NESSie.jl – Efficient and intuitive finite element and boundary element methods for nonlocal protein electrostatics in the Julia language
| Autor: | Andreas Hildebrandt, Sergej Rjasanow, Thomas Kemmer |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
0301 basic medicine
Rapid prototyping General Computer Science business.industry Computer science Computation Usability Python (programming language) Finite element method Theoretical Computer Science NESSIE Computational science 03 medical and health sciences 030104 developmental biology Modeling and Simulation business MATLAB Boundary element method computer computer.programming_language |
| Zdroj: | Journal of Computational Science. 28:193-203 |
| ISSN: | 1877-7503 |
| DOI: | 10.1016/j.jocs.2018.08.008 |
| Popis: | The development of scientific software can be generally characterized by an initial phase of rapid prototyping and the subsequent transition to computationally efficient production code. Unfortunately, most programming languages are not well-suited for both tasks at the same time, commonly resulting in a considerable extension of the development time. The cross-platform and open-source Julia language aims at closing the gap between prototype and production code by providing a usability comparable to Python or MATLAB alongside high-performance capabilities known from C and C++ in a single programming language. In this paper, we present efficient protein electrostatics computations as a showcase example for Julia. More specifically, we present both finite element and boundary element solvers for computing electrostatic potentials of proteins in structured solvents. By modeling the latter in an implicit but nonlocal fashion, we account for correlation of molecular polarization due to the solvent structure around the solute and sustain accuracy without suffering from infeasible runtimes as compared to the explicit case. In this context, we show that our implementation is on par with optimized C code and highlight the components of the implementation that can be transferred to more general tasks. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect