metapredict 2.4.2__tar.gz → 2.5__tar.gz

{metapredict-2.4.2 → metapredict-2.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: metapredict
-Version: 2.4.2
+Version: 2.5
 Summary: UNKNOWN
 Home-page: https://github.com/idptools/metapredict.git
 Author: Ryan Emenecker - Holehouse Lab - WUSM
@@ -14,9 +14,16 @@ License-File: LICENSE
 # metapredict: A machine learning-based tool for predicting protein disorder.
 
 **Important update** - as of February 15, 2022 we have updated metapredict to V2.0. This comes with important changes that improve the accuracy of metapredict. Please see the section on the update *Major update to metapredict predictions to increase overall accuracy* below. In addition, this update changes the functionality of the *predict_disorder_domains()* function, so please read the documentation on that function if you were using it previously. 
-*A write up with more info on this update will be available soon.*
 
-**metapredict** uses a bidirectional recurrent neural network trained on the consensus disorder values from 8 disorder predictors from 12 proteomes that were obtained from [MobiDB](https://mobidb.bio.unipd.it/). The creation of metapredict was made possible by [parrot](https://github.com/idptools/parrot).
+**metapredict** uses a bidirectional recurrent neural network trained on the consensus disorder values from 8 disorder predictors from 12 proteomes that were obtained from [MobiDB](https://mobidb.bio.unipd.it/). In addition, as of version 2, metapredict incorporates an additional layer of predictions by counter-selecting based on structure predictions from AlphaFold2. The creation of metapredict was made possible by [parrot](https://github.com/idptools/parrot).
+
+## How can I use metapredict?
+Metapredict can be used in four different ways:
+
+1. As a stand-alone command-line tool (installable via pip - the code in this repository).
+2. As a Python library for integrating into your favorite bioinformatics pipeline (installable via pip - the code in this repository).
+3. As a web-server for examining disorder predictions on individual sequences found at [https://metapredict.net/](https://metapredict.net/).
+4. *NEW as of August 2022:* as a Google Colab notebook for batch-predicting disorder scores for larger numbers of sequences: [**LINK HERE**](https://colab.research.google.com/drive/1UOrOxun9i23XDE8lFo_4I89Tw8P3Z1D-?usp=sharing). Performance-wise, batch mode can predict the entire yeast proteome in ~6 min.
 
 
 
@@ -62,6 +69,12 @@ https://metapredict.readthedocs.io/en/latest/getting_started.html
 For disorder predictions using our server, please see:
 https://metapredict.net
 
+
+## What's new?
+
+The most recent update to metapredict on March 15, 2023 introduced multiprocessing support to the command-line interface (you're welcome, Jeff). This is only implemented for the CLI at this time. **We were able to predict disorder in the entire human proteome in about 4 minutes with metapredict V2 and in about 2 minutes using legacy metapredict!** *We didn't even have to close the 40 chrome tabs we definitely had to have open for testing.* If there is additional interest in speeding up other functionality by using multiple cores, let us know!
+
+
 ## How to cite metapredict
 
 If you use metapredict for your work, please cite the metapredict paper - 
@@ -119,6 +132,28 @@ If you would like to specify where to save the output, simply use the ``-o`` or
 **Using the original metapredict predictor**
 To use the original metapredict predictor as opposed to our new, updated predictor, use the ``-l`` or ``--legacy`` flag!
 
+### Predicting Disorder from a fasta file using multiple cores
+
+The ``metapredict-multipredict`` command from the command line takes a .fasta file as input and returns disorder scores for the sequences in the FASTA file. It works the same as metapredict-predict-disroder but splits the sequences across the number of cores you specify. USE THIS AT YOUR OWN RISK! If you specify too many cores, you're going to have a bad time.
+
+	$ metapredict-multipredict <Path to .fasta file> <number of cores>
+
+**Example - using 5 cores**
+
+	$ metapredict-multipredict /Users/thisUser/Desktop/interestingProteins.fasta 5
+
+**Additional Usage**
+
+**specifying where to save the output -** 
+If you would like to specify where to save the output, simply use the ``-o`` or ``--output-file`` flag and then specify the file path and file name. By default this command will save the output file as disorder_scores.csv to your current working directory. However, you can specify the file name in the output path.
+
+**Example**
+
+    $ metapredict-multipredict /Users/thisUser/Desktop/interestingProteins.fasta 5 -o /Users/thisUser/Desktop/disorder_predictions/my_disorder_predictions.csv
+
+**Using the original metapredict predictor**
+To use the original metapredict predictor as opposed to our new, updated predictor, use the ``-l`` or ``--legacy`` flag!
+
 ### Predicting Disorder from a sequence
 
 ``metapredict-quick-predict`` is a command that will let you input a sequence and get disorder values immediately printed to the terminal. The only argument that can be input is the sequence.
@@ -1098,6 +1133,18 @@ Example data that can be used with metapredict can be found in the metapredict/d
 
 This section is a log of recent changes with metapredict. My hope is that as I change things, this section can help you figure out why a change was made and if it will break any of your current work flows. The first major changes were made for the 0.56 release, so tracking will start there. Reasons are not provided for bug fixes for because the reason can assumed to be fixing the bug...
 
+
+#### V2.5
+Changes:
+
+* Added the first multicore support to metapredict. Currently limited to metapredict-predict-disorder functionality.
+=======
+#### V2.4.3
+Changes:
+
+* Updated the default names for `metapredict-predict-idrs` so that the FASTA output file is now called `idrs.fasta` instead of the inappropriate `shephard_idrs.tsv`.
+* Added link to the new batch-mode Google Colab notebook!
+
 #### V2.4.2 
 Changes:
 

{metapredict-2.4.2 → metapredict-2.5}/README.md

@@ -1,9 +1,16 @@
 # metapredict: A machine learning-based tool for predicting protein disorder.
 
 **Important update** - as of February 15, 2022 we have updated metapredict to V2.0. This comes with important changes that improve the accuracy of metapredict. Please see the section on the update *Major update to metapredict predictions to increase overall accuracy* below. In addition, this update changes the functionality of the *predict_disorder_domains()* function, so please read the documentation on that function if you were using it previously. 
-*A write up with more info on this update will be available soon.*
 
-**metapredict** uses a bidirectional recurrent neural network trained on the consensus disorder values from 8 disorder predictors from 12 proteomes that were obtained from [MobiDB](https://mobidb.bio.unipd.it/). The creation of metapredict was made possible by [parrot](https://github.com/idptools/parrot).
+**metapredict** uses a bidirectional recurrent neural network trained on the consensus disorder values from 8 disorder predictors from 12 proteomes that were obtained from [MobiDB](https://mobidb.bio.unipd.it/). In addition, as of version 2, metapredict incorporates an additional layer of predictions by counter-selecting based on structure predictions from AlphaFold2. The creation of metapredict was made possible by [parrot](https://github.com/idptools/parrot).
+
+## How can I use metapredict?
+Metapredict can be used in four different ways:
+
+1. As a stand-alone command-line tool (installable via pip - the code in this repository).
+2. As a Python library for integrating into your favorite bioinformatics pipeline (installable via pip - the code in this repository).
+3. As a web-server for examining disorder predictions on individual sequences found at [https://metapredict.net/](https://metapredict.net/).
+4. *NEW as of August 2022:* as a Google Colab notebook for batch-predicting disorder scores for larger numbers of sequences: [**LINK HERE**](https://colab.research.google.com/drive/1UOrOxun9i23XDE8lFo_4I89Tw8P3Z1D-?usp=sharing). Performance-wise, batch mode can predict the entire yeast proteome in ~6 min.
 
 
 
@@ -49,6 +56,12 @@ https://metapredict.readthedocs.io/en/latest/getting_started.html
 For disorder predictions using our server, please see:
 https://metapredict.net
 
+
+## What's new?
+
+The most recent update to metapredict on March 15, 2023 introduced multiprocessing support to the command-line interface (you're welcome, Jeff). This is only implemented for the CLI at this time. **We were able to predict disorder in the entire human proteome in about 4 minutes with metapredict V2 and in about 2 minutes using legacy metapredict!** *We didn't even have to close the 40 chrome tabs we definitely had to have open for testing.* If there is additional interest in speeding up other functionality by using multiple cores, let us know!
+
+
 ## How to cite metapredict
 
 If you use metapredict for your work, please cite the metapredict paper - 
@@ -106,6 +119,28 @@ If you would like to specify where to save the output, simply use the ``-o`` or
 **Using the original metapredict predictor**
 To use the original metapredict predictor as opposed to our new, updated predictor, use the ``-l`` or ``--legacy`` flag!
 
+### Predicting Disorder from a fasta file using multiple cores
+
+The ``metapredict-multipredict`` command from the command line takes a .fasta file as input and returns disorder scores for the sequences in the FASTA file. It works the same as metapredict-predict-disroder but splits the sequences across the number of cores you specify. USE THIS AT YOUR OWN RISK! If you specify too many cores, you're going to have a bad time.
+
+	$ metapredict-multipredict <Path to .fasta file> <number of cores>
+
+**Example - using 5 cores**
+
+	$ metapredict-multipredict /Users/thisUser/Desktop/interestingProteins.fasta 5
+
+**Additional Usage**
+
+**specifying where to save the output -** 
+If you would like to specify where to save the output, simply use the ``-o`` or ``--output-file`` flag and then specify the file path and file name. By default this command will save the output file as disorder_scores.csv to your current working directory. However, you can specify the file name in the output path.
+
+**Example**
+
+    $ metapredict-multipredict /Users/thisUser/Desktop/interestingProteins.fasta 5 -o /Users/thisUser/Desktop/disorder_predictions/my_disorder_predictions.csv
+
+**Using the original metapredict predictor**
+To use the original metapredict predictor as opposed to our new, updated predictor, use the ``-l`` or ``--legacy`` flag!
+
 ### Predicting Disorder from a sequence
 
 ``metapredict-quick-predict`` is a command that will let you input a sequence and get disorder values immediately printed to the terminal. The only argument that can be input is the sequence.
@@ -1085,6 +1120,18 @@ Example data that can be used with metapredict can be found in the metapredict/d
 
 This section is a log of recent changes with metapredict. My hope is that as I change things, this section can help you figure out why a change was made and if it will break any of your current work flows. The first major changes were made for the 0.56 release, so tracking will start there. Reasons are not provided for bug fixes for because the reason can assumed to be fixing the bug...
 
+
+#### V2.5
+Changes:
+
+* Added the first multicore support to metapredict. Currently limited to metapredict-predict-disorder functionality.
+=======
+#### V2.4.3
+Changes:
+
+* Updated the default names for `metapredict-predict-idrs` so that the FASTA output file is now called `idrs.fasta` instead of the inappropriate `shephard_idrs.tsv`.
+* Added link to the new batch-mode Google Colab notebook!
+
 #### V2.4.2 
 Changes:
 

{metapredict-2.4.2 → metapredict-2.5}/metapredict/_version.py

@@ -8,11 +8,11 @@ import json
 
 version_json = '''
 {
- "date": "2022-08-02T10:18:02-0500",
+ "date": "2023-03-15T15:58:20-0500",
  "dirty": false,
  "error": null,
- "full-revisionid": "a85bd401c15347961d81941b6ee0424b43aca8b4",
- "version": "v2.4.2"
+ "full-revisionid": "2dd04449f97b17e4b7a0e46558e0922134e517f1",
+ "version": "v2.5"
 }
 '''  # END VERSION_JSON
 

{metapredict-2.4.2 → metapredict-2.5}/metapredict/backend/meta_tools.py

@@ -1,4 +1,4 @@
-
+import protfasta
 import re
 from metapredict.metapredict_exceptions import MetapredictError
 
@@ -230,8 +230,78 @@ def write_caid_format(input_dict, output_file):
     current_output.close()
 
 
+def split_fasta(protfasta_fasta_list, number_splits):
+    '''
+    function to split the dict returned from
+    a fasta file read in by protfasta to make
+    a specific number of dictionaries with
+    approximately equal numbers of proteins. 
+    '''
+    all_nums=[num for num in range(0, len(protfasta_fasta_list))]
+    num_per_split=int(len(protfasta_fasta_list)/number_splits)
+    all_vals=[]
+    for val in range(0, number_splits):
+        temp_list=[]
+        for subval in range(0, num_per_split):
+            temp_list.append(all_nums.pop())
+        if val==number_splits-1:
+            for val in all_nums:
+                temp_list.append(val)
+        all_vals.append(temp_list)
+    # keep track of added prots
+    num_prots=0
+    split_lists=[]
+    for val in all_vals:
+        temp_list=[]
+        for prot in val:
+            num_prots+=1
+            temp_list.append([protfasta_fasta_list[prot][0], protfasta_fasta_list[prot][1]])
+        split_lists.append(temp_list)
+    # make sure all proteins added
+    if num_prots!= len(protfasta_fasta_list):
+        raise Exception('splitting of fasta file did not get all proteins')
+    return split_lists
+
+
+def append_to_file(outpath, idrs, mode):
+    # append to the file we are writing to.
+    fh = open(outpath, 'a')
+    # depending on mode...
+    # dict to hold vals...
+    return_dictionary = {}
+    if mode == 'fasta':
+        # for each sequence
+        for s in idrs:
+            # d is IDR start and end positions
+            for d in idrs[s][2]:
+                return_dictionary[f'{s:s} IDR_START={d[0]:d} IDR_END={d[1]:d}'] =  d[2]
+        # write out fasta.
+        protfasta.write_fasta(return_dictionary, outpath)
+
+    elif mode == 'shephard-domains':
+        fh = open(outpath, 'w')
+        for s in idrs:
+            # d is IDR start and end positions
+            for d in idrs[s][2]:
+                # note need +1 for shephard format
+                start = d[0]+1
+                end = d[1]
+                fh.write(f'{s}\t{start}\t{end}\tIDR\n')
+
+    elif mode == 'shephard-domains-uniprot':
+        fh = open(outpath, 'w')
+        for s in idrs:
+            # d is IDR start and end positions
+            for d in idrs[s][2]:
+
+                uid = s.split('|')[1]
+                start = d[0]+1
+                end = d[1]
+                fh.write(f'{uid}\t{start}\t{end}\tIDR\n')
 
-
+    else:
+        raise Exception('no mode specified!')
+                
 
 # ..........................................................................................
 #

{metapredict-2.4.2 → metapredict-2.5}/metapredict/meta.py

@@ -8,7 +8,7 @@
 ##Handles the primary functions
 
 # NOTE - any new functions must be added to this list!
-__all__ =  ['predict_disorder_domains', 'predict_disorder', 'graph_disorder', 'predict_all', 'percent_disorder', 'predict_disorder_fasta', 'graph_disorder_fasta', 'predict_disorder_uniprot', 'graph_disorder_uniprot', 'predict_disorder_domains_uniprot', 'predict_disorder_domains_from_external_scores', 'graph_pLDDT_uniprot', 'predict_pLDDT_uniprot', 'graph_pLDDT_fasta', 'predict_pLDDT_fasta', 'graph_pLDDT', 'predict_pLDDT', 'predict_disorder_caid']
+__all__ =  ['predict_disorder_domains', 'predict_disorder', 'graph_disorder', 'predict_all', 'percent_disorder', 'predict_disorder_fasta', 'graph_disorder_fasta', 'predict_disorder_uniprot', 'graph_disorder_uniprot', 'predict_disorder_domains_uniprot', 'predict_disorder_domains_from_external_scores', 'graph_pLDDT_uniprot', 'predict_pLDDT_uniprot', 'graph_pLDDT_fasta', 'predict_pLDDT_fasta', 'graph_pLDDT', 'predict_pLDDT', 'predict_disorder_caid', 'predict_for_multithread']
  
 import os
 import sys
@@ -1612,3 +1612,18 @@ def predict_disorder_caid(input_fasta, output_file):
     _meta_tools.write_caid_format(output_dict, output_file)
 
