A regulation probability model-based meta-analysis of multiple transcriptomics data sets for cancer biomarker...

Area of Science:

• Bioinformatics
• Computational Biology
• Genomics

Background:

• Omics data integration presents challenges in bioinformatics due to large data volumes.
• Identifying consistent gene activity patterns across studies requires careful handling of data heterogeneity.

Purpose of the Study:

• To develop a robust meta-analysis method for identifying differentially expressed genes (DEGs).
• To address and manage data heterogeneity in multi-study omics data integration.

Main Methods:

• Proposes jGRP, a regulation probability model-based meta-analysis.
• Integrates multiple transcriptomics datasets in a gene regulatory space, not expression space.
• Transforms gene expression profiles into united gene regulation profiles by estimating probabilities of gene regulation events.

Main Results:

• jGRP accurately and flexibly identifies DEGs in gene regulation space.
• Evaluated on simulation and real-world cancer datasets, demonstrating effectiveness and efficiency.
• The method successfully captures and manages data heterogeneity across different studies and platforms.

Conclusions:

• Data heterogeneity significantly impacts DEG identification in meta-analysis.
• jGRP offers a united framework to manage study heterogeneity, outperforming other methods in sensitivity.
• jGRP is a versatile standalone tool for robust meta-analysis of omics data.