Elastic network models for RNA: a comparative assessment with molecular dynamics and SHAPE experiments

Abstract

Elastic network models (ENMs) are valuable and efficient tools for characterizing the collective internal dynamics of proteins based on the knowledge of their native structures. The increasing evidence that the biological functionality of RNAs is often linked to their innate internal motions poses the question of whether ENM approaches can be successfully extended to this class of biomolecules. This issue is tackled here by considering various families of elastic networks of increasing complexity applied to a representative set of RNAs. The fluctuations predicted by the alternative ENMs are stringently validated by comparison against extensive molecular dynamics simulations and SHAPE experiments. We find that simulations and experimental data are systematically best reproduced by either an all-atom or a three-beads-per-nucleotide representation (sugar-base-phosphate), with the latter arguably providing the best balance of accuracy and computational complexity.

Combining Simulations and Solution Experiments as a Paradigm for RNA Force Field Refinement

“…[1][2][3][4][5][6][7][8][9] In the case of RNA folding, a few partly successful atomistic simulations have been reported. [10][11][12][13] However, recent extensive simulations of unstructured oligonucleotides for which converged sampling is affordable have unambiguously shown that current force-field parameters are not accurate enough to reproduce solution experiments.…”

Barnaba: Software for Analysis of Nucleic Acids Structures and Trajectories

“…Barnaba contains routines to construct ENM of nucleic acids and proteins, and, as unique feature, makes it possible to calculate 2uctuations between consecutive C2-C2 atoms. In a previous work Pinamonti et al ( 2015 ), we have shown this quantity to correlate with 2exibility measurements performed with selective 2-hydroxyl acylation analyzed by primer extension (SHAPE) experiments Merino et al ( 2005 ). Here, we show an example of ENM analysis on two RNA molecules: the 174-nucleotide sensing domain of the Thermotoga maritima lysine riboswitch (PDB ID: 3DIG), and the Escherichia coli 5S rRNA (PDB ID: 1C2X).…”

“…Execution time for the ENM calculation using sparse matrices (yellow) or dense matrices (red) on a 2.3 GHz Dual-Core Intel Core i5 processor, as a function of the number of residues in the RNA molecule. Results are shown both for sugar-base-phosphate (SBP) ENM (triangles) and all-atom-ENM (AA-ENM) (circles), as defined in Pinamonti et al ( 2015 ). Left panel shows the time for the interaction matrix diagonalization only, right panel shows the total time including the calculation of C2-C2 fluctuations.…”

“…The optimal value of the parameter R c depends on this choice, as described in Ref. Pinamonti et al ( 2015 ).…”

Predicting the Kinetics of RNA Oligonucleotides Using Markov State Models

“…15 We notice that in a previous work we did not find a significant dependence of RNA dynamics on monovalent ion concentration. 61 …”