b2bTools 3.0.7b1__tar.gz → 3.0.7b2__tar.gz

{b2bTools-3.0.7b1 → b2bTools-3.0.7b2}/MANIFEST.in

@@ -1,6 +1,8 @@
 prune */__pycache__
 recursive-exclude */__pycache__ *
 
+include requirements*.txt
+
 include b2bTools/*
 include b2bTools/singleSeq/*
 include b2bTools/singleSeq/DisoMine/*
@@ -30,9 +32,8 @@ include b2bTools/nmr/shiftCrypt/models/new_commons.mtorch
 include b2bTools/nmr/shiftCrypt/models/new_full.mtorch
 include b2bTools/nmr/shiftCrypt/models/new_NH.mtorch
 include README.md
+include setup.py
 
-exclude b2bTools/requirements.txt
-exclude b2bTools/requirements-dev.txt
 exclude b2bTools/.gitignore
 exclude b2bTools/.DS_Store
 exclude b2bTools/test/*

{b2bTools-3.0.7b1 → b2bTools-3.0.7b2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: b2bTools
-Version: 3.0.7b1
+Version: 3.0.7b2
 Summary: bio2Byte software suite to predict protein biophysical properties from their amino-acid sequences
 Home-page: https://bio2byte.be
 Author: Wim Vranken
@@ -17,6 +17,7 @@ Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: License :: OSI Approved :: GNU General Public License v3 (GPLv3)
 Classifier: Operating System :: MacOS
 Classifier: Operating System :: POSIX :: Linux
@@ -26,20 +27,76 @@ Classifier: Topic :: Scientific/Engineering :: Physics
 Classifier: Intended Audience :: Science/Research
 Classifier: Intended Audience :: Education
 Classifier: Development Status :: 5 - Production/Stable
-Requires-Python: >=3.7, <=3.12
+Requires-Python: >=3.7, <3.13
 Description-Content-Type: text/markdown
-Requires-Dist: apricot-select>=0.6.1
-Requires-Dist: biopython~=1.81
-Requires-Dist: matplotlib~=3.5.3
-Requires-Dist: numpy~=1.24
-Requires-Dist: pandas~=1.1
-Requires-Dist: pomegranate~=0.14
-Requires-Dist: requests~=2.0
-Requires-Dist: scikit-learn~=1.0.2
-Requires-Dist: scipy>=1.6.2
-Requires-Dist: torch==1.13.1
-Requires-Dist: torchvision==0.14.1
-Requires-Dist: urllib3~=1.26.6
+Requires-Dist: biopython~=1.81; python_version == "3.7"
+Requires-Dist: Cython~=0.29.37; python_version == "3.7"
+Requires-Dist: matplotlib~=3.5.3; python_version == "3.7"
+Requires-Dist: numpy~=1.21; python_version == "3.7"
+Requires-Dist: pandas~=1.1; python_version == "3.7"
+Requires-Dist: pomegranate~=0.14; python_version == "3.7"
+Requires-Dist: requests~=2.0; python_version == "3.7"
+Requires-Dist: scikit-learn~=1.0.2; python_version == "3.7"
+Requires-Dist: scipy==1.7.3; python_version == "3.7"
+Requires-Dist: torch==1.13.1; python_version == "3.7"
+Requires-Dist: torchvision==0.14.1; python_version == "3.7"
+Requires-Dist: urllib3~=1.26.6; python_version == "3.7"
+Requires-Dist: biopython~=1.81; python_version == "3.8"
+Requires-Dist: Cython~=0.29.37; python_version == "3.8"
+Requires-Dist: matplotlib~=3.5.3; python_version == "3.8"
+Requires-Dist: numpy~=1.24; python_version == "3.8"
+Requires-Dist: pandas~=1.1; python_version == "3.8"
+Requires-Dist: pomegranate~=0.14; python_version == "3.8"
+Requires-Dist: requests~=2.0; python_version == "3.8"
+Requires-Dist: scikit-learn~=1.0.2; python_version == "3.8"
+Requires-Dist: scipy==1.10.1; python_version == "3.8"
+Requires-Dist: torch==1.13.1; python_version == "3.8"
+Requires-Dist: torchvision==0.14.1; python_version == "3.8"
+Requires-Dist: urllib3~=1.26.6; python_version == "3.8"
+Requires-Dist: biopython~=1.81; python_version == "3.9"
+Requires-Dist: Cython~=0.29.37; python_version == "3.9"
+Requires-Dist: matplotlib~=3.5.3; python_version == "3.9"
+Requires-Dist: numpy~=1.24; python_version == "3.9"
+Requires-Dist: pandas~=1.1; python_version == "3.9"
+Requires-Dist: pomegranate~=0.14; python_version == "3.9"
+Requires-Dist: requests~=2.0; python_version == "3.9"
+Requires-Dist: scikit-learn~=1.0.2; python_version == "3.9"
+Requires-Dist: scipy==1.10.1; python_version == "3.9"
+Requires-Dist: torch==1.13.1; python_version == "3.9"
+Requires-Dist: torchvision==0.14.1; python_version == "3.9"
+Requires-Dist: urllib3~=1.26.6; python_version == "3.9"
+Requires-Dist: biopython~=1.83; python_version == "3.10"
+Requires-Dist: Cython~=0.29.37; python_version == "3.10"
+Requires-Dist: matplotlib~=3.5.3; python_version == "3.10"
+Requires-Dist: numpy~=1.26.4; python_version == "3.10"
+Requires-Dist: pandas~=1.5.3; python_version == "3.10"
+Requires-Dist: pomegranate~=0.14; python_version == "3.10"
+Requires-Dist: requests~=2.31.0; python_version == "3.10"
+Requires-Dist: scikit-learn~=1.0.2; python_version == "3.10"
+Requires-Dist: scipy~=1.12.0; python_version == "3.10"
+Requires-Dist: torch~=1.13.1; python_version == "3.10"
+Requires-Dist: torchvision~=0.14.1; python_version == "3.10"
+Requires-Dist: urllib3~=1.26.6; python_version == "3.10"
+Requires-Dist: biopython~=1.83; python_version == "3.11"
+Requires-Dist: matplotlib~=3.8.3; python_version == "3.11"
+Requires-Dist: numpy~=1.26.4; python_version == "3.11"
+Requires-Dist: pandas~=1.5.3; python_version == "3.11"
+Requires-Dist: pomegranate~=0.14.9; python_version == "3.11"
+Requires-Dist: requests~=2.31.0; python_version == "3.11"
+Requires-Dist: scikit-learn~=1.4.1.post1; python_version == "3.11"
+Requires-Dist: scipy~=1.12.0; python_version == "3.11"
+Requires-Dist: torch~=2.2.0; python_version == "3.11"
+Requires-Dist: urllib3~=1.26.18; python_version == "3.11"
+Requires-Dist: biopython~=1.83; python_version == "3.12"
+Requires-Dist: matplotlib~=3.8.3; python_version == "3.12"
+Requires-Dist: numpy~=1.26.4; python_version == "3.12"
+Requires-Dist: pandas~=2.1.0; python_version == "3.12"
+Requires-Dist: pomegranate~=0.14.9; python_version == "3.12"
+Requires-Dist: requests~=2.31.0; python_version == "3.12"
+Requires-Dist: scikit-learn~=1.4.1.post1; python_version == "3.12"
+Requires-Dist: scipy~=1.12.0; python_version == "3.12"
+Requires-Dist: torch~=2.2.0; python_version == "3.12"
+Requires-Dist: urllib3~=1.26.6; python_version == "3.12"
 
