The overarching goal of the thesis project is to develop technologies to infer pathways implicated in periodontal diseases and tissue regeneration. Specifically, we will use methods to computationally annotate and predict protein interactions to investigate sub-networks within the human interactome that comprise these pathways. The thesis research will have two phases: (1) Development of a periodontal protein interaction database (PPID) and automated interaction predictions. The PPID will include known protein interactions, automated prediction of interactions using state of the art computational methods, and integration into visualized interaction networks. Automating interaction predictions will require integration of the phylogenetic profile, domain fusion, interolog, and gene neighbor techniques. Access to the database and prediction tools will be developed within the lab servers. (2) Experimental verification of predicted interactions. To evaluate prediction methods, proteins with high probability of interaction will be tested for functional interactions. Experiments will focus on protein pairs predicted to exist in a pathway with the NH2-propeptide of procollagen type I (N-propeptide). N-propeptide provides feedback for extracellular matrix production [1], an important aspect of bone formation. Verification of physical interactions will include co-immunoprecipitation, yeast two hybrid, and cell-based binding studies. Co-expression testing will include quantitative RT-PCR, SAGE, and/or gene array methods.
Showing the most recent 10 out of 13 publications