Exploring Bistable Molecular Switches and Their Functional Roles Using a Meta-Analysis That Integrates Expression Data with Regulatory Networks
- Type: Conference Presentation
- Conference Type:
AIChE Annual Meeting
- Presentation Date:
November 8, 2010
- Skill Level:
You will be able to download and print a certificate for PDH credits once the content has been viewed. If you have already viewed this content, please click here to login.
An important topic in systems biology is to understand the dynamic properties of the biological networks. Previous studies have provided important insights into the mechanistic principles of dynamic behaviors by applying kinetic modeling to condition-specific and very small networks. However, it remains unclear how to bridge such dynamic behaviors described by bottom-up mechanisms with top-down global functional analysis, i.e. to associate the dynamic features of single molecules with the large-scale network structure and functionality. Here we explore potential bistable molecular switches based on gene expression profiles across many different conditions, representing various biological states. Distinct from most approaches proposed on the meta-analysis of microarray data that attempt to identify differentially expressed genes, we analyze the expression profiles in the context of multistability as a dynamic property of kinetic systems. We apply mixture Gaussian model to characterize the bimodality in gene expression profiles and perform Expectation Maximization algorithm together with information criterions to identify potential bistable states. We hypothesize that the dynamic property of the molecular components within a network is contingent on the network organization and the resulting dynamic behavior corresponds to specific biological functions, i.e. positive feedback. Our study integrates yeast microarray, transcriptional regulatory network, protein interaction network and genetic knock-out data to reveal the relationship between bistability, local network structure and functionality, and provides proof-of-concept of our proposed meta-analysis approach. Finally we apply our meta-analysis approach on human gene expression datasets comparing multiple cancer cell types and tissues with non-cancerous diseases, to identify potential “molecular switches” that are universal across cancer types. These common modules may define the switch-like behavior and thereby determine the status of the biological systems and whether it could be perturbed into tumorigenesis.