Predicting RNA structure and stability from sequence requires a deep integration of experimental thermodynamics with computational modeling. Research in the Znosko lab has systematically quantified the energetics of fundamental RNA motifs (hairpins, internal loops, mismatches, and non-canonical pairs) using optical melting. These data have expanded and refined nearest-neighbor parameters, improving the accuracy of RNA secondary-structure prediction. Complementary algorithmic development and computational studies have enabled the incorporation of these parameters into predictive tools. Bioinformatic analyses have evaluated sequence–structure relationships across large datasets and identified sequence families that exhibit similar structural features, informing both modeling and design. Together, these efforts bridge experiment and computation to advance a more predictive framework for RNA folding, with implications for understanding natural RNA function and engineering RNA-based systems.