PHARMACOGENOMICS
Modeling Genomics and Proteomics Time Series
Despite the availability of a variety of alternative paradigms, compartmental modeling is the dominant approach for understanding the dynamics of drug concentration and effect in the pharmaceutical and pharmacological sciences. This paradigm offers a rich body of behaviors for modeling the effects of drugs and models and model equations can be systematically built from intuitive compartmental elements. A good pharmacokinetic/pharmacodynamic (PK/PD) model could deliver the several advantages, including but not limited to:

• A parsimonious, reduced dimensionality representation of the original data
• Insights into the potential mechanisms or means by which input stimuli are transformed into the time courses of output
• The model parameters convey quantitative information that can be used in simulations and what-if analysis or combined with other methodologies already in use for array data analysis, e.g., the parameters of a model could provide a basis for clustering.

However, the PK/PD modeling community is still struggling to handle gene expression data effectively for a variety of reasons. This is because compartmental approaches require very high levels of supervision and the system identification can be very time-intensive even for data sets containing only a limited number of PD endpoints. State-of-the art PK/PD system modeling software packages such as ADAPT, SAAM II and WinNonLin (to name a few) also require high levels of user intervention and the fitting of thousands of mRNA expression profiles to even a limited set of model types, while technically feasible, is inconvenient enough to be impractical.

Murali Ramanathan, PhD

Associate Professor of
Pharmaceutical Sciences
and Neurology

543 Cooke Hall
Dept. of Pharmaceutical Sciences
State University of New York
Buffalo, NY 14260

Ph:  (716)645-2842 ext 242
Fax: (716)645-3693

E-mail: murali@buffalo.edu

Publications