Sensitivity analysis is a cornerstone of scientific modeling and engineering design, offering systematic ways to quantify how variations in model inputs influence outputs. It plays a crucial role in improving model understanding, guiding data collection, enhancing robustness, and supporting decision-making under uncertainty. In the current era of digital twins and scientific machine learning, the importance of sensitivity analysis has become even more pronounced. As models become increasingly complex and data-driven, the analysis offers a principled framework for combining data with mechanistic models, tackling challenges such as identifiability, model selection, and interpretability. Different methods enable rigorous assessment of how data inform model predictions, helping to bridge the gap between theory and experiment while promoting transparency and reliability in predictive modeling. In this mini-symposium, we aim to showcase recent advances and applications of (global) sensitivity analysis in science and engineering. We welcome contributions ranging from theoretical developments and computational methods to applied case studies. In particular, we highlight (without restricting the scope to) three emerging focus areas for sensitivity analysis: 1. Sensitivity analysis with given data. 2. Identifiability and parameter estimation. 3. Explainability in scientific machine learning. We invite researchers from academia and industry to contribute to this discussion, fostering dialogue between developers and practitioners on the evolving role of sensitivity analysis in contemporary computational challenges.