Large-eddy simulation (LES) has gained significant importance as a high-fidelity technique for the numerical resolution of complex turbulent flow. The low-pass filtering of the conservation equations significantly reduces the computational cost of solving turbulence, however, at the expense of modeling the subgrid-scale (SGS) physics. In consequence, the assumptions introduced in the closure formulations may result in potential sources of structural uncertainty that can affect the quantities of interest (QoI), especially in multi-physics phenomena, e.g., combustion processes and interfacial flow, where the small-scale dynamics are crucial to the development of the large scale flow. Therefore, in order to facilitate the analysis, the aim of this work is to characterize SGS model-form uncertainty and estimate the impact on the QoIs by means of eigenspace-based, controlled perturbations within plausible physical bounds [Jofre et al. Flow Turbulence Combust (2018) 100:341]. In the presentation, the strategy will be described in detail and investigations based on LES of canonical turbulent flows will be discussed.