Quantifying RNA Folding Equilibria from Unclassified Cryo-EM Particle Images

Cryo-electron microscopy (cryo-EM) has traditionally been used to solve the structures of proteins, but functional RNA structures often exhibit greater flexibility compared to these proteins. The dynamic nature of RNA structures arises from flexible junctions and weak tertiary contacts, leading to a range of possible conformations. However, cryo-EM particle classification analyses often discard unaligned particles, which contain valuable information required to quantify these broad conformational ensembles.

In this study, researchers devised a method to infer compositional and conformational distributions from unclassified particle images by employing Bayesian ensemble reweighting. They applied this approach to the P4-P6 domain of the Tetrahymena thermophila group I intron, a well-known model system that demonstrates magnesium (Mg2+)-dependent RNA folding.

While the stable monomeric and dimeric states were successfully reconstructed with high resolution, the analysis of the prevalent, flexible conformations necessitated the use of Bayesian ensemble inference.

The researchers' method effectively captured the Mg2+-dependent folding transition, demonstrating that cryo-EM can accurately report the abundance of coexisting RNA states, aligning with measurements obtained from bulk solution. These findings highlight the potential for quantitative studies of RNA folding by integrating cryo-EM measurements with statistical ensemble-inference tools.