Computational approaches for RNA energy parameter estimation

Abstract

Methods for efficient and accurate prediction of RNA structure are increasingly valuable, given the current rapid advances in understanding the diverse functions of RNA molecules in the cell. To enhance the accuracy of secondary structure predictions, we developed and refined optimization techniques for the estimation of energy parameters. We build on two previous approaches to RNA free-energy parameter estimation: (1) the Constraint Generation (CG) method, which iteratively generates constraints that enforce known structures to have energies lower than other structures for the same molecule; and (2) the Boltzmann Likelihood (BL) method, which infers a set of RNA free-energy parameters that maximize the conditional likelihood of a set of reference RNA structures. Here, we extend these approaches in two main ways: We propose (1) a max-margin extension of CG, and (2) a novel linear Gaussian Bayesian network that models feature relationships, which effectively makes use of sparse data by sharing statistical strength between parameters. We obtain significant improvements in the accuracy of RNA minimum free-energy pseudoknot-free secondary structure prediction when measured on a comprehensive set of 2518 RNA molecules with reference structures. Our parameters can be used in conjunction with software that predicts RNA secondary structures, RNA hybridization, or ensembles of structures. Our data, software, results, and parameter sets in various formats are freely available at http://www.cs.ubc.ca/labs/beta/Projects/RNA-Params.

StreAM- $$T_g$$ : Algorithms for Analyzing Coarse Grained RNA Dynamics Based on Markov Models of Connectivity-Graphs

“…We reduce these simulated trajectories to nucleotides represented by their (C3 ′ ) atoms. From there, we represent RNA structure as an undirected graph [20] using each C3 ′ as a vertex and distance dependent interactions as edges [3]. It is well known that nucleotide-based molecular interactions take place between more than one partner [21].…”

“…2 The final implementation StreAM-T g is integrated in a Julia repository. 3 We created plots using the AssayToolbox library for R [39,40]. We generate all random graphs using a generator for dynamic graphs 4 derived for vertex combination.…”

“…In order to perform benchmarks with increasing k, we chose randomly nodes k ∈ [3,10] and repeated this 500 times for different numbers of snapshots (every 10,000 steps). We determined the slope (speed frames ms ) of compute time vs. k for random and MD graphs with different parameters (Table 1).…”

“…So far, available models and simulation tools to design and analyze functionally complex RNA based devices are very limited [2]. Although several tools are available to assess secondary as well as tertiary RNA structure [3], current capabilities to simulate dynamics are still underdeveloped [4] and rely heavily on atomistic molecular dynamics (MD) techniques [5]. RNA structure is largely modular and composed of repetitive motifs [4] that form structural elements such as hairpins and stems based on hydrogen-bonding patterns [6].…”

Exact Learning of RNA Energy Parameters from Structure

“…Recently, massively feature-rich models empowered by parameter estimation algorithms have been proposed. Despite significant progress in the last three decades, made possible by the work of Turner and others [20] on measuring RNA thermodynamic energy parameters and the work of several groups on novel algorithms [21,22,23,24,25,26,27,28] and machine learning approaches [29,30,31], the RNA structure prediction accuracy has not reached a satisfactory level yet [32].…”