 <h1 align="center">
   <a href="bio2byte.be/b2btools" target="_blank" ref="noreferrer noopener">
@@ -94,37 +151,61 @@ Use this example as an entry point when you have a FASTA file containing one or
 #### Predicting biophysical features in JSON format
 
 ```python
+# Import necessary modules:
+# 'json' for file operations
+# and specific functions/classes from 'b2bTools'.
 import json
 from b2bTools import SingleSeq, constants
 
+# Define the path to the input FASTA file containing the protein sequences.
 input_fasta = "/path/to/example.fasta"
 
+# Initialize a SingleSeq object with the input FASTA file.
 single_seq = SingleSeq(input_fasta)
 
-# Predict features using 'dynamine', 'disomine', 'efoldmine', 'agmata', 'psper'
+# Use the predict method of the SingleSeq object to run predictions using a predefined set of tools.
+# All the available tools are specified in 'constants.PREDICTOR_NAMES', which includes 'dynamine', 'disomine', 'efoldmine', 'agmata', 'psper'.
 single_seq.predict(tools=constants.PREDICTOR_NAMES)
 
+# Retrieve all the predictions made by the previously specified tools and store them in a dictionary.
 predictions = single_seq.get_all_predictions()
 
-for sequence_id, prediction_values in predictions.items():
+# Extract the protein predictions from the overall predictions dictionary.
+protein_predictions = predictions['proteins']
+
+# Iterate through the protein predictions. Each iteration provides a sequence ID and its associated prediction values.
+for sequence_id, prediction_values in protein_predictions.items():
+    # Open a new JSON file for each sequence ID to write the prediction values.
+    # The file is named using the input FASTA file path and the sequence ID.
     with open(f"{input_fasta}_{sequence_id}.json", "w") as fp:
+        # Write the prediction values to the file in a pretty-printed JSON format.
         json.dump(prediction_values, fp, indent=4)
+
+# Additionally, save the execution metadata associated with the predictions into a CSV file.
+with open(f"{input_fasta}_exec_metadata.json", "w") as fp:
+    # Write the execution metadata to the file in a pretty-printed JSON format.
+    json.dump(predictions['metadata'], fp, indent=4)
 ```
 
