sgrpy 0.1.9

Python package for fragmenting colored graphs

SGR-Py (SubGraph Regularization)

This is a Python module which performs molecule encoding, linear feature regression, and subgraph regularization.

About the project

This project provides various information-based graph routines (including the titular subgraph regularization algorithm, implemented in Rust), linear algebra routines, and an RDKit-based shim which translates molecules and SMILES strings into labeled graphs via several possible methods.

Getting started

Installation

Installation from PyPI

1. Enter a virtual environment, if you want to use one.

2. Run pip install sgrpy, or pip install sgrpy[rdkit] if you want to include the optional RDKit shim.

Installation from Wheel

1. Navigate to relevant Tag or Release (right sidebar).

2. Download relevant .whl file for your system.

3. Enter a virtual environment, if you want to use one.

4. Run pip install rdkit to install RDKit if you want to use the optional RDKit shim.

5. Run pip install sgrpy-*.whl on the wheel file you downloaded to install the main package.

Installation from Source

Windows

1. Install Visual Studio 2017 or later (or Build Tools for Visual Studio with Visual C++ option)

2. Install Rust toolchain

3. Install Python and RDKit if you want to use the RDKit shim.

4. Enter a virtual environment, if you want to use one.

5. Clone this repo, then install using pip install . in the top-level directory.

MacOS/Linux

1. Install Rust toolchain

2. Install Python and RDKit if you want to use the RDKit shim.

3. Enter a virtual environment, if you want to use one.

4. Clone this repo, then install using pip install . in the top-level directory.

Tutorials

• CLI Tutorial
• Python API Tutorial (to be released)

Contact

William Sprague (main developer) - [e-mail]

Project Link: https://github.com/wsprague-nu/sgrpy