Uniform generation of RNA pseudoknot structures with genus filtration

TL;DR

This paper introduces a linear-time uniform sampling algorithm for RNA pseudoknot structures of fixed topological genus, enabling efficient generation and energy-weighted sampling of complex RNA structures.

Contribution

It presents the first linear-time algorithms for uniform and energy-weighted sampling of RNA structures with fixed genus, advancing computational RNA modeling.

Findings

Linear-time uniform sampling algorithm for fixed genus RNA structures

Efficient energy-weighted sampling with a single partition function computation

Applicable to arbitrary topological genus g > 0

Abstract

In this paper we present a sampling framework for RNA structures of fixed topological genus. We introduce a novel, linear time, uniform sampling algorithm for RNA structures of fixed topological genus $g$, for arbitrary $g>0$. Furthermore we develop a linear time sampling algorithm for RNA structures of fixed topological genus $g$ that are weighted by a simplified, loop-based energy functional. For this process the partition function of the energy functional has to be computed once, which has $O(n^2)$ time complexity.