TEMPO

TEMPO models pathway-level temporal dysregulation in gene expression by constructing virtual time series from cross-sectional microarray and RNA-seq datasets to detect pathway-specific temporal outliers.

Key Features:

• Pathway-Based Outlier Detection: Applies a pathway-centric approach to identify gene sets that deviate from expected temporal expression patterns.
• Virtual Time Series Construction: Synthesizes virtual time series by integrating temporal information across individuals from microarray and RNA-seq datasets to approximate longitudinal profiles.
• Cross-sectional Data Handling: Captures dynamic processes when experiments are limited to single time points by leveraging cross-sectional transcriptional profiling.
• Application to Complex Diseases: Detects temporal dysregulation in biologically relevant pathways in autism spectrum disorders, Huntington’s disease, Alzheimer’s disease, and chronic obstructive pulmonary disease (COPD).
• Signal over Noise: Distinguishes meaningful temporal signals from noise in virtual time series derived from cross-sectional data.
• Pathway-level Insights: Produces pathway-level results that can provide distinct insights compared to standard temporal or gene set analysis methods.

Scientific Applications:

• Disease temporal dysregulation analysis: Identifying temporally disrupted pathways in autism spectrum disorders, Huntington’s disease, Alzheimer’s disease, and COPD.
• Developmental and functional inference: Inferring functional, temporal, or developmental processes from transcriptional data when longitudinal sampling is unavailable.
• Pathway-level hypothesis generation: Generating hypotheses about pathway dysregulation that may not be evident from individual-gene analyses.

Methodology:

Integrates temporal information from cross-sectional microarray and RNA-seq datasets to construct virtual time series and applies pathway-centric outlier detection to identify temporal dysregulation in gene sets.