Add infer_singletons to workflow

ppanggolin/utils.py

@@ -441,6 +441,9 @@ def check_option_workflow(args):
     if not any([args.fasta, args.anno]):
         raise Exception("At least one of --fasta or --anno must be given")
 
+    if args.infer_singletons and args.clusters is None:
+        logging.getLogger("PPanGGOLiN").warning("--infer_singleton works only with --clusters given.")
+
 
 def parse_config_file(yaml_config_file: str) -> dict:
     """

ppanggolin/workflow/all.py

@@ -54,7 +54,8 @@ def launch_workflow(args: argparse.Namespace, panrgp: bool = True,
 
         start_anno = time.time()
         read_annotations(pangenome, args.anno, pseudo=args.annotate.use_pseudo,
-                         cpu=args.annotate.cpu, translation_table=args.annotate.translation_table, disable_bar=args.disable_prog_bar)
+                         cpu=args.annotate.cpu, translation_table=args.annotate.translation_table,
+                         disable_bar=args.disable_prog_bar)
         anno_time = time.time() - start_anno
 
         start_writing = time.time()
@@ -63,6 +64,7 @@ def launch_workflow(args: argparse.Namespace, panrgp: bool = True,
 
         if args.clusters is not None:
             start_clust = time.time()
+            print(args.cluster)
             read_clustering(pangenome, args.clusters, disable_bar=args.disable_prog_bar,
                             infer_singleton=args.cluster.infer_singletons)
         else:  # args.cluster is None
@@ -362,6 +364,10 @@ def add_workflow_args(parser: argparse.ArgumentParser):
     optional.add_argument("--identity", required=False, type=restricted_float, default=0.8,
                           help="Minimal identity percent for two proteins to be in the same cluster")
 
+    optional.add_argument("--infer_singletons", required=False, action="store_true",
+                          help="When reading a clustering result with --clusters,"
+                               " if a gene is not in the provided file")
+
     optional.add_argument("-K", "--nb_of_partitions", required=False, default=-1, type=int,
                           help="Number of partitions to use. Must be at least 2. If under 2, "
                                "it will be detected automatically.")