 
+def predict_for_multithread(nested_list_of_seqs, 
+    outpath, mode, threshold_val, legacy):
+    '''
+    not multithreaded itself. Just a function
+    needed to predicted batches of sequences
+    for ultimately using in metapredict-multipredict
+    '''
+    all_vals={}
+    for name_seq in nested_list_of_seqs:
+        all_vals[name_seq[0]] = predict_disorder_domains(name_seq[1], disorder_threshold=threshold_val, legacy=legacy, return_list=True)
+    
+    # append vals to outpath
+    _meta_tools.append_to_file(outpath, all_vals, mode=mode)
+
+

{metapredict-2.4.2 → metapredict-2.5}/metapredict.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: metapredict
-Version: 2.4.2
+Version: 2.5
 Summary: UNKNOWN
 Home-page: https://github.com/idptools/metapredict.git
 Author: Ryan Emenecker - Holehouse Lab - WUSM
@@ -14,9 +14,16 @@ License-File: LICENSE
 # metapredict: A machine learning-based tool for predicting protein disorder.
 
 **Important update** - as of February 15, 2022 we have updated metapredict to V2.0. This comes with important changes that improve the accuracy of metapredict. Please see the section on the update *Major update to metapredict predictions to increase overall accuracy* below. In addition, this update changes the functionality of the *predict_disorder_domains()* function, so please read the documentation on that function if you were using it previously. 
-*A write up with more info on this update will be available soon.*
 
-**metapredict** uses a bidirectional recurrent neural network trained on the consensus disorder values from 8 disorder predictors from 12 proteomes that were obtained from [MobiDB](https://mobidb.bio.unipd.it/). The creation of metapredict was made possible by [parrot](https://github.com/idptools/parrot).
+**metapredict** uses a bidirectional recurrent neural network trained on the consensus disorder values from 8 disorder predictors from 12 proteomes that were obtained from [MobiDB](https://mobidb.bio.unipd.it/). In addition, as of version 2, metapredict incorporates an additional layer of predictions by counter-selecting based on structure predictions from AlphaFold2. The creation of metapredict was made possible by [parrot](https://github.com/idptools/parrot).
+
+## How can I use metapredict?
+Metapredict can be used in four different ways:
+
+1. As a stand-alone command-line tool (installable via pip - the code in this repository).
+2. As a Python library for integrating into your favorite bioinformatics pipeline (installable via pip - the code in this repository).
+3. As a web-server for examining disorder predictions on individual sequences found at [https://metapredict.net/](https://metapredict.net/).
+4. *NEW as of August 2022:* as a Google Colab notebook for batch-predicting disorder scores for larger numbers of sequences: [**LINK HERE**](https://colab.research.google.com/drive/1UOrOxun9i23XDE8lFo_4I89Tw8P3Z1D-?usp=sharing). Performance-wise, batch mode can predict the entire yeast proteome in ~6 min.
 
 
 
@@ -62,6 +69,12 @@ https://metapredict.readthedocs.io/en/latest/getting_started.html
 For disorder predictions using our server, please see:
 https://metapredict.net
 
+
+## What's new?
+
+The most recent update to metapredict on March 15, 2023 introduced multiprocessing support to the command-line interface (you're welcome, Jeff). This is only implemented for the CLI at this time. **We were able to predict disorder in the entire human proteome in about 4 minutes with metapredict V2 and in about 2 minutes using legacy metapredict!** *We didn't even have to close the 40 chrome tabs we definitely had to have open for testing.* If there is additional interest in speeding up other functionality by using multiple cores, let us know!
+
+
 ## How to cite metapredict
 
 If you use metapredict for your work, please cite the metapredict paper - 
@@ -119,6 +132,28 @@ If you would like to specify where to save the output, simply use the ``-o`` or
 **Using the original metapredict predictor**
 To use the original metapredict predictor as opposed to our new, updated predictor, use the ``-l`` or ``--legacy`` flag!
 
