octad

OCTAD provides a platform for virtually screening compounds targeting precise cancer patient groups by identifying drugs that reverse the gene expression signature of disease.

Key Features:

• Deep-learning based reference tissue selection: Identifies appropriate normal tissue references for comparison with disease samples.
• Disease gene expression signature creation: Generates gene expression signatures that distinguish disease from normal tissue.
• Pathway enrichment analysis: Identifies biological pathways associated with disease gene expression signatures.
• Drug reversal potency scoring: Quantifies the ability of compounds to reverse disease gene expression signatures.
• Cancer cell line selection: Identifies relevant cancer cell lines for experimental validation.
• Drug enrichment analysis: Assesses statistical enrichment of drug classes or mechanisms among top-scoring compounds.
• In silico hit validation: Provides computational validation of predicted drug hits.

Scientific Applications:

• Cancer therapeutic discovery: Screens ~12,000 distinct compounds to identify potential therapeutics for ~50 cancer types using ~20,000 patient tissue samples.
• Precision oncology: Targets drug screening to precise cancer patient groups based on their specific gene expression profiles.
• Drug repurposing: Identifies existing compounds that may reverse cancer gene expression signatures by suppressing over-expressed genes and stimulating weakly expressed ones.

Methodology:

OCTAD uses deep learning for reference tissue selection and computes drug reversal potency scores based on the degree to which compounds reverse disease gene expression signatures.