cagqsar 1.0.0__tar.gz

cagqsar-1.0.0/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Sathish Kumar M Ponnaiya
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

cagqsar-1.0.0/PKG-INFO

@@ -0,0 +1,171 @@
+Metadata-Version: 2.4
+Name: cagqsar
+Version: 1.0.0
+Summary: A complete CLI QSAR pipeline for drug discovery and predictive toxicology
+Author: Sathish Kumar M Ponnaiya
+Classifier: Programming Language :: Python :: 3
+Classifier: Operating System :: POSIX :: Linux
+Classifier: Intended Audience :: Science/Research
+Classifier: Topic :: Scientific/Engineering :: Chemistry
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: numpy
+Requires-Dist: pandas
+Requires-Dist: scipy
+Requires-Dist: scikit-learn
+Requires-Dist: xgboost
+Requires-Dist: matplotlib
+Requires-Dist: rdkit
+Provides-Extra: torch
+Requires-Dist: torch; extra == "torch"
+Dynamic: license-file
+
+# CAG-QSAR CLI Tool
+
+A complete, production-grade command-line QSAR (Quantitative Structure-Activity Relationship) modeling pipeline. This tool automates the process of data curation, molecular descriptor calculation, feature selection, data splitting, model building, and rigorous validation.
+
+---
+
+## Installation
+
+You can install this package in Linux/WSL using three different methods, depending on your environment.
+
+### Prerequisites
+Make sure you have Python (version >= 3.8) and `pip` installed:
+```bash
+sudo apt update
+sudo apt install python3 python3-pip python3-venv -y
+```
+
+### Option A: Install from PyPI (Once Published)
+After publishing the package to PyPI, you can create a virtual environment and install the tool globally or locally using `pip`:
+```bash
+# 1. Create a virtual environment
+python3 -m venv qsar_env
+source qsar_env/bin/activate
+
+# 2. Install the package from PyPI
+pip install cagqsar
+
+# 3. Run the CLI tool
+cagqsar --help
+```
+
+### Option B: Local Source Install (No root access required)
+You can build and install the package locally from the repository folder:
+```bash
+# 1. Navigate to the repository directory
+cd git_qsar
+
+# 2. Create and activate a virtual environment
+python3 -m venv qsar_env
+source qsar_env/bin/activate
+
+# 3. Install the package locally
+pip install .
+
+# 4. Run the CLI tool
+cagqsar --help
+```
+
+### Option C: Editable Development Mode
+If you plan to modify the source code of the pipeline and want changes to reflect instantly:
+```bash
+pip install -e .
+```
+
+---
+
+## Command Line Usage
+
+After installation, run the application from any directory in your terminal:
+
+```bash
+cagqsar --data <dataset_csv> --smiles <smiles_column> --activity <activity_column> [options]
+```
+
+### Core CLI Arguments:
+* `--data`: Path to the CSV dataset (Required).
+* `--smiles`: Column name containing SMILES strings (Required).
+* `--activity`: Column name containing activities in nM (Required).
+* `--model`: Regression algorithm to train: `mlr` (MLR), `pls` (PLS), `rf` (Random Forest), `svr` (SVM), `xgb` (XGBoost), or `gnn` (Graph Neural Network) (Default: `pls`).