 #### Predicting biophysical features in tabular formats (CSV, TSV)
 
 ```python
+# Import necessary libraries:
 import json
 from b2bTools import SingleSeq, constants
 
+# Define the path to the input FASTA file containing the protein sequences.
 input_fasta = "/path/to/example.fasta"
 
 single_seq = SingleSeq(input_fasta)
-
-# Predict features using 'dynamine', 'disomine', 'efoldmine', 'agmata', 'psper'
 single_seq.predict(tools=constants.PREDICTOR_NAMES)
 
+# Output the predictions in a tabular format to a CSV file.
+# The method 'get_all_predictions_tabular' is called
+# with the filename (appending '_residues.csv' to the input FASTA path) and the separator set to a comma (',')
 single_seq.get_all_predictions_tabular(f"{input_fasta}_residues.csv", sep=",")
+
+# Additionally, output the predictions in a tabular format to a TSV file.
 single_seq.get_all_predictions_tabular(f"{input_fasta}_residues.tsv", sep="\t")
 ```
 
@@ -154,31 +235,47 @@ The output contains these columns:
 In case you need to plot the prediction values:
 
 ```python
-import json
+# Import necessary libraries.
 
+import json
 from b2bTools import SingleSeq, constants
 from matplotlib import pyplot as plt
 
+# Define the path to the input FASTA file containing the protein sequences.
 input_fasta = "/path/to/example.fasta"
 
-single_seq = SingleSeq(input_fasta)
+# Specify the ID of the sequence for which predictions are to be made.
 sequence_id = "SEQ001"
 
-# Predict features using 'dynamine'
+# Predict features using 'DynaMine'.
+single_seq = SingleSeq(input_fasta)
 single_seq.predict(tools=constants.TOOL_DYNAMINE)
 
 predictions = single_seq.get_all_predictions()
+protein_predictions = predictions['proteins']
 
-backbone_pred = predictions[sequence_id]['backbone']
-sidechain_pred = predictions[sequence_id]['sidechain']
+# Extract backbone and side chain predictions for the specified sequence ID. These predictions are part of the
+# protein features predicted by DynaMine.
+backbone_pred = protein_predictions[sequence_id]['backbone']
+sidechain_pred = protein_predictions[sequence_id]['sidechain']
 
+# Plot the backbone and sidechain predictions using matplotlib. Two lines are plotted: one for backbone predictions
+# and another for side chain predictions. The x-axis represents amino acid positions, and the y-axis represents
+# the prediction values.
 plt.plot(range(len(backbone_pred)), backbone_pred, label = "Backbone")
 plt.plot(range(len(backbone_pred)), sidechain_pred, label = "Sidechain")
 
+# Add a legend to the plot to differentiate between the backbone and sidechain lines.
 plt.legend()
+
+# Label the x-axis as 'aa_position' to indicate amino acid positions and the y-axis as 'pred_values' to indicate
+# the values of the predictions.
 plt.xlabel('aa_position')
 plt.ylabel('pred_values')
-plt.show()
+
+# Save the plot as an image file. The file path includes the input FASTA file path with a '.png' extension,
+# indicating the plot is saved in PNG format.
+plt.savefig("/path/to/example.fasta.png")
 ```
 
 #### Extracting metadata
@@ -186,160 +283,21 @@ plt.show()
 For extracting metadata from the prediction values in tabular format when analyzing Single Sequence input:
 
 ```python
+# Import necessary libraries.
 from b2bTools import SingleSeq, constants
 
+# Define the path to the input FASTA file containing the protein sequences.
 input_fasta = "/path/to/example.fasta"
 
 single_seq = SingleSeq(input_fasta)
 
-# Predict features using 'dynamine', 'disomine', 'efoldmine', 'agmata', 'psper'
+# Predict features using 'dynamine', 'disomine', 'efoldmine', 'agmata', 'psper'.
 single_seq.predict(tools=constants.PREDICTOR_NAMES)
 
-predictions = single_seq.get_all_predictions()
+# Retrieve and save the metadata associated with the predictions to a CSV file.
 single_seq.get_metadata(f"{input_fasta}.csv")
 ```
 
