StreAM-[Formula: see text]: algorithms for analyzing coarse grained RNA dynamics based on Markov models of...

Area of Science:

• Computational Biology
• Biophysics
• Bioinformatics

Background:

• RNA molecules possess modular structures, making them suitable for coarse-grained representations based on secondary structure.
• Molecular dynamics simulations generate interaction patterns that can be modeled as adjacency matrices of k-nucleotides.
• Transitions between RNA states can be defined as changes in these adjacency matrices, forming Markovian dynamics.

Purpose of the Study:

• To introduce a novel coarse-grained representation for RNA dynamics using adjacency matrices.
• To develop an efficient algorithm, StreAM, for generating Markov models of RNA dynamics.
• To apply the developed method to a riboswitch design problem.

Main Methods:

• Utilizing adjacency matrices derived from molecular dynamics simulations to represent RNA interactions.
• Defining transitions between states as changes in k-vertex adjacency matrices, forming Markovian dynamics.
• Developing StreAM, a stream-based algorithm to overcome computational demands in generating transition probability matrices.

Main Results:

• StreAM was benchmarked on random and RNA unit sphere dynamic graphs, demonstrating robustness across various parameters.
• The algorithm was applied to molecular dynamics simulations of a synthetic tetracycline-dependent riboswitch interacting with antibiotics.
• StreAM provides insights into nucleotide-based RNA dynamics, outperforming conventional metrics like root-mean square fluctuation.

Conclusions:

• The StreAM algorithm effectively models large-scale simulated and real-world dynamic graphs.
• StreAM offers valuable insights into RNA dynamics at the nucleotide level, complementing traditional methods.
• The findings present significant design opportunities for riboswitches, supported by experimental data.