Dynamic modularity of functional categories associated with protein complexes in cellular processes

Abstract

A living cell is a complex and dynamic system, where gene or protein activities, and its interactions (gene-gene, protein-protein, and protein-gene) exhibit various expression profiles and spatial compartmentalization in the level of transcription, post-transcription, post-translation. Understanding how the phenotypes and behaviors of cells are controlled is one of the major challenges in biological research. The responses of cells to genetic perturbations or environmental cues are controlled by functional modules. For the purpose of identifying the perturbed activities of functional modules, integrative method using gene expression profiles is often used in the field of bioinformatics. It is reported that functional module-based approaches represent more efficient performance than individual gene-based approaches. Previous study using Gene Ontology and KEGG pathways, however, create large list of annotations that make it difficult to derive consolidated information and often include over-generalized terms. The majority of Gene Set Enrichment Analysis (GSEA) approaches identify the differentially expressed functional modules under different conditions. Differential expressions-based approaches are not always significant change of functional modules. To improve these limitations, we need to employ the concrete and informative functional modules. In addition, functional module-based approaches must consider the differential expressions as well as the modular changes of functional modules. In this thesis, we developed two methods for understanding cellular processes under certain conditions: the composite function enrichment and the dynamic functional modularity analysis. The composite function enrichment using $\emph{apriori}$ association rules infers significant association between well characterized annotation terms such as functional modules. The interrelationship of annotation terms can be more clearly delineated by integrating protein complexes with biolog...