+### Predicting Disorder from a fasta file using multiple cores
+
+The ``metapredict-multipredict`` command from the command line takes a .fasta file as input and returns disorder scores for the sequences in the FASTA file. It works the same as metapredict-predict-disroder but splits the sequences across the number of cores you specify. USE THIS AT YOUR OWN RISK! If you specify too many cores, you're going to have a bad time.
+
+	$ metapredict-multipredict <Path to .fasta file> <number of cores>
+
+**Example - using 5 cores**
+
+	$ metapredict-multipredict /Users/thisUser/Desktop/interestingProteins.fasta 5
+
+**Additional Usage**
+
+**specifying where to save the output -** 
+If you would like to specify where to save the output, simply use the ``-o`` or ``--output-file`` flag and then specify the file path and file name. By default this command will save the output file as disorder_scores.csv to your current working directory. However, you can specify the file name in the output path.
+
+**Example**
+
+    $ metapredict-multipredict /Users/thisUser/Desktop/interestingProteins.fasta 5 -o /Users/thisUser/Desktop/disorder_predictions/my_disorder_predictions.csv
+
+**Using the original metapredict predictor**
+To use the original metapredict predictor as opposed to our new, updated predictor, use the ``-l`` or ``--legacy`` flag!
+
 ### Predicting Disorder from a sequence
 
 ``metapredict-quick-predict`` is a command that will let you input a sequence and get disorder values immediately printed to the terminal. The only argument that can be input is the sequence.
@@ -1098,6 +1133,18 @@ Example data that can be used with metapredict can be found in the metapredict/d
 
 This section is a log of recent changes with metapredict. My hope is that as I change things, this section can help you figure out why a change was made and if it will break any of your current work flows. The first major changes were made for the 0.56 release, so tracking will start there. Reasons are not provided for bug fixes for because the reason can assumed to be fixing the bug...
 
+
+#### V2.5
+Changes:
+
+* Added the first multicore support to metapredict. Currently limited to metapredict-predict-disorder functionality.
+=======
+#### V2.4.3
+Changes:
+
+* Updated the default names for `metapredict-predict-idrs` so that the FASTA output file is now called `idrs.fasta` instead of the inappropriate `shephard_idrs.tsv`.
+* Added link to the new batch-mode Google Colab notebook!
+
 #### V2.4.2 
 Changes:
 

{metapredict-2.4.2 → metapredict-2.5}/metapredict.egg-info/SOURCES.txt

@@ -79,6 +79,7 @@ metapredict/tests/input_data/three_seqs.fasta
 metapredict/tests/output/test.csv
 scripts/metapredict-graph-disorder
 scripts/metapredict-graph-pLDDT
+scripts/metapredict-multipredict
 scripts/metapredict-name
 scripts/metapredict-predict-disorder
 scripts/metapredict-predict-idrs

metapredict-2.5/scripts/metapredict-multipredict

@@ -0,0 +1,108 @@
+#!/usr/bin/env python
+
+# executing script for multithreaded prediction from command line
+
+'''
+Specifically for the 'predict_disorder_domains'
+functionality.
+
+Jeff wanted disorder predictions to be faster.
+I made this for Jeff.
+You're welcome, Jeff.
+'''
+
+
+# import stuff for making CLI
+import os
+import argparse
+import protfasta
+import multiprocessing
+
+# for the predictions...
+from metapredict.backend.meta_tools import split_fasta
+import metapredict as meta
+
+if __name__ == "__main__":
+
+    # Parse command line arguments.
+    parser = argparse.ArgumentParser(description='Predict IDRs for all sequences in a FASTA file. Uses multiple cores for speeeeeeed!')
+
+    parser.add_argument('data_file', help='Path to fasta file containing sequences to be predicted.')
+
+    parser.add_argument('number_cores', help='Number of cores to run simultaneously. Do not specify more than you have..', type=int)
+
+    parser.add_argument('-o', '--output-file', help='Filename for where to save the outputfile. Default = idrs_shephard.tsv', default='idrs_shephard.tsv')
+
+    parser.add_argument('-l', '--legacy', action='store_true', help='Optional. Use this flag to use the original legacy version of metapredict.')
+
+    parser.add_argument('--invalid-sequence-action', help="For parsing FASTA file, defines how to deal with non-standard amino acids. See https://protfasta.readthedocs.io/en/latest/read_fasta.html for details. Default='convert' ", default='convert')
+
+    parser.add_argument('--mode', help='Defines the mode in which IDRs are reported. By default this generates a FASTA file with header format that matches  the input file with an additional set of fields that are "IDR_START=$START   IDR_END=$END" where $START and $END are the starting and ending IDRs. If mode is set to shephard-domains than a SHEPHAD-compliant domains file is generated. If shephard-domains-uniprot the uniprot ID is extracted from the header assuming standard uniprot formatting. Default = fasta', default='fasta')
+
+    parser.add_argument('--threshold', help='Defines the threshold used to define a region as disordered or not. Default=0.42 which is recommended.', default=0.42, type=float)
+    parser.add_argument('--verbose', help='If included then prints out status updates', action='store_true')
+
+    args = parser.parse_args()
+
+    if args.mode not in ['fasta', 'shephard-domains','shephard-domains-uniprot', ]:
+        raise Exception("--mode must be set to one of 'fasta', 'shephard-domains', or 'shephard-domains-uniprot'")
+    
+    if args.legacy:
+        use_legacy = True
+        threshold_val = args.threshold_val
+
+    else:
+        use_legacy=False
+        # if not using legacy and the default legacy value is still being used, adjust it to 0.5.
+        if args.threshold == 0.42:
+            threshold_val = 0.5
+
+        # if the user sets their own threshold value that isn't 0.42, keep it.
+        else:
+            threshold_val = args.threshold
+    
+    if not os.path.isfile(args.data_file):
+        print(f'Error: Could not find passed fasta file [{args.data_file:s}]')
+
+    # read in sequences
+    sequences = protfasta.read_fasta(args.data_file, invalid_sequence_action=args.invalid_sequence_action, return_list=True)
+    if args.verbose:
+        print('Read in FASTA file')
+
+    # make number cores an int
+    num_cores_used = args.number_cores
+
+    # split sequences into sublists for multithreading
+    split_seqs = split_fasta(sequences, num_cores_used)
+
+    # create the file
+    with open(args.output_file, 'w') as fh:
+        pass
+
+    # first split fasta into list of lists.
+    protfasta_seqs = protfasta.read_fasta(args.data_file, 
+        invalid_sequence_action=args.invalid_sequence_action, return_list=True)
+    
+    # split into list of lists where each list contains lists 
+    # where that list 0=header, 1=sequence
+    split_fasta_seqs = split_fasta(protfasta_seqs, number_splits=num_cores_used)
+
+    # iterate through batches
+    cur_threads = []
+    for sub_batch in range(0, num_cores_used):
+        cur_seq_batch = split_fasta_seqs[sub_batch]
+        #startup the thread
+        t = multiprocessing.Process(target = meta.predict_for_multithread, args = (cur_seq_batch, args.output_file, args.mode, threshold_val, use_legacy))
+        cur_threads.append(t)
+        
+    # run cur threds
+    for thr in cur_threads:
+        thr.start()
+
+    # join threads
+    for thr in cur_threads:
+        thr.join()    
+    
+                
+        
+

{metapredict-2.4.2 → metapredict-2.5}/scripts/metapredict-predict-idrs

@@ -18,13 +18,13 @@ if __name__ == "__main__":
 
     parser.add_argument('data_file', help='Path to fasta file containing sequences to be predicted.')
 
