With advances in biology and medicine, there is a need for new decision support systems that can integrate the knowledge of these domains and enhance the decision making process. Several issues need to be addressed before we can design an intelligent biomedical decision support system. With rapid speed of development and innovation, biomedical information is continuously changing, so systems adaptive to change in knowledge are needed. Furthermore, successful integration of knowledge from experimental data as well as that stored in textual databases is needed. In this paper, we discuss some of the challenges in an adaptive modeling framework for complex systems. We focus on systems biology and discuss the challenges in two aspects—modeling from experimental data and modeling from scientific text articles. Firstly, we focus on learning from experimental data and address why adaptive behaviour is required. Secondly, we discuss the importance of having a general adaptive system that may be able to extract knowledge from text for several domains rather than one specific domain as is done in most of the current state-of-the art systems.