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

• Brinkhaus, H.O., Rajan, K., Zielesny, A. et al. RanDepict: Random chemical structure depiction generator. J Cheminform 14, 31 (2022). https://doi.org/10.1186/s13321-022-00609-4

Acknowledgements

• We would like to thank M. Isabel Agea and Tulay Muezzinoglu for their valuable contributions.

More information about our research group