Embedding mRNA stability in correlation analysis of time-series gene expression data.

Area of Science:

• Genomics
• Systems Biology
• Bioinformatics

Background:

• Gene co-regulation identification is crucial for understanding gene networks.
• Existing methods often overlook mRNA stability and dynamic variations.
• Gene expression time profiles are key to inferring regulatory relationships.

Purpose of the Study:

• Introduce a novel similarity metric, lead-lag R(2), for gene co-regulation analysis.
• Improve the accuracy of identifying co-regulated genes from time-series data.
• Explore the application of lead-lag R(2) in understanding protein complex formation dynamics.

Main Methods:

• Developed the lead-lag R(2) similarity metric.
• Applied the metric to yeast cell-cycle time-series gene expression data.
• Compared lead-lag R(2) performance against standard similarity measures.

Main Results:

• Lead-lag R(2) significantly outperforms standard metrics in identifying co-regulated genes.
• Identified a large number of novel putatively co-regulated genes.
• Demonstrated the metric's utility in uncovering relationships between gene expression and protein complex dynamics, including transient complexes.

Conclusions:

• Lead-lag R(2) is a powerful tool for accurate gene co-regulation identification.
• The metric enhances gene network inference by incorporating dynamic properties like mRNA stability.
• Lead-lag R(2) offers new insights into the temporal dynamics of biological processes, such as protein complex assembly.