Choosing weights for simulated tempering
Sanghyun Park, Vijay S. Pande
Abstract
Sanghyun Park, Vijay S. Pande
Abstract
Simulated tempering is a method to enhance simulations of complex systems by periodically raising and lowering the temperature. Despite its advantages, simulated tempering has been overshadowed by its parallel counterpart, replica exchange (also known as parallel tempering), due to the difficulty of weight determination in simulated tempering. Here we propose a simple and fast method to obtain near-optimal weights for simulated tempering, and demonstrate its effectiveness in a molecular dynamics simulation of Ala(10) polypeptide in explicit solvent. We believe simulated tempering now deserves another look.
OpenAlex reports 94 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Simulated tempering is a method to enhance simulations of complex systems by periodically raising and lowering the temperature. Despite its advantages, simulated tempering has been overshadowed by its parallel counterpart, replica exchange (also known as parallel tempering), due to the difficulty of weight determination in simulated tempering. Here we propose a simple and fast method to obtain near-optimal weights for simulated tempering, and demonstrate its effectiveness in a molecular dynamics simulation of Ala(10) polypeptide in explicit solvent. We believe simulated tempering now deserves another look.
Key concepts: Tempering, Parallel tempering, Replica, Molecular dynamics, Computer science, Simple (philosophy), Raising (metalworking), Statistical physics