Mathematical modeling is to approximate a real world system, (i.e., cell), to an extent that the prediction can be tested against observable properties of the system. Therefore, the sophistication of the model is tied to that of techniques to make observations of the system. More sophisticated modeling may lead to the development of better measurement techniques and vice versa. Recently, new technologies that make possible genomic and proteomic profiling of cellular behavior have been developed, providing enormous amounts of information for cellular systems behavior. These genomic and proteomic observations could, and have successfully been used to, identify molecular markers for certain kinds of disease such as cancers. However, the monitoring and modeling of genetic regulatory behaviors of the cell could benefit most from the advancement of technology. Therefore, the need for mathematical models to better describe cellular behavior is critical as measurement technologies advance.

Classification of cancers or other genetic diseases at the molecular level and identifying those genes responsible for different phenotypes (molecular markers), may lead to the discovery of new drug targets. Various pattern classification algorithms have been developed and applied not only for the design of classifiers for cancers but also for the identification of novel genes that can discriminate different types of cancers. This will be extended and studied more extensively in the future for the improvement of integrating biological prior-knowledge with data analysis. The genes identified as molecular markers may also become important clue to the study of genetic regulatory mechanisms.

Learning causative relationships as well as predictive relationships among genes from steady-state observations are of great importance since many of biological measurements are from steady-state. We are exploring diverse mathematical and algorithmic approaches to attack this problem. Understanding a genetic regulatory network will require interdisciplinary collaboration and hopefully lead to the mathematical modeling of the system. Of many proposed approaches, the probabilistic Boolean network is promising for qualitative description of biological systems and can be extended in the perspective of biological context. However, it is also desirable to develop more biologically oriented mathematical models to describe biological systems in a more quantitative manner. This will require collaborative studies among biologists, mathematicians, computer scientists, and/or engineers.

Molecular Diagnostics and Target Validation Division and TGen provides a unique opportunity to achieve my research goals since the division has research units to develop various measurement (CGH and expression arrays) and perturbation (siRNA) technologies, to seek for molecular targets for cancer treatment (melanoma and breast cancers), and to utilize integrated diagnostic tools (Nanobiomics) for clinical uses.

Dr. Kim received his BS degree in Agricultural Engineering at the Seoul National University in Seoul, Korea in 1993. He received his MS in Agricultural Engineering in 1995 also from the Seoul National University. Dr. Kim received a PhD in Electrical Engineering from Texas A&M University in 2001.