Installation¶

Requirements¶

Python 3.8+
OncoKB API token (free for research use)
NCBI PubMed API key (recommended)

Install from GitHub¶

git clone https://github.com/joonan-lab/IDAP-pipeline.git
cd IDAP-pipeline
pip install -r requirements.txt

Dependencies¶

The main dependencies are listed in requirements.txt:

Package	Purpose
`pandas`	Data manipulation
`requests`	API queries (PubMed, ClinicalTrials.gov)
`numpy`	Numerical operations
`torch`	TxGNN knowledge graph loading
`dgl`	Graph neural network framework
`networkx`	Network visualization
`matplotlib`	Plotting
`reportlab`	PDF report generation
`xlsxwriter`	Excel report generation
`txgnn`	Biomedical knowledge graph

TxGNN Setup¶

The TxGNN module requires the TxGNN package and its associated data:

# Install TxGNN
pip install txgnn

# Or clone from source
git clone https://github.com/mims-harvard/TxGNN.git

The knowledge graph data will be automatically downloaded on first run.

API Tokens¶

IDAP requires API tokens passed as command-line arguments. Do not hardcode tokens in source files.

# OncoKB: Request at https://www.oncokb.org/api-access
# PubMed: Request at https://www.ncbi.nlm.nih.gov/account/settings/

python main_pipeline.py \
    --oncokb_token YOUR_ONCOKB_TOKEN \
    --pubmed_token YOUR_PUBMED_TOKEN \
    ...

Tip

Store tokens in environment variables for convenience:

export ONCOKB_TOKEN="your_token_here"
export PUBMED_TOKEN="your_token_here"