A dependency graph approach for the analysis of differential gene expression profiles

Area of Science:

• Bioinformatics
• Systems Biology
• Genomics
• Proteomics
• Transcriptomics

Background:

• Interpreting differential gene expression data at the biological process and pathway level remains challenging.
• Translating descriptive omics data into meaningful biological context requires advanced analytical frameworks.

Purpose of the Study:

• To develop a gene-centric dependency graph approach for interpreting omics profiles.
• To enable the identification of affected biological networks and cellular states from omics data.

Main Methods:

• Constructed a dependency graph using gene functional categorization, tissue-specific expression, protein interactions, and subcellular localization.
• Computed pairwise gene dependencies and represented them as weighted edges.
• Mapped omics profiles onto the graph to identify connecting features and characterize cellular states.

Main Results:

• The gene-centric dependency graph approach successfully supports the interpretation of omics profiles at the network level.
• Identified disease-associated subgraphs by analyzing differential gene expression data in B-cell lymphoma.
• Demonstrated the utility of the graph in pinpointing key biological networks relevant to specific cellular conditions.

Conclusions:

• The proposed dependency graph method offers a robust framework for understanding complex omics data.
• This approach facilitates the discovery of biologically relevant subgraphs and network motifs.
• The method is effective for characterizing cellular states and identifying disease-specific molecular signatures.