CeBiTec Colloquium: 2008/11/17 Dr. Imtiaz Khan, Department of Pathology, Cardiff University Hospital, Cardiff, United Kingdom

2008/11/17, 17:15

CeBiTec Laboratory Building, Room G2-104

Bioinformatics methodology to integrate genomic and cellular data - comprehensive view on the dynamic biological system

Abstract:

The molecular basis for understanding organism behaviour started with the premise that organisms are an assembly of different components which can be described in a hierarchical fashion according to their functionality, size etc.  Different ‘–omic’ approaches like genomics, proteomics have facilitated us to identify and characterize the components and structures of biological networks. But for multi-cellular organisms it is the network of networks that fundamentally defines the biological processes and leads to a combinatorial number of billions of possible interactions to test.  Again from the experimental point of view, these large scale ‘-omic’ data are not bona fide representations of the innate cellular conditions, due to the methods of collection and exclusion of natural time dependent complexity and heterogeneity and thus the paradigm that the molecular basis of knowledge is the key for understanding the disease process and biology at the system level (bottom-up approach), has a biased and often limited view.  Which in turn invokes the strong requirements for top-down views where a living system is more than the sum of it parts and it acquires emergent properties that its individual components may not have.  Explaining these often counterintuitive properties in terms of the underlying components requires the cell to be placed as the irreducible and integrating unit that links molecular information with behavioural information.  According to differentiation status, cells represent the elementary functional units of multi-cellular organisms, and disease represents molecular alterations that impact upon the integrity and functions of cellular systems determined by both genotype and external or internal influences.  The challenge for establishing a theoretical framework suitable to predict disease and related biological processes, is to generate a integrative multidimensional data (usually 5D – x,y,z,t,g).  Such a five-dimensional data set could describe the expression patterns and subcellular localization patterns for proteins and genes (g) at certain time point (t) and place (xyz) in three dimensions in an organism.  Bifurcating cell lineage map provide a theoretical solution for linking cell physiology (phenotypic events, xyzt) with the protein activity and gene expression. (molecular events, tg).  The hypothesis is that integrating this rich yet ‘static’ molecular information (eg microarray data) with the ‘dynamic’ responses of the cell (eg a phenotype of delayed cell cycle arrest) is a prerequisite for our understanding of biology at a systems level including disease processes.