update msd output parsing

lib/msd.ml

@@ -4,7 +4,7 @@ open Bistro.EDSL
 open Bistro_bioinfo.Std
 open File_formats
 
-let env = docker_image ~account:"carinerey" ~name:"msd" ~tag:"07232018" ()
+let env = docker_image ~account:"carinerey" ~name:"msd" ~tag:"08082018" ()
 
 let msd ~(faa:aminoacid_fasta workflow) ~(tree_sc:_ workflow) : [`msd] directory workflow =
   let map_table = tmp // "map.tsv" in
@@ -22,7 +22,7 @@ let msd ~(faa:aminoacid_fasta workflow) ~(tree_sc:_ workflow) : [`msd] directory
     cmd "msd" ~env [
       opt "-t" int 1;
       opt "-o"  ident out ;
-      opt "-e" float 1. ;
+      opt "-e" float 0.05 ;
       dep tree_sc;
       ident map_table;
       dep faa;

lib/scripts/parse_output_msd.py

@@ -94,23 +94,10 @@ res_line = lines[1].strip()
 res_line_split = res_line.split("\t")
 
 n_sites=int(res_line_split[2])
-prob_post=["NA"] * (n_sites)
 
-try:
-    p_value = 1-float(res_line_split[5])
-except:
-    p_value = 1
 
-try:
-    convergent_sites = res_line_split[7].split(",")
-except: # if no convergent sites, convergent sites field don't exist
-    convergent_sites = []
-
-for cs in convergent_sites:
-    prob_post[int(cs) -1] = p_value
-
-Sites = [i +1 for i in range(n_sites)]
-df_final = pd.DataFrame({'Sites': Sites, '1MinusP' : prob_post})
+df_final = pd.read_csv(InputFile.name, sep="\t", header=None, skiprows=[0,1,2], names = ["Sites", "p_value"])
+df_final ["1MinusP"] = 1 - df_final["p_value"]
 
 df_final = df_final[["Sites","1MinusP"]]