+* `--split`: Splitting method: `random` or `pca` (Kennard-Stone PCA-distance split) (Default: `pca`).
+* `--test_size`: Fraction of data allocated to the test set (Default: `0.2`).
+* `--var_thresh`: Variance filter threshold for dropping constant descriptors (Default: `0.01`).
+* `--corr_thresh`: Correlation threshold for collinearity filter (Default: `0.85`).
+* `--y_rand_runs`: Number of Y-randomization validation loops (Default: `50`).
+* `--fingerprints`: Flag to compute 2D fingerprints (Morgan/ECFP + MACCS keys) in addition to physical descriptors.
+* `--out_dir`: Directory to export curated data, model reports, trained model binaries, and evaluation plots (Default: `qsar_output`).
+
+---
+
+## Programmatic Import in Python
+
+Once the package is installed, you can import and use any of its internal logic (like the structure curator or descriptor calculator) in your own scripts:
+
+```python
+from cagqsar import curate_molecule, get_rdkit_descriptors
+
+# 1. Clean a SMILES structure and remove salt fragments
+clean_smiles, mol = curate_molecule("CN(C)C(=O)c1ccccc1.Cl", Chem.SaltRemover.SaltRemover())
+
+# 2. Extract standard RDKit descriptors
+descriptors = get_rdkit_descriptors(mol)
+```
+
+---
+
+## Publishing to GitHub & PyPI
+
+Follow these instructions to publish your code for public access.
+
+### 1. Publishing to GitHub
+Initialize the local git repository, commit the files, and push to GitHub:
+```bash
+# 1. Initialize repository
+git init
+
+# 2. Add files (automatically respects .gitignore)
+git add .
+
+# 3. Create initial commit
+git commit -m "feat: initial release of cagqsar v1.0.0"
+
+# 4. Set main branch name
+git branch -M main
+
+# 5. Add remote GitHub link and push
+git remote add origin https://github.com/YOUR_USERNAME/cagqsar.git
+git push -u origin main
+```
+
+### 2. Publishing to PyPI
+To make the tool installable globally via `pip install cagqsar`, build and upload the package distributions to the Python Package Index (PyPI):
+
+```bash
+# 1. Install packaging build tools
+pip install --upgrade build twine
+
+# 2. Compile source distribution (sdist) and binary wheel (bdist_wheel)
+python3 -m build
+
+# 3. Verify build files
+twine check dist/*
+
+# 4. Upload to PyPI (requires PyPI API Token)
+python3 -m twine upload dist/*
+```
+
+---
+
+## Acknowledgments & Credits
+
+* **Concept, Idea & Planning**: Sathish Kumar M Ponnaiya (SKM Ponnaiya).
+* **Infrastructure & Support**: **Ponnaiya's Code And Genome Pvt Ltd, Madurai** (System support, server resources, internet facilities, and infrastructure).
+* **AI Coding Partner**: Pair-programmed and optimized using **Antigravity**, a Google DeepMind agentic coding system.
+* **Large Language Model (LLM)**: Driven by Google's **Gemini 3.5 Flash**.
+* **Access Provider**: Grateful to **Jio** for enabling Gemini Premium access.
+* **Test Dataset**: Sourced from the public **BindingDB database**.
+
+*This software is open-access and free for all users under the terms of the MIT License.*

