Conditional Prediction of RNA Secondary Structure Using NMR Chemical Shifts

Inspired by methods that utilize chemical-mapping data to guide secondary structure prediction, we sought to develop a framework for using assigned chemical shift data to guide RNA secondary structure prediction. We first used machine learning to develop classifiers which predict the base-pairing status of individual residues in an RNA based on their assigned chemical shifts. Then, we used these base-pairing status predictions as restraints to guide RNA folding algorithms. Our results showed that we could recover the correct secondary folds for nearly all of the 108 RNAs in our dataset with remarkable accuracy. Finally, we assessed whether we could conditionally predict the structure of the model RNA, microRNA-20b (miR-20b), by folding it using folding restraints derived from chemical shifts associated with two distinct conformational states, one a free (apo) state and the other a protein-bound (holo) state. For this test, we found that by using folding restraints derived from chemical shifts, we could recover the two distinct structures of the miR-20b, confirming our ability to conditionally predict its secondary structure. A command-line tool for Chemical Shifts to Base-Pairing Status (CS2BPS) predictions in RNA has been incorporated into our CS2Structure Git repository and can be accessed via: https://github.com/atfrank/CS2Structure.