Genome-scale metabolic network models represent the link between the genotype and phenotype of the organism, which are usually reconstructed based on the genome sequence annotation and relevant biochemical and physiological information. These models provide a holistic view of the organism's metabolism, and constraint-based metabolic flux analysis methods have been used extensively to study genome-scale cellular metabolic networks. It is clear that the quality of the metabolic network model determines the outcome of the application. Therefore, it is critically important to determine the accuracy of a genome-scale model in describing the cellular metabolism of the modeled strain. However, because of the model complexity, which results in a system with very high degree of freedom, a good agreement between measured and computed substrate uptake rates and product secretion rates is not sufficient to guarantee the predictive capability of the model. To address this challenge, in this work we presenta novel system identification based framework to extract the qualitative biological knowledge embedded in the quantitative simulation results from the metabolic network models. The extracted knowledge can serve two purposes: model validation during model development phase, which is the focus of this work, and knowledge discovery once the model is validated. This framework bridges the gap between the large amount of numerical results generated from genome-scale models and the knowledge that can be easily understood by biologists. The effectiveness of the proposed framework is demonstrated by its application to the analysis of two recently published genome-scale models of Scheffersomyces stipitis. (C) 2015 Wiley Periodicals, Inc.