PAGE: an R package for network detection of multivariate error-prone gene expression data with the availability of...

Area of Science:

• Bioinformatics and Computational Biology
• Systems Biology
• Genomics

Background:

• Gene expression data often involves high-dimensional and multivariate variables, necessitating network structure analysis for disease pathway identification.
• Incorporating auxiliary variables (covariates) can enhance network detection but poses challenges in selection and relationship modeling.
• Measurement errors in gene expression data can compromise the accuracy of network inference.

Purpose of the Study:

• To develop a robust analytical tool for uncovering gene expression network structures.
• To address challenges in variable selection, measurement error correction, and network estimation.
• To provide a publicly accessible R package for bioinformatics research.

Main Methods:

• Development of the R package PAGE, incorporating functions for measurement error correction, variable selection, and network estimation.
• Utilized both linear and nonlinear modeling frameworks to characterize response-covariate relationships.
• Demonstrated package functionality using a yeast cell cycle dataset.

Main Results:

• The PAGE package effectively handles complex network structures in gene expression data.
• Simulation studies confirmed the critical importance of measurement error correction.
• PAGE demonstrates utility in analyzing error-prone biological datasets.

Conclusions:

• The R package PAGE is a valid tool for complex gene network analysis.
• Measurement error correction is essential for accurate biological network inference.
• PAGE provides a valuable solution for handling error-prone gene expression data.