feat: adds -n, --nb-models option

This option allows to choose the number of target models generated by MODELLER

Author: thomas-bailly

asmc/modeling.py

@@ -9,12 +9,13 @@
 ## Functions ##
 ###############
 
-def modeling(aln, outdir):
+def modeling(aln, outdir, n):
     """Build model
 
     Args:
         aln (str): String containing the path to the alignment and the reference name
-        outdir (pathlib.Path): Path to the output directory 
+        outdir (pathlib.Path): Path to the output directory
+        n (int): number of target models generated by MODELLER
 
     Returns:
         (int): 0
@@ -35,7 +36,7 @@ def modeling(aln, outdir):
                                             automodel.assess.GA341))
 
     a.starting_model = 1
-    a.ending_model = 2
+    a.ending_model = n
     a.make()
 
     model_name = ""
@@ -63,7 +64,9 @@ def modeling(aln, outdir):
                         " ref name(s), e.g: /home/data/target.ali+refA+refB+refC")
     parser.add_argument("-o", "--outdir", type=str, metavar="", default="./",
                         help="output directory [default: ./]")
+    parser.add_argument("-n", "--nb-models", type=int, default=2, metavar="",
+                        help="number of target models generated by MODELLER")
 
     args = parser.parse_args()
 
-    modeling(args.ali, Path(args.outdir))
+    modeling(args.ali, Path(args.outdir), args.nb_models)

run_asmc.py

@@ -149,13 +149,14 @@ def run_build_ali(ref, seq, pocket, outdir, pid, log):
 
     return ret
 
-def modeling(job_file, outdir, threads):
+def modeling(job_file, outdir, threads, n):
     """Prepares data and execute in parallel the run_modeling function
 
     Args:
         job (pathlib.Path): Path of file containing the list of inputs
         outdir (pathlib.Path): Path to the output directory
         threads (int): Number of parallel jobs
+        n (int): number of target models generated by MODELLER
 
     Returns:
         ret (subprocess.CompletedProcess): The completed process
@@ -176,7 +177,7 @@ def modeling(job_file, outdir, threads):
     job_list = []
     with open(job_file, "r") as f:
         for line in f:
-            job_list.append(f"python3 {src_path} -a {line.strip()} -o {model_dir}")
+            job_list.append(f"python3 {src_path} -a {line.strip()} -o {model_dir} -n {n}")
 
     with multiprocessing.Pool(processes=threads) as pool:
         ret = set(pool.map(run_modeling, job_list))
@@ -387,6 +388,8 @@ def run_usalign(job, usalign, log):
                             help="percent identity cutoff between target and " 
                             "reference to build a model of the target, only " 
                             "used with -s, --seqs [default: 30.0]")
+    targts_opt.add_argument("-n", "--nb-models", type=int, default=2, metavar="",
+                            help="number of target models generated by MODELLER")
     dbscan_opt = parser.add_argument_group("Clustering options")
     dbscan_opt.add_argument("-e", "--eps", type=str, metavar="", default="auto",
                             help="maximum distance between two samples for them"
@@ -545,7 +548,7 @@ def run_usalign(job, usalign, log):
                 sys.exit(1)
             else:
                 start_model = datetime.datetime.now()
-                ret_model = modeling(job_file, outdir, args.threads)
+                ret_model = modeling(job_file, outdir, args.threads, args.nb_models)
                 logging.info("modeling duration: "
                              f"{datetime.datetime.now() - start_model}")