Home | Research » Projects > Modeling Biological Networks

Modeling Biological Networks


IV.1 Coordinators
IV.2 Participants
IV.3 Introduction
IV.4 Background and Significance
IV.5 Research Plan
IV.6 Specific Subprojects IV.7 Connection to Specific Projects 2 (cytoskeleton) and 3 (organogenesis)
IV.8 Timeline

< Previous | Page 15 of 35 | Next >


IV.6.ii.e Preliminary Results and Research Plan:
IV.6.ii.e.1 Aim 1:

Using the existing data on protein-protein interactions in yeast, we have constructed a topological representation of these interactions. The topological map clearly demonstrates that cytoskeletal and signaling proteins tend to cluster, indicating a significant correlation between these two groups of proteins.

We are currently characterizing this correlation quantitatively. One way is to define an abstract distance measure on the topological map and to analyze the "distance" between the individual proteins. Correlation between two particular proteins would require that the distance between them be shorter than between two randomly chosen proteins. Our preliminary results indicate that the average distance between proteins within the pools of cytoskeletal and signaling proteins is significantly shorter than between proteins that do not belong to these pools (several proteins belong to both pools), suggesting that proteins associated with the cytoskeleton have correlated functions. The enhanced correlation for signaling proteins suggests that the various signaling pathways may "communicate" with each other, as is being discovered experimentally (Tarricone, et al., 2001; Nishita et al., 2000). More significantly, the distance between cytoskeletal and signaling proteins is also shorter than in the rest of the protein pool, suggesting that the cytoskeleton actively participates in intracellular signaling.

Our research will develop other methods to investigate possible correlations between signaling and cytoskeleton-associated proteins. We will calculate from the topological map, the number of "nearest neighbors" (and more distant neighbors) of each cytoskeletal protein that belongs either to the pool of cytoskeletal proteins or to the pool of signaling proteins, or to neither. We will analyze this nearest neighbor distribution to establish whether signaling proteins preferentially accumulate near cytoskeletal proteins, as we expect.

IV.6.ii.e.2 Aim 2:

Though cytoskelatal and signaling proteins tend to cluster more than randomly chosen, similar size subsets on the protein interaction map, we must prove their preferential correlation by investigating their interactions with other functionally related protein classes.