A crystallographic and geometric approach for the computational prediction of conformational epitopes
Heugle, Ravi John
MetadataShow full item record
The conformational epitope(CE) is a spatially grouped cluster of residues belonging to an antigen to which an antibody or B cell immunoglobulins binds (Muller and Jacoby, 2009). Accurate and reliable predictions of CEs are necessary for rational vaccine design and immunodiagnostics (Ponomarenko and Regenmortel, 2009). In this work, we explore the relationship between residues involved in crystal contacts and those belonging to the CE. With this purpose in mind, we present a new computational approach for predicting portions of CEs based on the premise that proteins establish specific crystal contacts to optimize the enthalpic contribution to the free energy during crystallization (Durbin and Feher, 1996) and in doing so mark those residues with high binding affinity - an attribute shared by epitope residues. Furthermore, it has been shown that a disproportionate amount of binding free energy of an antibody-antigen complex is accounted for by a small cluster of residues at the interface, regarded as a "hot spot"(Van Regenmortel, 2002). By identifying the crystal contacts of antigens, we explore the notion that these residues likely belong to the epitope and correspondingly, the hot spot. We hypothesize that crystal contacts are established at specific locations to involve residues that largely contribute to the binding free energy of an antibody-antigen complex. In our approach, we first find the antigen's crystal contacts, before calculating the solvent accessible surface area(SA) and energy of interaction for each crystal contact. Additionally, we calculate the protrusion index of the residues involved in the crystal contact using the computational package ElliPro (Ponomarenko et al., 2008). We measure the viability that the residues of a specific contact are participating in the epitope based on a combined measure of their solvent accessible surface area, interaction energy, and protrusivity. A benchmark dataset of bound antibody-antigen complexes and corresponding unbound antigens is used for analysis. We show that we can with predict a subset of residues belonging to the epitope with high relatively high accuracy. Furthermore, our approach illustrates that the residues driving the establishment of crystal contacts are shared by those used to establish antibody-antigen binding.
Thesis (M.Sc.Eng.) PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at firstname.lastname@example.org. Thank you.