TOX

Final Project for CS272.

Introduction to Biomedical Informatics Research
Methodology at Stanford University.

Team

 Ayush    Kevin    Shreyas    Tom 

Setup

Configuring your system for the project.

Conda Environment

Preparing the local environment.

1. conda create -n tox python=3.9

2. conda activate tox

3. conda install -c conda-forge biopython

4. conda install -c pytorch pytorch

5. conda install scikit-learn

Jupyter Notebook

Making Conda available in Jupyter.

1. conda install -c anaconda ipykernel

2. python -m ipykernel install --user --name=tox

Files

 

    Contains data from previous papers,
    including ToxIBTL, ToxDL & ToxinPred.

 

    Contains Python files for exploratory data analysis.

    This includes the reading in and wrangling of data into
    a standard format ( sequences and toxic / non-toxic ),
    identification of duplicate sequences, division of data
    into training and test, as well as analysis of sequence
    similarity.

 

    Contains data related to CD-HIT, which we use to
    determine sequences that are at least 40% similar.

 

    Contains Jupyter notebooks used in the
    process of developing our ToxIN model.

    The /ToxIBTL/ folder contains
    original code from ToxIBTL.