Limits...
Functional organization and its implication in evolution of the human protein-protein interaction network.

Zhao Y, Mooney SD - BMC Genomics (2012)

Bottom Line: However, being able to capture topological properties does not necessarily mean it correctly reproduces how networks emerge and evolve.Consistently, we further found that the topological unit is also the functional unit of the network.Given our observations, we suggest that its significance should not be overlooked when studying network evolution.

View Article: PubMed Central - HTML - PubMed

Affiliation: Buck Institute for Research on Aging, Novato, California, USA.

ABSTRACT

Background: Based on the distinguishing properties of protein-protein interaction networks such as power-law degree distribution and modularity structure, several stochastic models for the evolution of these networks have been purposed, motivated by the idea that a validated model should reproduce similar topological properties of the empirical network. However, being able to capture topological properties does not necessarily mean it correctly reproduces how networks emerge and evolve. More importantly, there is already evidence suggesting functional organization and significance of these networks. The current stochastic models of evolution, however, grow the network without consideration for biological function and natural selection.

Results: To test whether protein interaction networks are functionally organized and their impacts on the evolution of these networks, we analyzed their evolution at both the topological and functional level. We find that the human network is shown to be functionally organized, and its function evolves with the topological properties of the network. Our analysis suggests that function most likely affects local modularity of the network. Consistently, we further found that the topological unit is also the functional unit of the network.

Conclusion: We have demonstrated functional organization of a protein interaction network. Given our observations, we suggest that its significance should not be overlooked when studying network evolution.

Show MeSH
A dendogram reflecting the functional relationships for proteins from different temporal groups. The tree suggests a progressive functional change of the human PPI network during evolution. The observed clustering would not have occurred by random chance (P < 0.001). As indicated by the arrow, TG3 and TG4 in the tree are separately grouped into two clades, which are compared to the accelerated topology changes during this period as shown in Figure 3. TG: Temporal group.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3375200&req=5

Figure 4: A dendogram reflecting the functional relationships for proteins from different temporal groups. The tree suggests a progressive functional change of the human PPI network during evolution. The observed clustering would not have occurred by random chance (P < 0.001). As indicated by the arrow, TG3 and TG4 in the tree are separately grouped into two clades, which are compared to the accelerated topology changes during this period as shown in Figure 3. TG: Temporal group.

Mentions: We next trace the functional evolution of a PPI network, which has not been clearly investigated by previous studies. If the functional evolution of the network is adaptive, a progressive change in functions along with evolutionary age is hypothesized. By contrast, this is not expected if the change in function occurs at a single point in time or is neutral. To test this hypothesis, we first defined functional distance using the Mahalanobis distance between a pair of genes based on their function annotations (For details, see Methods). Compared to the functional similarity measurements used by other studies [27,28,30], the functional distance method used here considers the overall annotation pattern and is more informative. We calculated the functional distance for all possible pairs of nodes in the network, and the distances were further averaged and grouped into a 6 X 6 matrix using 6 X 6 TG combinations as indices (Table 4). By reordering the 6 X 6 matrix using agglomerative hierarchical clustering [34], as shown in Figure 4, the pattern of the tree obtained is consistent with the evolutionary time-scale, and with adjacent TGs being clustered with shorter functional distances (or higher functional similarity). To test whether the observed clustering is statistically significant, we calculate a p-value by comparing the observed summed distances along the tree against a distribution produced by randomly permuting TG information of each protein (1000 permutations). In an attempt to quantify the relationship, we correlated the averaged functional distance with differences in group age, which has a coefficient of 0.725 (Spearman rho, p < 0.001), confirming a progressive functional change of the network. Interestingly, TG3 and TG4 in the tree are grouped separately into two clades. Considering our observation of accelerated topological changes during this period, this further indicates an association between topological evolution and functional evolution.


Functional organization and its implication in evolution of the human protein-protein interaction network.

Zhao Y, Mooney SD - BMC Genomics (2012)

A dendogram reflecting the functional relationships for proteins from different temporal groups. The tree suggests a progressive functional change of the human PPI network during evolution. The observed clustering would not have occurred by random chance (P < 0.001). As indicated by the arrow, TG3 and TG4 in the tree are separately grouped into two clades, which are compared to the accelerated topology changes during this period as shown in Figure 3. TG: Temporal group.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3375200&req=5

Figure 4: A dendogram reflecting the functional relationships for proteins from different temporal groups. The tree suggests a progressive functional change of the human PPI network during evolution. The observed clustering would not have occurred by random chance (P < 0.001). As indicated by the arrow, TG3 and TG4 in the tree are separately grouped into two clades, which are compared to the accelerated topology changes during this period as shown in Figure 3. TG: Temporal group.
Mentions: We next trace the functional evolution of a PPI network, which has not been clearly investigated by previous studies. If the functional evolution of the network is adaptive, a progressive change in functions along with evolutionary age is hypothesized. By contrast, this is not expected if the change in function occurs at a single point in time or is neutral. To test this hypothesis, we first defined functional distance using the Mahalanobis distance between a pair of genes based on their function annotations (For details, see Methods). Compared to the functional similarity measurements used by other studies [27,28,30], the functional distance method used here considers the overall annotation pattern and is more informative. We calculated the functional distance for all possible pairs of nodes in the network, and the distances were further averaged and grouped into a 6 X 6 matrix using 6 X 6 TG combinations as indices (Table 4). By reordering the 6 X 6 matrix using agglomerative hierarchical clustering [34], as shown in Figure 4, the pattern of the tree obtained is consistent with the evolutionary time-scale, and with adjacent TGs being clustered with shorter functional distances (or higher functional similarity). To test whether the observed clustering is statistically significant, we calculate a p-value by comparing the observed summed distances along the tree against a distribution produced by randomly permuting TG information of each protein (1000 permutations). In an attempt to quantify the relationship, we correlated the averaged functional distance with differences in group age, which has a coefficient of 0.725 (Spearman rho, p < 0.001), confirming a progressive functional change of the network. Interestingly, TG3 and TG4 in the tree are grouped separately into two clades. Considering our observation of accelerated topological changes during this period, this further indicates an association between topological evolution and functional evolution.

Bottom Line: However, being able to capture topological properties does not necessarily mean it correctly reproduces how networks emerge and evolve.Consistently, we further found that the topological unit is also the functional unit of the network.Given our observations, we suggest that its significance should not be overlooked when studying network evolution.

View Article: PubMed Central - HTML - PubMed

Affiliation: Buck Institute for Research on Aging, Novato, California, USA.

ABSTRACT

Background: Based on the distinguishing properties of protein-protein interaction networks such as power-law degree distribution and modularity structure, several stochastic models for the evolution of these networks have been purposed, motivated by the idea that a validated model should reproduce similar topological properties of the empirical network. However, being able to capture topological properties does not necessarily mean it correctly reproduces how networks emerge and evolve. More importantly, there is already evidence suggesting functional organization and significance of these networks. The current stochastic models of evolution, however, grow the network without consideration for biological function and natural selection.

Results: To test whether protein interaction networks are functionally organized and their impacts on the evolution of these networks, we analyzed their evolution at both the topological and functional level. We find that the human network is shown to be functionally organized, and its function evolves with the topological properties of the network. Our analysis suggests that function most likely affects local modularity of the network. Consistently, we further found that the topological unit is also the functional unit of the network.

Conclusion: We have demonstrated functional organization of a protein interaction network. Given our observations, we suggest that its significance should not be overlooked when studying network evolution.

Show MeSH