RNA is unique among all biomolecules as it can be both information storing and enzymatic. Indeed, in addition to the genetic code, RNA possesses a second layer of information integrated in its secondary structure that can regulate processes as varied as splicing, localization, translation efficiency and protein binding. However, our current ability to explore the complexity of RNA structure is limited. Prediction algorithms do not account for intra-cellular interactions such as the role played by proteins in RNA folding or dynamic unwinding of structured regions by ribosomes. Other existing assays based on RNA chemical probing (DMS, SHAPE) can explore RNA structure inside the cell but fail to untangle the multiple conformations RNA may assume inside a single cell, leading to a potentially false interpretation of its structure. We developped a method that is capable of detecting alternative RNA structures which form from the same underlying sequence, both in vitro and ex vivo. We applied our approach to HIV-1 and revealed genomic RNA structure heterogeneity with novel functional implications.