Exploration of Quantitative Scoring Metrics to Compare Systems Biology Modeling Approaches

Abstract

In this paper, we report a focused case study to assess whether quantitative metrics are useful to evaluate molecular-level system biology models on cellular metabolism. Ideally, the bio-modeling community shall be able assess systems biology models based on objective and quantitative metrics. This is because metric-based model design not only can accelerate the validation process, but also can improve the efficacy of model design. In addition, the metric will enable researchers to select models with any desired quality standards to study biological pathway. In this case study, we compare popular systems biology modeling approaches such as Michaelis-Menten kinetics and generalized mass action and flux balance analysis to examine the difficulties in developing quantitative metrics for bio-model assessment. We created a set of guidelines in evaluating the efficacy of various bio-modeling approaches and system analysis in several "bio-systems of interest". We found that quantitative scoring metrics are essential aids for (i) model adopters and users to determine fundamental distinctions among bio-models, and (ii) model developers to improve key areas in bio-modeling. Eventually, we want to extend this evaluation practice to broad systems biology modeling.