ENH Rename "whole" -> "global"

This matches the manuscript

Author: luispedro

ChangeLog

@@ -4,7 +4,8 @@ Unreleased
 	mmseqs format
 	* Fix bug with non-standard characters in sample names (#68)
 	* Provide pretrained models from soil, cat gut, human oral,
-	pig gut, mouse gut, built environment, wastewater and whole(training from all samples).
+	pig gut, mouse gut, built environment, wastewater and global (training
+	from all environments).
 
 Version 0.5.0 Fri Jan 7 2022 by BigDataBiology
 	* Faster `SemiBin --version`

README.md

@@ -107,7 +107,9 @@ In this example, SemiBin will download GTDB to
 `$HOME/.cache/SemiBin/mmseqs2-GTDB/GTDB`. You can change this default using the
 `-r` argument.
 
-You can use `--environment` with (`human_gut`, `dog_gut`, `ocean`, `soil`,  `cat_gut`, `human_oral`, `mouse_gut`, `pig_gut`, `built_environment`, `wastewater` and `whole`) to use one of our built-in models. (_NOTE_: This is a recommended way, which will save much time for contig annotations and model training, and also get very good results.) 
+You can use `--environment` with `human_gut`, `dog_gut`, `ocean`, `soil`,
+`cat_gut`, `human_oral`, `mouse_gut`, `pig_gut`, `built_environment`,
+`wastewater`, or `global` to use one of our built-in models.
 
 ```bash
 SemiBin single_easy_bin -i contig_S1.fna -b S1.bam -o output --environment human_gut
@@ -183,11 +185,12 @@ includes the following steps: `generate_cannot_links`, `generate_data_multi` and
 
 In advanced mode, you can also use our built-in pre-trained model in
 single-sample binning mode. Here we provide pre-trained models for human gut,
-dog gut, marine, soil, cat gut, human oral, mouse gut, pig gut, built_environment, wastewater and whole (train from whole environments) environment. You can just use these models for single-sample
-binning and it will save much time for contig annotations and model training.
+dog gut, marine, soil, cat gut, human oral, mouse gut, pig gut,
+built_environment, wastewater and global (train from all environments listed)
+environment.
 
 A very easy way to run SemiBin with a built-in model
-(`human_gut`/`dog_gut`/`ocean`/`soil`/`cat_gut`/`human_oral`/`mouse_gut`/`pig_gut`/`built_environment`/`wastewater`/`whole` for single-sample binning):
+(`human_gut`/`dog_gut`/`ocean`/`soil`/`cat_gut`/`human_oral`/`mouse_gut`/`pig_gut`/`built_environment`/`wastewater`/`global` for single-sample binning):
 
 ```bash
 SemiBin single_easy_bin -i contig_S1.fna -b S1.bam -o output --environment human_gut

SemiBin/whole_model.h5 renamed to SemiBin/global_model.h5

@@ -175,7 +175,7 @@ def parse_args(args):
     for p in [single_easy_bin, binning]:
         p.add_argument('--environment',
                        required=False,
-                       help='Environment for the built-in model (available choices: human_gut/dog_gut/ocean/soil/cat_gut/human_oral/mouse_gut/pig_gut/built_environment/wastewater/whole).',
+                       help='Environment for the built-in model (available choices: human_gut/dog_gut/ocean/soil/cat_gut/human_oral/mouse_gut/pig_gut/built_environment/wastewater/global).',
                        dest='environment',
                        default=None,
                        )

SemiBin/main.py

@@ -368,39 +368,20 @@ def split_data(data, sample, separator, is_combined = True):
     return part_data
 
 def get_model_path(env):
-    if env == 'human_gut':
-        model_path = os.path.join(os.path.split(__file__)[0], 'human_gut_model.h5')
-        return model_path
-    elif env == 'dog_gut':
-        model_path = os.path.join(os.path.split(__file__)[0], 'dog_gut_model.h5')
-        return model_path
-    elif env == 'ocean':
-        model_path = os.path.join(os.path.split(__file__)[0], 'ocean_model.h5')
-        return model_path
-    elif env == 'soil':
-        model_path = os.path.join(os.path.split(__file__)[0], 'soil_model.h5')
-        return model_path
-    elif env == 'cat_gut':
-        model_path = os.path.join(os.path.split(__file__)[0], 'cat_gut_model.h5')
-        return model_path
-    elif env == 'human_oral':
-        model_path = os.path.join(os.path.split(__file__)[0], 'human_oral_model.h5')
-        return model_path
-    elif env == 'mouse_gut':
-        model_path = os.path.join(os.path.split(__file__)[0], 'mouse_gut_model.h5')
-        return model_path
-    elif env == 'pig_gut':
-        model_path = os.path.join(os.path.split(__file__)[0], 'pig_gut_model.h5')
-        return model_path
-    elif env == 'built_environment':
-        model_path = os.path.join(os.path.split(__file__)[0], 'built_environment_model.h5')
-        return model_path
-    elif env == 'wastewater':
-        model_path = os.path.join(os.path.split(__file__)[0], 'wastewater_model.h5')
-        return model_path
-    elif env == 'whole':
-        model_path = os.path.join(os.path.split(__file__)[0], 'whole_model.h5')
-        return model_path
+    if env in [
+            'human_gut',
+            'dog_gut',
+            'ocean',
+            'soil',
+            'cat_gut',
+            'human_oral',
+            'mouse_gut',
+            'pig_gut',
+            'built_environment',
+            'wastewater',
+            'global',
+            ]:
+        return os.path.join(os.path.split(__file__)[0], f'{env}_model.h5')
     else:
         sys.stderr.write(
             f"Error: Expected environment '{env}' does not exist\n")

