A simple approach to ranking differentially expressed gene expression time courses through Gaussian process...

Area of Science:

• Genomics
• Computational Biology
• Bioinformatics

Background:

• Gene expression time-series analysis is crucial for biological studies.
• Current methods often ignore the temporal nature of time-series data.
• Key analyses include filtering inactive genes and identifying differential expression.

Purpose of the Study:

• To propose a simple Gaussian process (GP) model for gene expression time-series analysis.
• To account for the temporal dependencies in gene expression data.
• To improve the accuracy of identifying active genes and differentially expressed genes.

Main Methods:

• Utilized Gaussian process regression to estimate continuous gene expression trajectories.
• Developed an approach for filtering "quiet" genes.
• Quantified differential gene expression using expression ratios.
• Assessed performance using ROC curves and compared against a hierarchical Bayesian model (BATS).

Main Results:

• The proposed Gaussian process approach effectively filters quiet genes and quantifies differential expression.
• Demonstrated superior performance compared to the BATS model on both simulated and experimental data.
• Achieved higher accuracy in ranking gene expression patterns.

Conclusions:

• Gaussian processes provide an efficient and user-friendly framework for analyzing microarray time-series data.
• The GP framework naturally handles biological replicates and missing values.
• Offers confidence intervals for estimated gene expression curves, making GPs a recommended standard tool.