Quick Start

Basic Usage

Run the full IDAP pipeline on a single MAF file:

python main_pipeline.py \
    --maf path/to/sample.maf \
    --cancer NSCLC \
    --oncokb_token YOUR_ONCOKB_TOKEN \
    --annotator path/to/oncokb-annotator/MafAnnotator.py \
    --pubmed_token YOUR_PUBMED_TOKEN \
    --txgnn_data path/to/txgnn/data \
    --txgnn_root path/to/TxGNN \
    --outdir output/sample_001 \
    --patient_id sample_001

Parameters

Parameter	Required	Description
--maf	Yes	Path to input MAF file
--cancer	Yes	Cancer type abbreviation (e.g., NSCLC, BRCA, COAD, GBM, PDAC)
--oncokb_token	Yes	OncoKB API token
--annotator	Yes	Path to OncoKB MafAnnotator.py
--pubmed_token	Yes	NCBI PubMed API key
--txgnn_data	Yes	Path to TxGNN data folder
--txgnn_root	Yes	Path to TxGNN package root
--outdir	No	Output directory (default: output)
--patient_id	No	Patient identifier for the report

Supported Cancer Types

Abbreviation	Full Name
NSCLC	Non-Small Cell Lung Cancer
LUAD	Lung Adenocarcinoma
LUSC	Lung Squamous Cell Carcinoma
BRCA	Breast Invasive Carcinoma
COAD	Colon Adenocarcinoma
CRC	Colon Adenocarcinoma
GBM	Glioblastoma Multiforme
PDAC	Pancreatic Adenocarcinoma
SKCM	Melanoma
OV	Ovarian Cancer
HNSC	Head and Neck Cancer

Output

After a successful run, the output directory contains:

output/sample_001/
├── oncokb_output.tsv          # OncoKB variant annotations
├── pubmed_output.tsv          # PubMed gene-drug mention counts
├── txgnn_output.tsv           # TxGNN graph-based drug candidates
├── clinicaltrials_output.tsv  # Clinical trial metadata
├── top20_drugs.png            # Bar plot of top 20 drugs
├── txgnn_graph.png            # Gene-drug network visualization
├── final_report.xlsx          # Full Excel report (7 sheets)
└── final_report.pdf           # PDF summary report

See Output Format for detailed descriptions of each file.