further refactoring bppseqgen

lib/bppsuite.ml

@@ -89,7 +89,7 @@ rate_distribution=Constant()
 
   let conf_file_bppseqgen_multi_profiles ~tree ~profile_f ~ne_c ~ne_a ~hypothesis ~nb_sites_per_profile =
     seq ~sep:"\n" (
-      [
+      bpp_config_F hypothesis @ [
         assign "input.tree.file" (dep tree) ;
         assign "input.tree.scale" (int 3) ;
         assign "PROFILE_F" (dep profile_f) ;
@@ -97,8 +97,8 @@ rate_distribution=Constant()
         assign "NE_1"            (float ne_a) ;
         assign "NE_C"            (float ne_c) ;
         assign "NE_T"            (float ne_c) ;
+        string "" ; (* finish with carriage return *)
       ]
-      @ bpp_config_F hypothesis
     )
 
   let bppseqgen_multi_profiles_script ~config  ~out ~profile_c ~seed =
@@ -130,31 +130,27 @@ rate_distribution=Constant()
 
 |}
 
-  let multi_profiles ?(descr="") ~profile_f ~profile_c ~tree_dataset ~hypothesis ~ne_c ~ne_a ~seed =
+  let multi_profiles ?(descr="") ~profile_f ~profile_c ~tree_dataset ~hypothesis ~seed =
     let nb_sites_per_profile = 1 in
-    (* let nb_combis = Pervasives.(nb_sites / nb_sites_per_profile) in *)
     let config_f = dest // "config.bpp" in
-    (* let profile_c_ok = tmp // "profiles_c.tsv" in *)
     let profile_c_ok = dep profile_c in
     let out = dest // "seq.fa" in
     let tree = Tree_dataset.tree tree_dataset `Simulation in
     let nodes = Tree_dataset.nodes tree_dataset hypothesis in
+    let ne_g = Convergence_hypothesis.neg_of_model hypothesis in
+    let ne_c = Convergence_hypothesis.nec_of_model hypothesis in
+    let ne_a = ne_g in
     workflow ~descr:("bppsuite.bppseqgen" ^ descr) [
       docker env (
         and_list [
           mkdir_p dest;
           mkdir_p tmp;
           cd tmp;
-          (*cmd "head" ~stdout:profile_c_ok [
-            opt "-n" int nb_sites;
-            dep profile_c
-            ];*)
           cmd "cat" ~stdout:config_f [
             file_dump (conf_file_bppseqgen_multi_profiles ~tree ~profile_f ~hypothesis ~ne_c ~ne_a ~nb_sites_per_profile) ;
-            file_dump (seq ~sep:"\n" [ string "\n" ]) ;
             dep nodes ;
           ];
-          cmd "bash" [(file_dump (bppseqgen_multi_profiles_script ~config:config_f  ~out  ~profile_c:profile_c_ok ~seed))];
+          cmd "bash" [ file_dump (bppseqgen_multi_profiles_script ~config:config_f ~out ~profile_c:profile_c_ok ~seed)];
         ]
       )
     ]

lib/bppsuite.mli

@@ -9,8 +9,6 @@ module Bppseqgen : sig
     profile_c: text_file workflow ->
     tree_dataset:[`tree_dataset] directory workflow ->
     hypothesis:Convergence_hypothesis.t ->
-    ne_c: float ->
-    ne_a: float ->
     seed:int ->
     [`bppseqgen] directory workflow
 

lib/pipeline.ml

@@ -54,22 +54,13 @@ let calc_fixed_seed ~(str:string) (seed:int) : int =
 let derive_from_model ~model ~input_tree ~tree_dataset ~tree_prefix ~profile ~preview ~ns ~seed =
   let model_prefix = Convergence_hypothesis.string_of_model model in
   let descr = "."^model_prefix^"."^tree_prefix in
-  (* only 1 profile or 1 couple of profiles*)
-  (*let config = Convergence_hypothesis.bpp_config nodes model in
-    let fna = Bppsuite.bppseqgen ~descr ~nb_sites ~tree ~config in
-  *)
   (* with several profiles or couples of profiles *)
-  let ne_g = neg_of_model model in
-  let ne_c = nec_of_model model in
-  let ne_a = ne_g in
   let profile_f = profile.profile_f in
   let profile_c = profile.profile_c in
-  (*let seed = Random.int Int.max_value in*)
   let seed = calc_fixed_seed ~str:descr seed in
-  let run_fna = Bppsuite.Bppseqgen.multi_profiles ~descr ~tree_dataset ~hypothesis:model ~profile_f ~profile_c ~ne_c ~ne_a ~seed in
+  let run_fna = Bppsuite.Bppseqgen.multi_profiles ~descr ~tree_dataset ~hypothesis:model ~profile_f ~profile_c ~seed in
   let fna = Bppsuite.Bppseqgen.alignment run_fna in
   let fna_infos = Some (Bppsuite.Bppseqgen.info run_fna) in
-
   let faa = Bppsuite.fna2faa ~fna in
   let ready_dataset = { Ready_dataset.input_tree = input_tree ; tree_dataset ; fna; faa; fna_infos } in
   { Dataset.model_prefix; is_real= false; tree_prefix; dataset = ready_dataset; seed }