Speaker:  Teresa Przytycka

Title: Co-evolution of interacting proteins: what is behind the mirrortree?

May 6, 2008,; 11 am; B2 Library

 

It has been observed that the evolutionary trees of interacting proteins often display a higher level of similarity than those 
of non-interacting proteins. This indicates that interacting proteins are subject to common evolutionary constraints and constitutes 
the basis of a method to predict protein interactions known as the mirrortree method. While several papers (including contributions 
from our group) demonstrated utility of the method, they also indicated that the method is far from perfect. To advance further this
 approach and to understand fully its limitations it is necessary to have a closer look at the nature of the co-evolutionary signal
 measured by the mirror tree method. In this talk I will describe our recent work directed towards separating the functional co-evolution 
of interacting partners from co-divergence due to common evolutionary history. A related issue that we have addressed is understanding 
of the source of the signal.   One possible explanation is the existence of compensatory mutations between partners’ binding sites 
to maintain proper binding. This explanation, however, has been recently challenged, and it has been argued that the signal of correlated 
evolution uncovered by the mirrortree method is unrelated to any co-variation between binding sites. We found these findings to be 
only partially true -  the correlation between evolutionary trees of interacting domains cannot exclusively be attributed to 
the co-evolution of the binding sites or to common evolutionary pressure exerted on the whole protein domain sequence but 
a is combination of the two.