1],'w')
+out.write("\t".join(['Chr', 'Pos','Strand', 'Log.like.ratio', 'Read_ID']) + "\n")
 with open(snakemake.input[0],'r') as fh:
     next(fh)
     for line in fh:
@@ -41,29 +38,29 @@
         seq = str(fields[10])
 
         if cpg_num == 1:
-            ch.append(chrom)
-            st.append(strand)
-            pos.append(start)
-            log.append(logRatio)
-            readID.append(read_name)
+            out.write(chrom + "\t" + str(start) + "\t" + strand + "\t" + str(logRatio) + "\t" + read_name + "\n")
 
         elif cpg_num > 1:
-            index = []
-            pos.append(start)
-            for match in  re.finditer(target, seq):
-                #print(match.start())
+            # index = []
+            # pos.append(start)
+            out.write(chrom + "\t" + str(start) + "\t" + strand + "\t" + str(logRatio) + "\t" + read_name + "\n")
+            for match in re.finditer(target, seq):
+                # print(match.start())
                 index.append(match.start())
                 length = len(index)
             for i in range(1, length):
                 new_start = start + (index[i] - index[0])
-                pos.append(new_start)
-                #print(new_start)
+                # pos.append(new_start)
+                out.write(chrom + "\t" + str(new_start) + "\t" + strand + "\t" + logRatio + "\t" + read_name + "\n")
+                # print(new_start)
+            '''
             ch.extend([chrom] * cpg_num)
             st.extend([strand] * cpg_num)
             log.extend([logRatio] * cpg_num)
             readID.extend([read_name] * cpg_num)
+            
 
 final = pd.DataFrame(list(zip(ch,pos,st,log,readID)), columns =['Chr', 'Pos','Strand', 'Log.like.ratio', 'Read_ID'])
 final.to_csv(snakemake.output[0], sep="\t", header=True, index=False)
-
+'''