-The output will include these columns:
-
-- sequence_id
-- length
-- Execution times:
-  - dynamine_execution_time
-  - disomine_execution_time
-  - efoldmine_execution_time
-  - agmata_execution_time
-  - psper_execution_time
-- DynaMine:
-  - Backbone:
-    - backbone_mean
-    - backbone_median
-    - backbone_variance
-    - backbone_stdev
-    - backbone_pvariance
-    - backbone_pstdev
-    - backbone_min
-    - backbone_max
-  - Side chain:
-    - sidechain_mean
-    - sidechain_median
-    - sidechain_variance
-    - sidechain_stdev
-    - sidechain_pvariance
-    - sidechain_pstdev
-    - sidechain_min
-    - sidechain_max
-  - ppII:
-    - ppII_mean
-    - ppII_median
-    - ppII_variance
-    - ppII_stdev
-    - ppII_pvariance
-    - ppII_pstdev
-    - ppII_min
-    - ppII_max
-  - Coil:
-    - coil_mean
-    - coil_median
-    - coil_variance
-    - coil_stdev
-    - coil_pvariance
-    - coil_pstdev
-    - coil_min
-    - coil_max
-  - Sheet:
-    - sheet_mean
-    - sheet_median
-    - sheet_variance
-    - sheet_stdev
-    - sheet_pvariance
-    - sheet_pstdev
-    - sheet_min
-    - sheet_max
-  - Helix:
-    - helix_mean
-    - helix_median
-    - helix_variance
-    - helix_stdev
-    - helix_pvariance
-    - helix_pstdev
-    - helix_min
-    - helix_max
-- EFoldMine:
-  - earlyFolding_mean
-  - earlyFolding_median
-  - earlyFolding_variance
-  - earlyFolding_stdev
-  - earlyFolding_pvariance
-  - earlyFolding_pstdev
-  - earlyFolding_min
-  - earlyFolding_max
-- AgMata:
-  - agmata_mean
-  - agmata_median
-  - agmata_variance
-  - agmata_stdev
-  - agmata_pvariance
-  - agmata_pstdev
-  - agmata_min
-  - agmata_max
-- DisoMine:
-  - disoMine_mean
-  - disoMine_median
-  - disoMine_variance
-  - disoMine_stdev
-  - disoMine_pvariance
-  - disoMine_pstdev
-  - disoMine_min
-  - disoMine_max
-- PSPer:
-  - Protein Score:
-    - protein_score
-  - Complexity:
-    - complexity_mean
-    - complexity_median
-    - complexity_variance
-    - complexity_stdev
-    - complexity_pvariance
-    - complexity_pstdev
-    - complexity_min
-    - complexity_max
-  - ARG:
-    - arg_mean
-    - arg_median
-    - arg_variance
-    - arg_stdev
-    - arg_pvariance
-    - arg_pstdev
-    - arg_min
-    - arg_max
-  - TYR:
-    - tyr_mean
-    - tyr_median
-    - tyr_variance
-    - tyr_stdev
-    - tyr_pvariance
-    - tyr_pstdev
-    - tyr_min
-    - tyr_max
-  - RRM:
-    - RRM_mean
-    - RRM_median
-    - RRM_variance
-    - RRM_stdev
-    - RRM_pvariance
-    - RRM_pstdev
-    - RRM_min
-    - RRM_max
-  - Disorder:
-    - disorder_mean
-    - disorder_median
-    - disorder_variance
-    - disorder_stdev
-    - disorder_pvariance
-    - disorder_pstdev
-    - disorder_min
-    - disorder_max
-
 ### Multiple Sequences Alignment predictions
 
