Saliency-based analysis can be applied to time-varying 3D datasets purpose of summarization, abstraction, and motion analysis. As the sizes of time-varying datasets continue to grow, it becomes more and more difficult to comprehend vast amounts of data and information in a short period of time. Automatically generated thumbnail images and previewing of time-varying datasets can help viewers explore and understand the datasets significantly faster as well as provide new insights. In this paper, we introduce a novel method for detecting salient frames for molecular dynamics simulations. Our method effectively detects crucial transitions in simulated mechanosensitive ion channel (MscS), in agreement with experimental data.