iFad: an integrative factor analysis model for drug-pathway association inference

Area of Science:

• Computational biology
• Systems pharmacology
• Bioinformatics

Background:

• Pathway-based drug discovery is crucial due to unknown drug mechanisms and off-target effects.
• Inferring drug-pathway associations is key for system-based pharmacological research.
• Transcriptome data (gene expression) can reflect pathway activity and aid in drug target identification.

Purpose of the Study:

• To develop a statistical model for jointly analyzing gene expression and drug sensitivity data.
• To investigate gene-pathway-drug-pathway associations.
• To incorporate prior knowledge of gene-pathway and drug-pathway relationships into the analysis.

Main Methods:

• Developed iFad, a Bayesian sparse factor analysis model.
• Utilized a collapsed Gibbs sampling algorithm for model inference.
• Applied the model to simulated datasets and the NCI-60 cell line data.

Main Results:

• The iFad model demonstrated satisfactory performance on both simulated and real-world datasets.
• The model successfully aids in the discovery of new drug-pathway association relationships.
• Results suggest iFad is a promising tool for identifying drug targets.

Conclusions:

• iFad provides a robust framework for pathway-based drug discovery.
• The model facilitates integrative analysis of multi-omics data.
• This approach enhances the understanding of drug action within biological pathways.