A Threshold-Minimization Scheme for Exploring the Energy Landscape of Biomolecules: Application to a Cyclic Peptide and a Disaccharide

Abstract

We present a scheme, called the threshold-minimization method, for globally exploring the energy landscapes of small systems of biomolecular interest where typical exploration moves always require a certain degree of subsequent structural relaxation in order to be efficient, e.g., systems containing small or large circular carbon chains such as cyclic peptides or carbohydrates. We show that using this threshold-minimization method we can not only reproduce the global minimum and relevant local minima but also overcome energetic barriers associated with different types of isomerism for the example of a cyclic peptide, cyclo-(Gly)4. We then apply the new method to the disaccharide α-d-glucopyranose-1-2-β-d-fructofuranose, report energetically preferred configurations and barriers to boat-chair isomerization in the glucopyranosyl ring, and discuss the energy landscape.

Exploring energy landscapes at the DFTB quantum level using the threshold algorithm: the case of the anionic metal cluster Au$$_{20}^{-}$$

“…[126] The exploration of the pocket typically takes place via many Monte Carlo random walks starting from the same minimum, but can also be performed using molecular dynamics or combinations thereof. [32,51] This ensemble of random walkers starting from the same point represents a unit of probability that has been placed into the starting minimum. Keeping track of this ensemble of walkers then corresponds to measuring the flow of probability on the energy landscape as function of threshold energy.…”

“…One class of approaches that systematically determine such entropic barriers are the so-called lid or threshold algorithms, [3,13,32,[49][50][51] where increasingly large pockets of the energy landscape that are accessible below a sequence of energy lids or thresholds are systematically explored and the probability flows between the regions that are connected below a given lid are measured as function of threshold energy. [52,53] Typically, one speaks of a lid algorithm, if the under-lying state space is discrete, e.g.…”

Prediction and clarification of structures of (bio)molecules on surfaces

“…Other methods suitable for biomolecules, in vacuum or on surfaces, are e.g. the recently proposed so-called threshold-minimization algorithm for the investigation of flexible molecules [153], and a variety of genetic or evolutionary algorithms [154][155][156], and numerous hybrid approaches combining elements of several methods.…”