Characterizing dynamic changes in the human blood transcriptional network.

Area of Science:

• Systems biology
• Transcriptomics
• Network analysis

Background:

• Systematic gene expression data over time reveals network changes.
• Food intake impacts blood gene expression and is linked to metabolic diseases.

Purpose of the Study:

• Identify genes driving gene expression changes due to time and food intake.
• Infer causal relationships in the blood transcriptional network.

Main Methods:

• Applied Granger causality test and dynamic Bayesian networks to time-series blood gene expression data.
• Utilized simulation to validate combining short time series for Granger causality inference.

Main Results:

• Identified PER1 as a key regulator of the blood transcriptional network, influencing circadian rhythm-regulated processes.
• Showed that over 72% of dynamic connections in fasted and fed networks are self-links.
• Demonstrated dynamic linking of previously disconnected processes like inflammation and lipid metabolism upon food intake.

Conclusions:

• Food intake profoundly impacts the dynamic co-regulation of metabolism, immune response, apoptosis, and circadian rhythm.
• Findings have implications for designing disease association and drug response studies.