-    parser.add_argument('-o', '--output-file', help='Filename for where to save the outputfile. Default = idrs_shephard.tsv', default='idrs_shephard.tsv')
+    parser.add_argument('-o', '--output-file', help='Filename for where to save the outputfile. Defaults = idrs.fasta (if mode=fasta) and shephard_idrs.tsv otherwise')
 
     parser.add_argument('-l', '--legacy', action='store_true', help='Optional. Use this flag to use the original legacy version of metapredict.')
 
     parser.add_argument('--invalid-sequence-action', help="For parsing FASTA file, defines how to deal with non-standard amino acids. See https://protfasta.readthedocs.io/en/latest/read_fasta.html for details. Default='convert' ", default='convert')
 
-    parser.add_argument('--mode', help='Defines the mode in which IDRs are reported. By default this generates a FASTA file with header format that matches  the input file with an additional set of fields that are "IDR_START=$START   IDR_END=$END" where $START and $END are the starting and ending IDRs. If mode is set to shephard-domains than a SHEPHAD-compliant domains file is generated. If shephard-domains-uniprot the uniprot ID is extracted from the header assuming standard uniprot formatting. Default = fasta', default='fasta')
+    parser.add_argument('--mode', help='Defines the mode in which IDRs are reported. By default this generates a FASTA file with header format that matches the input file with an additional set of fields that are "IDR_START=$START  IDR_END=$END" where $START and $END are the starting and ending IDRs. If mode is set to shephard-domains than a SHEPHAD-compliant domains file is generated. If shephard-domains-uniprot the uniprot ID is extracted from the header assuming standard uniprot formatting. Default = fasta', default='fasta')
 
     parser.add_argument('--threshold', help='Defines the threshold used to define a region as disordered or not. Default=0.42 which is recommended.', default=0.42, type=float)
     parser.add_argument('--verbose', help='If included then prints out status updates', action='store_true')
@@ -33,6 +33,15 @@ if __name__ == "__main__":
 
     if args.mode not in ['fasta', 'shephard-domains','shephard-domains-uniprot', ]:
         raise Exception("--mode must be set to one of 'fasta' or 'shephard-domains'")
+
+    if args.output_file is None:
+        if args.mode == 'fasta':
+            outfile_name = 'idrs.fasta'
+        else:
+            outfile_name = 'shephard_idrs.tsv'
+
+    else:
+        outfile_name = args.output_file
     
     if args.legacy:
         use_legacy = True
@@ -80,10 +89,10 @@ if __name__ == "__main__":
                 return_dictionary[f'{s:s} IDR_START={d[0]:d} IDR_END={d[1]:d}'] =  d[2]
                 
         
-        protfasta.write_fasta(return_dictionary, args.output_file)
+        protfasta.write_fasta(return_dictionary, outfile_name)
 
     elif args.mode == 'shephard-domains':
-        fh = open(args.output_file, 'w')
+        fh = open(outfile_name, 'w')
         for s in idrs:
             # d is IDR start and end positions
             for d in idrs[s][2]:
@@ -94,7 +103,7 @@ if __name__ == "__main__":
                 fh.write(f'{s}\t{start}\t{end}\tIDR\n')
 
     elif args.mode == 'shephard-domains-uniprot':
-        fh = open(args.output_file, 'w')
+        fh = open(outfile_name, 'w')
         for s in idrs:
             # d is IDR start and end positions
             for d in idrs[s][2]:

{metapredict-2.4.2 → metapredict-2.5}/setup.py

@@ -52,7 +52,8 @@ setup(
              'scripts/metapredict-predict-idrs',
              'scripts/metapredict-graph-pLDDT',
              'scripts/metapredict-predict-pLDDT',
-             'scripts/metapredict-name'],             
+             'scripts/metapredict-name',
+             'scripts/metapredict-multipredict'],             
          
 
     # Additional entries you may want simply uncomment the lines you want and fill in the data