Discovering Biology in Periodic Data through Phase Set Enrichment Analysis (PSEA)

Area of Science:

• Genomics
• Systems Biology
• Bioinformatics

Background:

• Existing gene enrichment tools struggle with periodic biological systems (e.g., cell cycle, circadian clock) due to assumptions of monotonicity and inaccurate phase distance measurements.
• This limitation hinders the analysis of time-series gene expression data crucial for understanding biological rhythms.

Purpose of the Study:

• To develop a novel method, Phase Set Enrichment Analysis (PSEA), for incorporating prior biological knowledge into the analysis of periodic gene expression data.
• To identify biologically relevant gene sets exhibiting temporally coordinated expression within periodic systems.

Main Methods:

• PSEA was developed to specifically address the challenges of analyzing periodic data, unlike existing enrichment tools.
• Performance was benchmarked against current methods using synthetic gene sets from von Mises distributions, evaluating sensitivity across various set sizes and distributions.
• PSEA was applied to four published datasets, including circadian atlases, restricted feeding paradigms, and cell-cycle studies.

Main Results:

• PSEA demonstrated enhanced sensitivity compared to existing methods across a range of periodic data characteristics.
• The sensitivity of PSEA was independent of the mean expression phase of gene sets, a key advantage over other tools.
• Analysis of published datasets revealed temporal orchestration in immune and cell-cycle pathways, disruption of metabolic synchrony in the liver under restricted feeding, distinct characteristics of CLOCK-bound circadian targets, and cell-cycle dysregulation in cancer cells.

Conclusions:

• PSEA provides a robust and sensitive approach for analyzing large-scale periodic biological data, offering significant advantages over existing methods.
• The tool enhances the ability to uncover biologically meaningful patterns and insights from complex, time-dependent gene expression datasets.
• PSEA has been implemented as a user-friendly Java application to facilitate its adoption and use within the research community.