SemiBin/utils.py

@@ -43,15 +43,15 @@ Reconstruct bins with single or co-assembly binning using one line command.
 * `-o/--output`: Output directory (will be created if non-existent).
 * `--cannot-name:` Name for the cannot-link file (Default: cannot).
 * `-r/--reference-db-data-dir`: GTDB reference directory (Default: $HOME/.cache/SemiBin/mmseqs2-GTDB). SemiBin will lazily download GTDB if it is not found there.
-* `-p/--processes/-t/--threads`: Number of CPUs used(0: use whole).
+* `-p/--processes/-t/--threads`: Number of CPUs used (0: use all).
 * `--minfasta-kbs`: minimum bin size in kilo-basepairs (Default: 200).
 * `--recluster` : [Deprecated] Does nothing (current default is to perform clustering).
 * `--epoches`: Number of epoches used in the training process(Default: 20).
 * `--batch-size`: Batch size used in the training process(Default: 2048).
 * `--max-node`: Percentage of contigs that considered to be binned(Default: 1).
 * `--max-edges`: The maximum number of edges that can be connected to one contig(Default: 200).
 * `--random-seed`: Random seed to reproduce results.
-* `--environment`: Environment for the built-in model(human_gut/dog_gut/ocean/soil/cat_gut/human_oral/mouse_gut/pig_gut/built_environment/wastewater/whole).
+* `--environment`: Environment for the built-in model (`human_gut`/`dog_gut`/`ocean`/`soil`/`cat_gut`/`human_oral`/`mouse_gut`/`pig_gut`/`built_environment`/`wastewater`/`global`).
 * `--ratio` : If the ratio of the number of base pairs of contigs between 1000-2500 bp smaller than this value, the minimal length will be set as 1000bp, otherwise2500bp. If you set -m parameter, you do not need to use this parameter. If you use SemiBin with multi steps and you use this parameter, please use this parameter consistently with all subcommands(Default: 0.05). 
 * `-m/--min-len` : Minimal length for contigs in binning. If you use SemiBin with multi steps and you use this parameter, please use this parameter consistently with all subcommands.(Default: SemiBin chooses 1000bp or 2500bp according the ratio of the number of base pairs of contigs between 1000-2500 bp).
 * `--ml-threshold` : Length threshold for generating must-link constraints.(By default, the threshold is calculated from the contig, and the default minimum value is 4,000 bp)

docs/index.md

@@ -9,7 +9,7 @@ Input: S1.fna and S1.bam
 ```bash
 SemiBin single_easy_bin -i S1.fna -b S1.bam -o output 
 ```
-Or with our built-in model(`human_gut`/`dog_gut`/`ocean`/`soil`/`cat_gut`/`human_oral`/`mouse_gut`/`pig_gut`/`built_environment`/`wastewater`/`whole`)
+Or with our built-in model(`human_gut`/`dog_gut`/`ocean`/`soil`/`cat_gut`/`human_oral`/`mouse_gut`/`pig_gut`/`built_environment`/`wastewater`/`global`)
 ```bash
 SemiBin single_easy_bin -i S1.fna -b S1.bam -o output --environment human_gut
 ```
@@ -31,7 +31,7 @@ SemiBin train -i S1.fna --data S1_output/train.csv --data-split S1_output/train_
 ```bash
 SemiBin bin -i S1.fna --model S1_output/model.h5 --data S1_output/data.csv -o output
 ```
-or with our built-in model(`human_gut`/`dog_gut`/`ocean`/`soil`/`cat_gut`/`human_oral`/`mouse_gut`/`pig_gut`/`built_environment`/`wastewater`/`whole`)
+or with our built-in model(`human_gut`/`dog_gut`/`ocean`/`soil`/`cat_gut`/`human_oral`/`mouse_gut`/`pig_gut`/`built_environment`/`wastewater`/`global`)
 ```bash
 SemiBin bin -i S1.fna --data S1_output/data.csv -o output --environment human_gut
 ```

docs/usage.md

@@ -8,14 +8,13 @@
 - The subcommand to generate cannot links is now called
   `generate_cannot_links`. The old name (`predict_taxonomy`) is kept as a
   deprecated alias.
-  
+- Provide pretrained models from soil, cat gut, human oral,pig gut, mouse gut,
+  built environment, wastewater and global (training from all samples).
+- Add `check_install` command and run `check_install` before easy command
+
 ### Bugfixes
 - Fix bug with non-standard characters in sample names (#68).
 
-### Internal improvements
-- Add `check_install` command and run `check_install` before easy command
-- Provide pretrained models from soil, cat gut, human oral,pig gut, mouse gut, built environment, wastewater and whole(training from all samples).
-
 ## Version 0.5
 
 *Released January 7 2022*