Systems biology--biomedical modeling

Abstract

Because of the complexity inherent in biological systems, many researchers frequently rely on a combination of global analysis and computational approaches to gain insight into both (i) how interacting components can produce complex system behaviors, and (ii) how changes in conditions may alter these behaviors. Because the biological details of a particular system are generally not taught along with the quantitative approaches that enable hypothesis generation and analysis of the system, we developed a course at Mount Sinai School of Medicine that introduces first-year graduate students to these computational principles and approaches. We anticipate that such approaches will apply throughout the biomedical sciences and that courses such as the one described here will become a core requirement of many graduate programs in the biological and biomedical sciences.

Fig. 1

Data sets can dictate the computational approaches used in systems biology. (Left) Omics technologies generate extremely large data sets that can be analyzed and organized into networks by using statistical modeling techniques. This strategy can be considered “top-down” modeling. (Right) When high-quality data are available, smaller-scale systems can be represented by dynamical models, and simulations with these models can generate quantitative predictions of system behavior. This strategy is sometimes called “bottom-up” modeling. Both approaches are important in systems biology, and a few cutting-edge studies combine the positive aspects of both.

Fig. 2

Complementary computational approaches used in systems biology. For either the top-down approach (left) or the bottom-up approach (right), the text describes the general strategy of computational studies (middle) and lists important techniques that are taught in the course.