Delete n (#10)

* introduced min_mut def

* updated args

* updated docs

README.md

@@ -69,6 +69,8 @@ options:
   -t 0, --threshold 0   display frequencies above this threshold (0-1)
   --delete, --no-delete
                         delete mutations that are present in all samples and their maximum frequency divergence is smaller than 0.5 (default: True)
+  -n None, --delete-n None
+                        do not show mutations that occur n times or less (default: Do not delete)                      
   --sort, --no-sort     sort sample names alphanumerically (default: False)
   --min-cov 20          display mutations covered at least x time (only if per base cov tsv files are provided)
   -v, --version         show program's version number and exit

virheat/__init__.py

@@ -1,3 +1,3 @@
 """plot vcf data as a heatmap mapped to a virus genome"""
 _program = "virheat"
-__version__ = "0.5.3"
+__version__ = "0.5.4"

virheat/command.py

@@ -72,6 +72,14 @@ def get_args(sysargs):
         default=True,
         help="delete mutations that are present in all samples and their maximum frequency divergence is smaller than 0.5"
     )
+    parser.add_argument(
+        "-n",
+        "--delete-n",
+        type=int,
+        metavar='None',
+        default=None,
+        help="do not show mutations that occur n times or less (default: Do not delete)"
+    )
     parser.add_argument(
         "--sort",
         action=argparse.BooleanOptionalAction,
@@ -114,8 +122,11 @@ def main(sysargs=sys.argv[1:]):
     # extract vcf info
     reference_name, frequency_lists, unique_mutations, file_names = data_prep.extract_vcf_data(vcf_files, threshold=args.threshold)
     frequency_array = data_prep.create_freq_array(unique_mutations, frequency_lists)
+    # user specified delete options (removes mutations based on various rationales)
     if args.delete:
         frequency_array = data_prep.delete_common_mutations(frequency_array, unique_mutations)
+    if args.delete_n is not None:
+        frequency_array = data_prep.delete_n_mutations(frequency_array, unique_mutations, args.delete_n)
     # annotate low coverage if per base coveage from qualimap was provided
     data_prep.annotate_non_covered_regions(args.input[0], args.min_cov, frequency_array, file_names, unique_mutations)
 

virheat/scripts/data_prep.py

@@ -195,6 +195,26 @@ def delete_common_mutations(frequency_array, unique_mutations):
 
     return np.delete(frequency_array, mut_to_del, axis=1)
 
+def delete_n_mutations(frequency_array, unique_mutations, min_mut):
+    """
+    delete mutations that are not present in more than n samples
+    """
+    mut_to_del = []
+
+    for idx in range(0, len(frequency_array[0])):
+        n_mutations = 0
+        for frequency_list in frequency_array:
+            if frequency_list[idx] > 0:
+                n_mutations += 1
+        # check if min_mut was reached and if not mark as to delete
+        if n_mutations <= min_mut:
+            mut_to_del.append(idx)
+    # delete the mutations that are found only min_mut times in all samples
+    for idx in sorted(mut_to_del, reverse=True):
+        del unique_mutations[idx]
+
+    return np.delete(frequency_array, mut_to_del, axis=1)
+
 
 def parse_gff3(file):
     """