Finding the optimum activation energy in DNA breathing dynamics: A Simulated Annealing approach

TL;DR

This paper applies Simulated Annealing to optimize activation energy parameters in DNA breathing dynamics, effectively handling noisy data to match observed bubble size distributions.

Contribution

It introduces a novel application of Simulated Annealing for parameter estimation in DNA breathing models, improving robustness against data noise.

Findings

Successfully identified optimal activation energy values

Method overcomes large noise in input data

Matches simulated bubble distributions with experimental data

Abstract

We demonstrate how the stochastic global optimization scheme of Simulated Annealing can be used to evaluate optimum parameters in the problem of DNA breathing dynamics. The breathing dynamics is followed in accordance with the stochastic Gillespie scheme with the denaturation zones in double stranded DNA studied as a single molecule time series. Simulated Annealing is used to find the optimum value of the activation energy for which the equilibrium bubble size distribution matches with a given value. It is demonstrated that the method overcomes even large noise in the input surrogate data.