Drug target prioritization by perturbed gene expression and network information

Area of Science:

• Pharmacology and Computational Biology
• Investigating drug mechanisms and target identification using systems biology approaches.

Background:

• Drugs modulate cancer cell transcriptomes through target proteins and downstream effectors.
• Understanding these perturbations is key to identifying drug targets and pathways.

Purpose of the Study:

• To systematically analyze drug perturbations and protein interaction networks for uncovering drug targets and key pathways.
• To evaluate the efficacy of network topological measures and introduce a novel metric for target prioritization.

Main Methods:

• Systematic analysis of over 500 drugs from the Connectivity Map dataset.
• Assessment of gene expression changes post-drug treatment.
• Application of network topological measures, including the novel 'local radiality' metric.

Main Results:

• Drug target gene expression is often not significantly affected by drug perturbation alone.
• Network topological measures, particularly local radiality, significantly improve drug target prioritization.
• Local radiality successfully identifies cancer-specific drug-target-gene pathways and suggests targets with potentially fewer side effects.

Conclusions:

• Gene expression changes alone are insufficient for identifying drug targets.
• Network analysis, especially using local radiality, is a powerful tool for prioritizing drug targets and understanding drug mechanisms.
• The local radiality measure offers a promising approach for discovering novel therapeutic targets with improved specificity.