 Use the following example as an entry point when you have a MSA file input. There is a live demo available on Google Colab: [link](https://colab.research.google.com/github/Bio2Byte/public_notebooks/blob/main/Bio2ByteTools_v3_multipleseq_pypi.ipynb)
@@ -347,80 +305,83 @@ Use the following example as an entry point when you have a MSA file input. Ther
 #### Predicting biophysical features in JSON format
 
 ```python
+# Import necessary modules: 'json' for handling JSON data, and functionalities from 'b2bTools'.
 import json
 from b2bTools import MultipleSeq, constants
 
+# Define the path to the input MSA (Multiple Sequence Alignment) file.
 input_msa = "/path/to/example.afa"
 
+# Create an instance of the MultipleSeq class, which is designed to handle operations on multiple sequences.
 multiple_seq = MultipleSeq()
+
+# Load the MSA file and specify the prediction tools to use. The 'from_aligned_file' method is used to read
+# the alignment from the specified file, and 'constants.PREDICTOR_NAMES' defines the list of available tools.
 multiple_seq.from_aligned_file(input_msa, tools=constants.PREDICTOR_NAMES)
 
+# Retrieve the predictions for the MSA.
 predictions = multiple_seq.get_all_predictions_msa()
 
-for sequence_id, prediction_values in predictions.items():
-    with open(f"{input_msa}_{sequence_id}.json", "w") as fp:
+# Extract the predictions specific to proteins from the overall predictions.
+proteins_predictions = predictions['proteins']
+
+# Iterate over the protein predictions.
+for sequence_id, prediction_values in proteins_predictions.items():
+    with open(f"{input_msa}_{sequence_id}_from_msa.json", "w") as fp:
         json.dump(prediction_values, fp, indent=4)
 
+# Retrieve distribution data for all predictions made on the MSA.
 distributions_dict = multiple_seq.get_all_predictions_msa_distrib()
 distributions = distributions_dict['results']
 
+# Save the distribution data to a JSON file.
 with open(f"{input_msa}_distributions.json", "w") as fp:
     json.dump(distributions, fp, indent=4)
 
+# Save the execution metadata to a JSON file.
+execution_metadata = multiple_seq.get_execution_metadata()
+with open(f"{input_msa}_execution_metadata.json", "w") as fp:
+    json.dump(execution_metadata, fp, indent=4)
 ```
 
 #### Predicting biophysical features in tabular formats (CSV, TSV)
 
 ```python
+# Import the MultipleSeq class and constants from the b2bTools package.
 from b2bTools import MultipleSeq, constants
 
+# Specify the path to the input MSA file.
 input_msa = "/path/to/example.afa"
 
+# Instantiate a MultipleSeq object.
 multiple_seq = MultipleSeq()
 multiple_seq.from_aligned_file(input_msa, tools=constants.PREDICTOR_NAMES)
 
+# Output the prediction results in a tabular format as a CSV file.
 multiple_seq.get_all_predictions_tabular(f"{input_msa}_residues.csv", sep=",")
+
+# Similarly, output the prediction results in a tabular format as a TSV file.
 multiple_seq.get_all_predictions_tabular(f"{input_msa}_residues.tsv", sep="\t")
 ```
 
-The output contains these columns:
-
-| Predictor | Value | Data type |
-| ----------| ----- | --------- |
-| None | sequence_id | String |
-| None | residue | Char |
-| None | residue_index | Integer |
-| PSPer | RRM | Float |
-| AgMata | agmata | Float |
-| PSPer | arg | Float |
-| DynaMine | backbone | Float |
-| DynaMine | coil | Float |
-| PSPer | complexity | Float |
-| DisoMine | disoMine | Float |
-| PSPer | disorder | Float |
-| EFoldMine | earlyFolding | Float |
-| DynaMine | helix | Float |
-| DynaMine | ppII | Float |
-| DynaMine | sheet | Float |
-| DynaMine | sidechain | Float |
-| PSPer | tyr | Float |
-
 #### Extracting metadata
 For extracting metadata from the prediction values in tabular format when analyzing MSA input:
 
 ```python
+# Import the MultipleSeq class and a constants module from the b2bTools package.
 from b2bTools import MultipleSeq, constants
 
+# Specify the file path to the input MSA file.
 input_msa = "/path/to/example.afa"
 
+# Create an instance of the MultipleSeq class.
 multiple_seq = MultipleSeq()
 multiple_seq.from_aligned_file(input_msa, tools=constants.PREDICTOR_NAMES)
 
-multiple_seq.get_metadata(f"{input_msa}.csv")
+# Extract and save the metadata associated with the analysis of the MSA file.
+multiple_seq.get_metadata(f"{input_msa}_metadata.csv")
 ```
 
-
-
 ## 💻 Installation
 
 From the official documentation:
@@ -494,7 +455,7 @@ The class `NefIO` provides the following static methods:
 Usage:
 
 ```python
-from b2bTools.general.parsers.alignments import AlignmentsIO
+from b2bTools.general.parsers.nef import NefIO
 ```
 
 #### 📄 NMR-STAR files