|
Title
|
|
|
|
A robust simulator for physiologically structured population models
|
|
|
Author
|
|
|
|
|
|
|
Abstract
|
|
|
|
| A framework to simulate physiologically structured population (PSP) models on high performance compute (HPC) infrastructure is built. Based on the model of a single cell, billions of cells can be simulated in an efficient way, allowing fast simulation of the interaction of an entire organ with other body parts. Trough combination of three state-of-the-art algorithms, the simulation time is decreased with multiple orders of magnitude. First: PSP modelling exploits the fact that a lot of the cells act the same at the same time which results in multiple orders of magnitude speed-up. Secondly, speed-up is achieved by using an unconditionally stable, partial differential equation solver which allows to trade speed for precision and allows big time stepping. Third speed-up is due to the fact that the framework is designed with HPC cluster use in mind. The PSP simulator is mathematically derived to have maximal stability.Simulation results are validated and simulation speed and accuracy are measured. |
|
|
|
Language
|
|
|
|
English
|
|
|
Source (journal)
|
|
|
|
IEEE/ACM transactions on computational biology and bioinformatics / Institute of Electrical and Electronics Engineers [New York, N.Y.] - New York, N.Y.
|
|
|
Publication
|
|
|
|
New York, N.Y.
:
2019
|
|
|
ISSN
|
|
|
|
1545-5963
|
|
|
DOI
|
|
|
|
10.1109/TCBB.2018.2810077
|
|
|
Volume/pages
|
|
|
|
16
:3
(2019)
, p. 851-864
|
|
|
ISI
|
|
|
|
000471070600017
|
|
|
Pubmed ID
|
|
|
|
29994586
|
|
|
Full text (Publisher's DOI)
|
|
|
|
|
|
|
Full text (open access)
|
|
|
|
|
|