Validating single crystal models that account for the complex deformation mechanisms in body-centered-cubic (bcc) materials has remained a prominent challenge. Uncertainty quantification (UQ) with the Bayesian statistical setting has been recently finding new avenues for supporting the validation of the bcc single crystal models with high uncertainty in their parameters. In this work, we address the deformation mechanisms of representative bcc molybdenum via UQ processes of two recently developed physics-based single crystal plasticity models: (1) a model with thermally-activated dislocation kinetics with a transition to phonon drag-dominated regime [1] and (2) a model with dislocation dynamics-informed hardening laws [2]. First, a Bayesian model calibration (BMC) approach that employs Markov chain Monte Carlo is utilized to constrain the parameter uncertainty to experimental stress data available at a wide range of strain rates, temperatures, and crystallographic orientations. In addition to the BMC, we perform a variance-based global sensitivity analysis (GSA) to better understand the mechanistic differences between the two single crystal models. The GSA results show apparent mechanistic differences in the dislocation density evolution mechanisms in the two single crystal models while the difference in the total residual output uncertainty between them is found to be insignificant after the BMC. To further elucidate the inelastic deformation mechanisms at mechanical extremes, calibrated models are validated against plate impact experiments. We perform the GSA and analyze how the parameter uncertainty propagates in each of the plate impact simulations to identify critical factors in predicting the complex plate impact responses. Our results show the UQ procedure that combines BMC and GSA enables efficient identification of the complex deformation mechanisms in the bcc single crystals subjected to moderate to extreme loading conditions. This work was funded by the National Research Foundation of Korea (Grant No. RS-2023-00279843).