diff --git a/workflow_zymo/Snakefile b/workflow_zymo/Snakefile
index ce0507e84e17feb8974f89c3cfe63d02280b0f63..abffb2a29be1c653635dc728a2c995d763cafcb3 100644
--- a/workflow_zymo/Snakefile
+++ b/workflow_zymo/Snakefile
@@ -238,12 +238,13 @@ rule comparison:
         for fname in input.cdhit:
             fname_col=os.path.basename(fname).split("__")[0] # column name to be used
             fname_pids=[]
-            with open(input.cdhit,'r') as fi:
+            with open(fname,'r') as fi:
                 for ln in fi:
                     if ln.startswith('>'):
-                        fname_pids.append(re.sub("^>", "", ln.strip().split(" ")[0]))
-            
+                        nam=re.sub("^>", "", ln.strip().split(" ")[0])
+                        fname_pids.append(nam.strip().split("__")[2])
+                
         df.loc[fname_pids,fname_col] = True
         
         # save
-        df.to_csv(output[0], sep='\t', header=0)
\ No newline at end of file
+        df.to_csv(output[0], sep='\t', header=0)