Determination of low-energy structures of a small RNA hairpin using Monte Carlo–based techniques

The energy landscape of RNA is known to be extremely rugged, and hence finding low-energy structures starting from a random structure is a challenging task for any optimization algorithm. In the current work, we have investigated the ability of one Monte Carlo–based optimization algorithm, Temperature Basin Paving, to explore the energy landscape of a small RNA T-loop hairpin. In this method, the history of the simulation is used to increase the probability of states less visited in the simulation. It has been found that using both energy and end-to-end distance as the biasing parameters in the simulation, the partially folded structure of the hairpin starting from random structures could be obtained.