GenRA-LTEA

Generalised Read Across (GenRA) in Python: Proof-of-Concept (LTEA)

Read-Across is extensively used to fill data gaps for compounds that have not been evaluated. We created Genralised Read-Across (GenRA) as a computational toxicology tool to simulate the manual reasoning of a human expert using similarity-weighted activity.

This repository provides a case example using genra-py, a Python 3 implementation of GenRA based on the scikit-learn estimator.  We show how to utilize genra-py to create a proof-of-concept utilizing published chemical structure, LTEA biological hit-call data, and toxicity data.

Getting Started

The entire analysis is implemented using open source tools and the genra-py package.

To get started:

pip install genra

Alternatively, either clone or download this repository:

Running the notebooks in this repository requires Python 3, Anaconda, Jupyter and some additional configuration.

1. Install Python 3, anaconda/conda and Jupyter Lab Clone this repo:

git clone https://github.com/i-shah/genra-ltea.git

2. Go into genra-ltea directory and create genra-py conda environment:

conda env create -f condaenv.yml

3. Activate conda environment:

conda activate genra-py

4. GenRA (genra-py) will need to be installed in this activated environment with: pip install genra

5. Open your Jupyter Notebook In the terminal, execute jupyter notebook {note: this may neet to be installed in environment.} Then open the notebook 001-tt-genra-ltea-2020 to start coding.

Further details are provided in the genra-py user manual

notebooks/manual/001-genra-py-user-manual.ipynb

Project Organization

No spaces in filenames!

├── 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       <- https://doi.org/10.1016/j.yrtph.2016.05.008
|
├── docs               <- A default Sphinx project; see sphinx-doc.org for details
│
├── notebooks          <- Jupyter notebooks (XXX-UU-DDDDDD.ipynb). Convention:- 
|   |                     XXX = numeric sequence 
|   |                     UU  = user initials
|   |                     DDDDDDD = descriptive string 
|   ├─is               <- Imran Shah
|   ├─gp               <- Grace Patlewicz
|   ├─tt               <- Tia Tate
│
├── requirements.txt   <- The requirements file for reproducing the analysis environment, e.g.
│                         generated with `pip freeze > requirements.txt`
├── condaenv.yaml      <- The spec for creating a conda environment. Generated using:
|                          conda env export > condaenv.yml
│                         Can create environment using:
|                          conda env create -f condaenv.yml
├── setup.py           <- makes project pip installable (pip install -e .) so src can be imported
|
├── tools              <- GenRA command-line tools (installed in /usr/local/bin) that depend on src 
└── src                <- Source code for use in this project.
    │
    └─genra          
        ├─db           <- Database access and etl
        ├─data         <- Data preparation / manipulation
        |  └─chm       <- chemical structure and physchem data processing 
        |  └─bio       <- bioactivity data processing             
        |  └─tox       <- toxicity data processing             
        ├─rax          <- Read Across prediction
        |  └─skl       <- Standalone based on scikit-learn
        |  └─srv       <- Server based on mongodb 
        ├─viz          <- Visualization 
        ├─utl          <- Utilities
        
       

Built using cookiecutter