Linking transcriptomic and proteomic data on the level of protein interaction networks

Area of Science:

• Systems Biology
• Genomics
• Proteomics
• Metabolomics

Background:

• Integrating multi-omics data (genome, transcriptome, proteome, metabolome) is crucial for a Systems Biology perspective.
• Directly merging transcriptomic and proteomic data is challenging due to limited overlap in differentially regulated features.

Purpose of the Study:

• To present procedures for integrating omics profiles using protein interaction networks.
• To exemplify this integration using transcriptomic and proteomic data from chronic kidney disease (CKD) studies.

Main Methods:

• Analysis of transcriptomic and proteomic datasets from CKD.
• Integration of omics profiles at the level of protein interaction networks (directed and undirected).
• Comparison of direct feature overlap versus pathway/process category overlap.

Main Results:

• Limited direct overlap of significantly affected genes and proteins was observed individually (e.g., collagen, uromodulin).
• Integration via protein interaction networks substantially increased overlap in pathway and process categories.
• Identified jointly relevant pathways include cell structure, cell adhesion, and immunity/defense mechanisms.

Conclusions:

• Protein interaction network-based integration effectively combines diverse omics data for functional interpretation.
• This approach enhances the understanding of cellular processes and disease mechanisms, particularly in complex conditions like CKD.
• Mapping multiple data sources onto networks reveals shared biological insights missed by individual analyses.