Studying the Properties of Cellular Materials with GPU Acceleration

Cell Biology;Computational Biology;Molecular Dynamics;Development;Soft Matter;Biophysics;Chemistry (Nanotechnology)

Madhikar, Pranav

University of Waterloo

关键词: Cell Biology;    Computational Biology;    Molecular Dynamics;    Development;    Soft Matter;    Biophysics;    Chemistry (Nanotechnology);   

Others  :  https://uwspace.uwaterloo.ca/bitstream/10012/9327/3/Madhikar_Pranav.pdf

瑞士|英语

来源: UWSPACE Waterloo Institutional Repository

【 摘 要 】

There has always been a great interest in cellular behaviour. From themolecular level, studying the chemistry of the reactions that occur incell, and the physical interactions between those molecules, to thescale of the cell itself and its behaviour in response to variousphenomena. Suffice it to say, that cellular behaviour is highlycomplex and, therefore, it is difficult to predict how cells willbehave or to even describe their behaviour in detail. Traditionallycell biology has been done solely in the laboratory. That has alwaysyielded interesting results and science. There are some aspects ofphenomena that, due to cost, time, or other factors, need to bestudied computationally. Especially if these stimuli occur on veryshort or long time scales. Therefore, a number of models have beenproposed in order to study cell behaviour.Unfortunately, these methods can only be used in certain situationsand circumstances. These methods can, and do, produce interesting andvalid results. Yet there is not really any model available that can beused to model more than one or two kinds of cell behaviour. Forexample, methods that can show cell sorting do not necessarily showpacking. Furthermore, many of the models in the literature representcells as collections of points, or polygons, so cellular interactionsat interfaces cannot be studied efficiently. The goal of the workpresented here was to develop a three dimensional model of cells usingMolecular Dynamics. Cells are represented as spherical meshes of masspoints. And these mass points are placed in a force field thatemulates cellular interactions such as adhesion, repulsion, andfriction. The results of this work indicates that the model developedcan reproduce qualitatively valid cellular behaviour. And the modelcan be extended to include other effects.It must also be recognized that gls{md} is very expensivecomputationally. Especially in the case of this model as many masspoints are needed in the cellular mesh to ensure adequate spatialresolution. Higher performance is always needed either to study largersystems or to iterate on smaller systems more quickly. The mostobvious way to alleviate this problem is too use high performancehardware. It will be shown that this performance is most accessible,after some effort, with gls{gpu} acceleration. The model developed inthis work will be implemented with gls{gpu} acceleration. Thecode generated in this way is quite fast.