Causal network analysis of omics data using prior knowledge databases.

Area of Science:

• Genomics
• Systems Biology
• Bioinformatics

Background:

• Identifying causal relationships in omics data is crucial for understanding biological processes.
• Challenges include complex molecular networks and limitations of observational data.
• Structured prior knowledge from databases is key for accurate causal inference.

Purpose of the Study:

• To systematically review data-driven causal omics analysis methods that integrate prior knowledge.
• To categorize these methods based on prior knowledge integration levels.
• To provide a practical guide for researchers on selecting and applying causal analysis methods.

Main Methods:

• Systematic literature review of causal omics analysis methods.
• Grouping methods into regulon-level, flow-level, and network-level approaches.
• Analysis of method strengths, limitations, and applications.

Main Results:

• Identified three main approaches: regulon-level, flow-level, and network-level.
• Each approach offers different trade-offs in interpretation, scope, and complexity.
• Methods demonstrated utility in diverse applications like cancer, kidney disease, and neurodegenerative disorders.

Conclusions:

• Causal omics analysis methods, guided by prior knowledge, are essential for biological discovery.
• The choice of method depends on the research question, data, and desired level of detail.
• Further research is needed to address limitations and enhance causal inference in omics data.