Hugh Nymeyer

The energy landscape theory of protein folding: insights into folding mechanisms and scenarios.

Publisher Summary This chapter presents an overview of the theoretical aspects of the “new view” of protein folding based on the energy landscape theory and the funnel concept. The chapter discusses the lattice and off-lattice models that have played a central role in verifying the general ideas associated with minimal frustration and the protein folding funnel. The chapter describes how a theoretical framework can relate the kinetics of protein folding to thermodynamic quantifies as a function of appropriate reaction coordinates. Using this framework, one then establishes the connection with experiments and detailed all-atom simulations, moving toward a fully quantitative theory for protein folding. The chapter also discusses how the hydrophobic effect is coupled to folding. The chapter focuses the difficulties that have to be overcomed to have a good folding sequence that can be of two different natures: energetic or topologic. Energetic frustration can be reduced with the appropriate design for the protein sequence. Topologic frustration is more complicated since it is a consequence of the polymeric nature of the chain and the shape of the folding motif. The chapter reveals the importance of new generation of experiments that have been devised to probe the early folding events and to explore the details of the landscape of small fast-folding proteins will play in increasing the understanding of this field.

Atomic Simulations of Protein Folding, Using the Replica Exchange Algorithm

Publisher Summary This chapter discusses the atomic simulations of protein folding, using the replica exchange (RE) algorithm. In REs method, several copies or replicas of a system are simulated in parallel, only occasionally exchanging temperatures through a Monte Carlo (MC) move that maintains detailed balance. This algorithm is ideal for a large cluster of poorly communicating processors because temperature exchanges can be relatively infrequent and require little data transfer. It was adapted for use with molecular dynamics and named the replica exchange molecular dynamics (REMD) method. The essence of the REMD method is to use the molecular dynamics to generate a suitable canonical ensemble in each replica rather than MC. REMD normally occurs in coordinate and momentum space instead of just coordinate space. The typical temperature fluctuations and their relation to conformational changes are demonstrated from an actual REMD simulation of a solvated peptide. The use of REMD to study the thermodynamics of helix-coil transitions in peptides that are experimentally characterized is also elaborated.

Quantitative computer simulations of biomolecules: A snapshot

A recent workshop titled “Quantitative Computational Biophysics” at Florida State University provided an overview of the state of the art in quantitative modeling of biomolecular systems. The presentations covered a wide range of interrelated topics, including the development and validation of force fields, the modeling of protein–protein interactions, the sampling of conformational space, and the assessment of equilibration and statistical errors. Substantial progress in all these areas was reported.

Exploring the protein folding funnel landscape

Abstract Energy landscape ideas have proved extremely useful in the understanding of the protein folding process. The central idea is that folding can be described as the movement of an ensemble of conformations on a funnel-like energy landscape, and that one can study folding by analyzing (or measuring) key statistical parameters that characterize this energy landscape. In this work simple lattice models of protein-like heteropolymers are used to identify important statistical parameters and demonstrate their effect on folding behavior. In these models, thermodynamic sampling can accurately characterize the kinetic behavior of the system and can be used with a simple diffusive theory to understand macroscopic folding behavior.

Data-Driven Time-Parallelization in the AFM Simulation of Proteins

Molecular dynamics is a popular technique to simulate the behavior of physical systems, with resolution at the atomic scale. One of its limitations is that an enormous computational effort is required to simulate to realistic time spans. Conventional parallelization strategies have limited effectiveness in dealing with this difficulty. We recently introduced a more scalable approach to parallelization, where data from prior, related, simulations are used to parallelize a simulation in the time domain. We demonstrated its effectiveness in nano-mechanics simulations. In this paper, we develop our approach so that it can be used in a soft-matter application involving the atomic force microscopy simulation of proteins. We obtain an order of magnitude improvement in performance when we combine time parallelization with conventional parallelization. The significance of this work lies in demonstrating the promise of data-driven time parallelization in soft-matter applications, which are more challenging than the hard-matter applications considered earlier.

Data-driven time parallelization

We present a new approach to parallelization of important scientific applications. It is based on the observation that results of prior, related, simulations are often available. We use such data to parallelize the time domain. We demonstrate the effectiveness of our approach in Molecular Dynamics (MD) simulations, which are widely used in nano and nano-bio sciences. An important limitation of MD is that the time-step size is around a femto-second. So a large number of time-steps are required to simulate to realistic time scales. Conventional parallelization is of limited effectiveness here -- the most scalable codes currently are not efficient at granularities finer than several milliseconds per iteration. Using our approach, Carbon Nanotube simulations scale to granularities as fine as around ten microseconds per iteration. We also present results on protein unfolding simulations of AFM pulling, where we obtain additional one order of magnitude scalability over conventional parallelization.