Skip to content

Latest commit

 

History

History
103 lines (68 loc) · 4.13 KB

README.md

File metadata and controls

103 lines (68 loc) · 4.13 KB

License Maintenance Workflow GitHub issues GitHub contributors GitHub release PyPI version fury.io versions DOI Documentation Status

GitHub Logo

This repository contains RanDepict, an easy-to-use utility to generate a big variety of chemical structure depictions (random depiction styles and image augmentations) based on RDKit, CDK, Indigo and PIKAChU.

Usage

  • To use RanDepict, clone the repository to your local disk and make sure you install all the necessary requirements.

Installation

$ git clone https://github.com/OBrink/RanDepict.git
$ cd RanDepict 
$ python -m pip install -U pip #Upgrade pip
$ pip install .

Alternative

$ python -m pip install -U pip #Upgrade pip
$ pip install git+https://github.com/OBrink/RanDepict.git

Install from PyPI

$ pip install RanDepict

Development

Note We recommend using RanDepict inside a Conda environment to facilitate the installation of the dependencies.

  • Conda can be downloaded as part of the Anaconda or the Miniconda plattforms (Python 3.8). We recommend miniconda3. On Linux you can get it with:
$ wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
$ bash Miniconda3-latest-Linux-x86_64.sh
$ echo -e "channels:\n - conda-forge\n - nodefaults" > ~/.condarc
$ conda update conda
$ conda install conda-libmamba-solver
$ conda config --set experimental_solver libmamba
$ conda create --name RanDepict python=3.8
$ conda activate RanDepict
# pypi has rdkit so not necessary to install it using conda
# $ conda install -c rdkit rdkit
$ pip install -e ".[dev]" # will install pytest and tox
# To run tests for python 3.8
$ tox -e py38

Basic usage:

from RanDepict import RandomDepictor

smiles = "CN1C=NC2=C1C(=O)N(C(=O)N2C)C"

with RandomDepictor() as depictor:
    image = depictor(smiles)

Have a look at the examples in our demo notebook. A more detailed documentation is provided here.

Here are some examples of depictions of caffeine without augmentations (left) and with augmentations (right) that were automatically created using RanDepict.

Cite Us

Acknowledgements

More information about our research group

GitHub Logo

Alt