Generalised Read Across (GenRA) in Python
Project description
Generalised Read-Across (GenRA) in Python
Read-Across is widely used to fill data-gaps for untested chemicals. We developed Genralised Read-Across (GenRA) as a computational toxicology tool to mimic a human expert’s manual reasoning based on similarity-weighted activity. This repository contains a Python 3 implementation for GenRA, called genra-py, which is based on the scikit-learn estimator. We also describe two potential uses-cases for genra-py that uses published chemical structure, bioactivity and toxicity data.
Easy starts
pip install genra
or try our Docker image from on dockerhub at [https://hub.docker.com/r/patlewig/genra-py]
The image contains the scipy Jupyter notebook, RDKit and a pip installable version of genra-py (https://github.com/i-shah/genra-py/).
In a terminal type:
docker run --rm -p 8888:8888 -e JUPYTER_ENABLE_LAB=yes --user $(id -u):$(id -g) --group-add users -v "$PWD":/home/jovyan patlewig/genra-py
Copy/paste the resulting url link into a new browser window. This should start the Jupyter lab session from your current working directory.
To avoid typing the long argument in the terminal, using docker-compose is an alternative means of running the container.
Type docker-compose -f genra-docker-compose.yml up
To stop the container simply type:
docker-compose -f genra-docker-compose.yml down
Alternatives
Running the notebooks in this repository requires Python 3, Anaconda, Jupter and some additional configuration.
- Install Python 3, anaconda/conda and Jupyter Lab
- Clone this repo:
git clone https://github.com/i-shah/genra-py.git
- Go into genra-py directory and create genra-py conda environment:
make -n create_environment
- Activate conda environment:
conda activate genra-py
- Add this conda environment as a kernel to jupyter-lab:
ipython kernel install --user --name=genra-py
- Copy the notebooks/dotenv file to notebooks/.env and edit the environemnt variables (replace path_to_top with the correct directory name):
TOP=path_to_top/genra-py SRC=path_to_top/genra-py/src DAT=path_to_top/genra-py/data FIG=path_to_top/genra-py/figs
Further details are provided in the notebooks/manual directory.
See https://github.com/patlewig/UNC_Rax and run the example using the Binder https://mybinder.org/
Project Organization
├── LICENSE
├── Makefile <- Makefile with commands like `make data` or `make train`
├── README.md <- The top-level README for developers using this project.
├── data <- Data from public domain sources.
│ └─ shah-2016 <- Data from https://doi.org/10.1016/j.yrtph.2016.05.008
│ └─ helman-2019 <- Data from https://doi.org/10.1016/j.yrtph.2016.05.008
|
├── notebooks <- Jupyter notebooks
| |
| ├─dotenv <- copy this to ".env" and edit this file
| ├─app-note <- use-cases described in manuscript
| └─manual <- user-manual as a jupyter notebook
│
├── requirements.txt <- The requirements file for reproducing the analysis environment, e.g.
│ generated with `pip freeze > requirements.txt`
├── genra-py.yml <- The spec for creating a conda environment.
| conda env create -f condaenv.yml
├── dist <- Source and Wheel Distributions
|
└── genra <- Source code
├─chm <- Chemical structure processing
├─rax <- Read Across prediction
└─utl <- Utilities
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for genra-0.1.5-py2.py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 30780b960327beaeb016d5b0f433f811573f8a82c9beaa20a2279d91847e0df8 |
|
MD5 | ca7690967afd716a9801c20cbf78bf76 |
|
BLAKE2b-256 | ec6d916947b7cd78be46ad43f7462f72565327bcb76e8c3802e072b402363853 |