switched to sequence divergence > 0.5 and updated docs

README.md

@@ -28,13 +28,8 @@ pip install virheat
 ```shell
 git clone https://github.com/jonas-fuchs/virHEAT
 cd virHEAT
-```
-and then install virHEAT with:
-```shell
 pip install -r requirements.txt
-```
-or:
-```shell
+# or
 pip install .
 ```
 That was already it. To check if it worked:
@@ -55,21 +50,20 @@ usage: 	virheat <folder containing vcfs> <output dir> -l or -g [additional argum
 
 ```
 positional arguments:
-  input                 folder containing vcf (and tsv) files and output folder
+  input                 folder containing input files and output folder
 
-optional arguments:
+options:
   -h, --help            show this help message and exit
   -l None, --genome-length None
                         length of the genome (needed if gff3 is not provided)
   -g None, --gff3-path None
                         path to gff3 (needed if length is not provided)
-  -a gene, --gff3-annotations gene
-                        annotations to display from gff3 file (standard: gene)
-  -t 0, --threshold 0   display frequencies above this threshold
+  -a [gene ...], --gff3-annotations [gene ...]
+                        annotations to display from gff3 file (standard: gene). Multiple possible.
+  -t 0, --threshold 0   display frequencies above this threshold (0-1)
   --delete, --no-delete
-                        delete mutations with frequencies present in all
-                        samples (default: True)
-  --sort, --no-sort     sort alphanumerically (default: False)
+                        delete mutations that are present in all samples and their maximum frequency divergence is smaller than 0.5 (default: True)
+  --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/command.py

@@ -64,19 +64,19 @@ def get_args(sysargs):
         type=float,
         metavar="0",
         default=0,
-        help="display frequencies above this threshold"
+        help="display frequencies above this threshold (0-1)"
     )
     parser.add_argument(
         "--delete",
         action=argparse.BooleanOptionalAction,
         default=True,
-        help="delete mutations that are present in all samples and their maximum frequency divergence is smaller than 0.2"
+        help="delete mutations that are present in all samples and their maximum frequency divergence is smaller than 0.5"
     )
     parser.add_argument(
         "--sort",
         action=argparse.BooleanOptionalAction,
         default=False,
-        help="sort alphanumerically"
+        help="sort sample names alphanumerically"
     )
     parser.add_argument(
         "--min-cov",

virheat/scripts/data_prep.py

@@ -185,9 +185,9 @@ def delete_common_mutations(frequency_array, unique_mutations):
         # check if all mutation in a column are zero (happens with some weird callers)
         if all(x == 0 for x in check_all):
             mut_to_del.append(idx)
-        # check if frequencies are present in all columns and the maximal diff is greater than 0.2
-        # example [0.8, 0.7, 0.6] is not deleted whereas [0.8, 0.7, 0.7] is deleted
-        elif all(x > 0 for x in check_all) and max(check_all)-min(check_all) >= 0.2:
+        # check if frequencies are present in all columns and the maximal diff is greater than 0.5
+        # example [0.8, 0.7, 0.3] is not deleted whereas [0.8, 0.7, 0.7] is deleted
+        elif all(x > 0 for x in check_all) and max(check_all)-min(check_all) < 0.5:
             mut_to_del.append(idx)
 
     for idx in sorted(mut_to_del, reverse=True):