Blind tests of RNA nearest-neighbor energy prediction

Abstract

The predictive modeling and design of biologically active RNA molecules requires understanding the energetic balance among their basic components. Rapid developments in computer simulation promise increasingly accurate recovery of RNA's nearest-neighbor (NN) freeenergy parameters, but these methods have not been tested in predictive trials or on nonstandard nucleotides. Here, we present, to our knowledge, the first such tests through a RECCES-Rosetta (reweighting of energy-function collection with conformational ensemble sampling in Rosetta) framework that rigorously models conformational entropy, predicts previously unmeasured NN parameters, and estimates these values' systematic uncertainties. RECCES-Rosetta recovers the 10 NN parameters for Watson-Crick stacked base pairs and 32 single-nucleotide dangling-end parameters with unprecedented accuracies: rmsd of 0.28 kcal/mol and 0.41 kcal/mol, respectively. For setaside test sets, RECCES-Rosetta gives rmsd values of 0.32 kcal/mol on eight stacked pairs involving G-U wobble pairs and 0.99 kcal/mol on seven stacked pairs involving nonstandard isocytidine-isoguanosine pairs. To more rigorously assess RECCES-Rosetta, we carried out four blind predictions for stacked pairs involving 2,6-diaminopurine-U pairs, which achieved 0.64 kcal/mol rmsd accuracy when tested by subsequent experiments. Overall, these results establish that computational methods can now blindly predict energetics of basic RNA motifs, including chemically modified variants, with consistently better than 1 kcal/mol accuracy. Systematic tests indicate that resolving the remaining discrepancies will require energy function improvements beyond simply reweighting component terms, and we propose further blind trials to test such efforts.RNA helix | ensemble prediction | simulated tempering | thermodynamics | blind prediction R NA plays central roles in biological processes, including translation, splicing, regulation of genetic expression, and catalysis (1, 2), and in bioengineering efforts to control these processes (3-5). These critical RNA functions are defined at their most fundamental level by the energetics of how RNA folds and interacts with other RNAs and molecular partners, and how these processes change upon naturally occurring or artificially introduced chemical modifications. Experimentally, the folding free energies of RNA motifs can be precisely measured by optical melting experiments, and a compendium of these measurements have established the nearest-neighbor (NN) model for the most basic RNA elements, including double helices with the four canonical ribonucleotides (6). In the NN model, the stability of a base pair is assumed to only be affected by its adjacent base pairs, and the folding free energy of a canonical RNA helix can be estimated based on NN parameters for each stacked pair, an initialization term for the entropic cost of creating the first base pair, and corrections for different terminal base pairs. Although next-NN effects and tertiary contacts are not treated in the NN mo...

Stability of RNA duplexes containing inosine·cytosine pairs

“… a For each nearest neighbor pair, the top sequence is written 5′ to 3′, and the bottom sequence is written 3′ to 5′. b The number of times that nearest neighbor pair appears in the sequences studied. c Predictions from computer model using the RECCES-Rosetta framework (38). d Difference between the experimentally derived Δ G ° 37 and RECCES-Rosetta model prediction. e Percent difference between the experimentally derived Δ G ° 37 and RECCES-Rosetta model prediction.…”

“…The Das lab recently reported blind tests of a method to computationally predict nearest neighbor energetic parameters (38). They developed a reweighting of energy-function collection with conformational ensemble sampling in a Rosetta (RECCES-Rosetta) framework.…”

“…They developed a reweighting of energy-function collection with conformational ensemble sampling in a Rosetta (RECCES-Rosetta) framework. This purely predictive model factors in separate component terms for hydrogen bonding, electrostatics, van der Waals interactions, base stacking, torsional potentials and an orientation-dependent solvation model (38) and does not require any experimental data. Those authors used this model to blindly predict the nearest neighbor parameters for all eight I·C nearest neighbors as well as a value for a terminal I·C.…”

“…for other RNA motif categories. Canonical pairs had the lowest rmsd value (0.28) while isoG–isoC pairs had the highest rmsd value (0.99), with the average of all motifs being an rmsd value of 0.50 (38). We are hopeful that these new experimental data can be used to refine the RECCES-Rosetta model to make it even more valuable to RNA researchers.…”

RNA-Puzzles Round III: 3D RNA structure prediction of five riboswitches and one ribozyme

“…The repulsion between molecular hydrogens was too weak in Rosetta. The effect was corrected for later targets and in other Rosetta RNA applications (Chou et al 2016). Independently, protein modeling work in Rosetta discovered a similar issue (Park et al 2016).…”