Computational chemogenomics: core concepts, applications, and critical aspects of model building
Dr. J.B. Brown (Kyoto University Graduate School of Medicine)
Chemogenomics is a field of chemical science concerned with the use of control probes to determine the potential protein targets in a biological system (forward chemogenomics), and the application of a chemical library on the reduced protein target space (reverse chemogenomics). The computational analog of chemogenomics is to use a database of annotated compound-protein interactions in order to either assess the contents of the database for patterns (the SAR matrix concept), or to build prediction models that can be used for predicting the bioactivity of new compounds. In this talk, I will discuss in larger detail three aspects of computational chemogenomics. First, I will cover the problem formulation and provide examples of real data with a running prediction program. Second, I will cover some studies experimentally validating the predictions and talk about issues of data and model quality. Finally, I will provide a preview into the future of chemogenomic modeling where model methods are currently being developed.
Biophysical views of protein interaction surfaces
Dr. Nobuyuki Uchikoga (Department of Physics Faculty of Science and Engineering, Chuo University)
To understanding protein-protein interaction (PPI) mechanisms, one of the most available methods is rigid-body docking process, generating many protein complexes (decoys) as candidates of the native complex structure. For PPI prediction cases using rigid-body docking, there is a serious case that protein decoys do not include near-natives. This problem results from insufficient docking search spaces. For overcoming this problem, I paid attention to properties of protein interaction surfaces, which are generated by rigid-body docking. Furthermore, physicochemical properties of interaction surfaces were observed and compared among difference ligand proteins docked with a certain receptor protein.