A dynamic Monte Carlo algorithm for exploration of dense conformational spaces in heteropolymers

This paper concerns configurational sampling methods for dense single chains for use in the study of low energy states of heteropolymers. The efficiency of current sampling techniques decreases with increasing density and breaks down completely when the volume fraction of the solvent approaches zero. Methods proposed for dense multichain systems are also ineffective for single chain conformational sampling. We propose a new elementary Monte Carlo move for dense single chains which generates new configurations by breaking and patching the chain. The effectiveness of this method is studied by testing its ability to identify global energy minima for maximally compact lattice chains. The algorithm is able to determine the ground state for a HP model protein of length 48 in around 37 000 steps. Computational results are presented for compact lattice chains of length up to 1000 on a 10×10×10 cubic lattice.