cagqsar-1.0.0/README.md

@@ -0,0 +1,148 @@
+# CAG-QSAR CLI Tool
+
+A complete, production-grade command-line QSAR (Quantitative Structure-Activity Relationship) modeling pipeline. This tool automates the process of data curation, molecular descriptor calculation, feature selection, data splitting, model building, and rigorous validation.
+
+---
+
+## Installation
+
+You can install this package in Linux/WSL using three different methods, depending on your environment.
+
+### Prerequisites
+Make sure you have Python (version >= 3.8) and `pip` installed:
+```bash
+sudo apt update
+sudo apt install python3 python3-pip python3-venv -y
+```
+
+### Option A: Install from PyPI (Once Published)
+After publishing the package to PyPI, you can create a virtual environment and install the tool globally or locally using `pip`:
+```bash
+# 1. Create a virtual environment
+python3 -m venv qsar_env
+source qsar_env/bin/activate
+
+# 2. Install the package from PyPI
+pip install cagqsar
+
+# 3. Run the CLI tool
+cagqsar --help
+```
+
+### Option B: Local Source Install (No root access required)
+You can build and install the package locally from the repository folder:
+```bash
+# 1. Navigate to the repository directory
+cd git_qsar
+
+# 2. Create and activate a virtual environment
+python3 -m venv qsar_env
+source qsar_env/bin/activate
+
+# 3. Install the package locally
+pip install .
+
+# 4. Run the CLI tool
+cagqsar --help
+```
+
+### Option C: Editable Development Mode
+If you plan to modify the source code of the pipeline and want changes to reflect instantly:
+```bash
+pip install -e .
+```
+
+---
+
+## Command Line Usage
+
+After installation, run the application from any directory in your terminal:
+
+```bash
+cagqsar --data <dataset_csv> --smiles <smiles_column> --activity <activity_column> [options]
+```
+
+### Core CLI Arguments:
+* `--data`: Path to the CSV dataset (Required).
+* `--smiles`: Column name containing SMILES strings (Required).
+* `--activity`: Column name containing activities in nM (Required).
+* `--model`: Regression algorithm to train: `mlr` (MLR), `pls` (PLS), `rf` (Random Forest), `svr` (SVM), `xgb` (XGBoost), or `gnn` (Graph Neural Network) (Default: `pls`).
+* `--split`: Splitting method: `random` or `pca` (Kennard-Stone PCA-distance split) (Default: `pca`).
+* `--test_size`: Fraction of data allocated to the test set (Default: `0.2`).
+* `--var_thresh`: Variance filter threshold for dropping constant descriptors (Default: `0.01`).
+* `--corr_thresh`: Correlation threshold for collinearity filter (Default: `0.85`).
+* `--y_rand_runs`: Number of Y-randomization validation loops (Default: `50`).
+* `--fingerprints`: Flag to compute 2D fingerprints (Morgan/ECFP + MACCS keys) in addition to physical descriptors.
+* `--out_dir`: Directory to export curated data, model reports, trained model binaries, and evaluation plots (Default: `qsar_output`).
+
+---
+
+## Programmatic Import in Python
+
+Once the package is installed, you can import and use any of its internal logic (like the structure curator or descriptor calculator) in your own scripts:
+
+```python
+from cagqsar import curate_molecule, get_rdkit_descriptors
+
+# 1. Clean a SMILES structure and remove salt fragments
+clean_smiles, mol = curate_molecule("CN(C)C(=O)c1ccccc1.Cl", Chem.SaltRemover.SaltRemover())
+
+# 2. Extract standard RDKit descriptors
+descriptors = get_rdkit_descriptors(mol)
+```
+
+---
+
+## Publishing to GitHub & PyPI
+
+Follow these instructions to publish your code for public access.
+
+### 1. Publishing to GitHub
+Initialize the local git repository, commit the files, and push to GitHub:
+```bash
+# 1. Initialize repository
+git init
+
+# 2. Add files (automatically respects .gitignore)
+git add .
+
+# 3. Create initial commit
+git commit -m "feat: initial release of cagqsar v1.0.0"
+
+# 4. Set main branch name
+git branch -M main
+
+# 5. Add remote GitHub link and push
+git remote add origin https://github.com/YOUR_USERNAME/cagqsar.git
+git push -u origin main
+```
+
+### 2. Publishing to PyPI
+To make the tool installable globally via `pip install cagqsar`, build and upload the package distributions to the Python Package Index (PyPI):
+
+```bash
+# 1. Install packaging build tools
+pip install --upgrade build twine
+
+# 2. Compile source distribution (sdist) and binary wheel (bdist_wheel)
+python3 -m build
+
+# 3. Verify build files
+twine check dist/*
+
+# 4. Upload to PyPI (requires PyPI API Token)
+python3 -m twine upload dist/*
+```
+
+---
+
+## Acknowledgments & Credits
+
+* **Concept, Idea & Planning**: Sathish Kumar M Ponnaiya (SKM Ponnaiya).
+* **Infrastructure & Support**: **Ponnaiya's Code And Genome Pvt Ltd, Madurai** (System support, server resources, internet facilities, and infrastructure).
+* **AI Coding Partner**: Pair-programmed and optimized using **Antigravity**, a Google DeepMind agentic coding system.
+* **Large Language Model (LLM)**: Driven by Google's **Gemini 3.5 Flash**.
+* **Access Provider**: Grateful to **Jio** for enabling Gemini Premium access.
+* **Test Dataset**: Sourced from the public **BindingDB database**.
+
+*This software is open-access and free for all users under the terms of the MIT License.*

cagqsar-1.0.0/cagqsar/__init__.py

@@ -0,0 +1,18 @@
+# CAG-QSAR Package Initialization
+# Exposes key pipeline functions for programmatic use in Python scripts
+
+from .pipeline import (
+    curate_molecule,
+    curate_dataset,
+    get_rdkit_descriptors,
+    get_2d_fingerprints,
+    generate_descriptors,
+    select_features,
+    kennard_stone_split,
+    split_dataset,
+    evaluate_qsar_model,
+    main
+)
+
+__version__ = "1.0.0"
+__author__ = "Sathish Kumar M Ponnaiya"