Document Type

Technical Report


Computer Science and Engineering

Publication Date






Technical Report Number



The ability to effectively model the interaction between proteins is an important and open problem. In molecular biology it is well accepted that from sequence arises form and from form arises function but relating structure to function remains a challenge. The function of a given protein is defined by its interactions. Likewise a malfunction or a change in protein-protein interactions is a hallmark of many diseases. Many researchers are studying the mechanisms of protein-protein interactions and one of the overarching goals of the community is to predict whether two proteins will bind, and if so what the final conformation will be. Attention is seldom paid to the association pathways that allow two proteins to bind. Evidence has shown that the information in the association pathways can play a vital role in understanding the interaction itself. This thesis presents a novel and scalable approach to computing association pathways between two proteins using the Probabilistic Roadmap (PRM) framework. We will discuss the challenges in extending PRM to the domain of protein-protein interactions such as performing structural mappings in a reduced space of flexibility, and sampling high dimensional conformation spaces. We will present analysis of individual association pathways as well as methods for estimating collective properties of the energy landscape. Our results indicate that these methods can discriminate between true and false protein binding interfaces. Finally, we will present condensing methods such as pathway clustering and visualization using dimensionality reduction that can be be applied to create compact representations of the interaction space.


Permanent URL: