Engineering Papers⌕ Search

DOE OSTI · 947873

Developing Improved MD Codes for Understanding Processive Cellulases

Abstract

The mechanism of action of cellulose-degrading enzymes is illuminated through a multidisciplinary collaboration that uses molecular dynamics (MD) simulations and expands the capabilities of MD codes to allow simulations of enzymes and substrates on petascale computational facilities. There is a class of glycoside hydrolase enzymes called cellulases that are thought to decrystallize and processively depolymerize cellulose using biochemical processes that are largely not understood. Understanding the mechanisms involved and improving the efficiency of this hydrolysis process through computational models and protein engineering presents a compelling grand challenge. A detailed understanding of cellulose structure, dynamics and enzyme function at the molecular level is required to direct protein engineers to the right modifications or to understand if natural thermodynamic or kinetic limits are in play. Much can be learned about processivity by conducting carefully designed molecular dynamics (MD) simulations of the binding and catalytic domains of cellulases with various substrate configurations, solvation models and thermodynamic protocols. Most of these numerical experiments, however, will require significant modification of existing code and algorithms in order to efficiently use current (terascale) and future (petascale) hardware to the degree of parallelism necessary to simulate a system of the size proposed here. This work will develop MD codes that can efficiently use terascale and petascale systems, not just for simple classical MD simulations, but also for more advanced methods, including umbrella sampling with complex restraints and reaction coordinates, transition path sampling, steered molecular dynamics, and quantum mechanical/molecular mechanical simulations of systems the size of cellulose degrading enzymes acting on cellulose.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Crowley, Michael F., Uberbacher, E. C., Brooks III, C. L., Walker, R. C., Nimlos, Mark R. (ORCID:000000017117775X), Himmel, Michael E.. 2008-01-01. Developing Improved MD Codes for Understanding Processive Cellulases. https://doi.org/10.1088/1742-6596%2F125%2F1%2F012049. https://engineeringpapers.com/record/osti%3A947873/developing-improved-md-codes-for-understanding-processive-cellulases

Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Engineering Papers by Indiciae: BibTeX · RIS.

KEEP EXPLORING

Related reports

A Framework to Analyze the Performance of Load Balancing Schemes for Ensembles of Stochastic Simulations

Ensembles of simulations are employed to estimate the statistics of possible future states of a system, and are widely used in important applications such as climate change and biological modeling. Ensembles of runs can naturally be executed in parallel. However, when the CPU times of individual simulations vary considerably, a simple strategy of assigning an equal number of tasks per processor can lead to serious work imbalances and low parallel efficiency. This paper presents a new probabilistic framework to analyze the performance of dynamic load balancing algorithms for ensembles of simulations where many tasks are mapped onto each processor, and where the individual compute times vary considerably among tasks. Four load balancing strategies are discussed: most-dividing, all-redistribution, random-polling, and neighbor-redistribution. Simulation results with a stochastic budding yeast cell cycle model are consistent with the theoretical analysis. It is especially significant that there is a provable global decrease in load imbalance for the local rebalancing algorithms due to scalability concerns for the global rebalancing algorithms. The overall simulation time is reduced by up to 25 %, and the total processor idle time by 85 %.

ALGORITHMS↗

Sequential Construction of Costly Networks

Natural disasters or attacks often disrupt infrastructure networks requiring a costly recovery. This motivates an optimization problem where the objecitve is to construct the nodes of a graph G(V;E), and the cost of each node is dependent on the number of its neighbors previously constructed, or more generally, any properties of the previously-completed subgraph. In this optimization problem the objective is to find a permutation of the nodes which results in the least construction cost. We prove that in the case where the cost of nodes is a convex function in the number of neighbors, the optimal construction sequence is to start at a single node and move outwards. We also introduce algorithms and heuristics for solving various instances of the problem. Those methods can be applied to help reduce the cost of recovering from disasters as well as to plan the deployment of new network infrastructure.

ALGORITHMS↗

Visualization and Data Analysis at the Exascale

The scope of our working group is scientific visualization and data analysis. Scientific visualization refers to the process of transforming scientific simulation and experimental data into images to facilitate visual understanding. Data analysis refers to the process of transforming data into an information-rich form via mathematical or computational algorithms to promote better understanding. We share scope on data management with the Storage group. Data management refers to the process of tracking, organizing and enhancing the use of scientific data. The purpose of our work is to enable scientific discovery and understanding. Visualization and data analysis has a broad scope as an integral part of scientific simulations and experiments, it is also a distinct separate service for scientific discovery and documentation purposes. Our scope includes an exascale software and hardware infrastructure that effectively supports visualization and data analysis.

ALGORITHMS↗