forked from ecell/epdp
-
Notifications
You must be signed in to change notification settings - Fork 11
/
Copy pathREADME
136 lines (103 loc) · 5.38 KB
/
README
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
:author: Thomas R. Sokolowski, Koichi Takahashi
=== eGFRD2 Prototype ===
originally based on the E-Cell Particle Dynamics Prototype
Copyright (C) 2009-2017 AMOLF
Copyright (C) 2008-2010 RIKEN
Copyright (C) 2005-2008 The Molecular Sciences Institute
About this package
========================
This package is tentatively named "eGFRD2 Prototype", and was
originally forked from RIKEN's E-Cell Particle Dynamics Prototype (EPDP).
It currently implements the enhanced Greens Function Reaction Dynamics
(eGFRD) algorithm working in all dimensions, the Reaction Brownian
Dynamics (RBD) simulation algorithm, and another BD simulator based on
the Rection Volume Method (rvm-BD). Implementation of the original
version of GFRD will possibly be added in future. The code was designed
and implemented with the hope that it will eventually be part of the
"E-Cell System Version 4" multi-algorithm, multi-space simulation
platform, or its subsequent versions.
The purpose of this prototype code written in mixed Python and C++ is
to establish a solid and practical implementation of the algorithms,
and to extend it into a form that is suitable for large-scale
biochemical and cell simulations.
The eGFRD algorithm first appeared and was used in a paper by
Takahashi, Tanase-Nicola and ten Wolde [1]. A detailed introduction
to eGFRD, and in particular a description of the extensions to lower
dimensions is found in the PhD Thesis of Sokolowski [2]. Review [3]
also contains a good introduction into the current eGFRD version.
Moreover, eGFRD2 is described in detail in a forthcoming paper,
currently available as a preprint on arXiv [4].
The Reaction Brownian Dynamics algorithm is described in a paper by
Morrelli and ten Wolde [5]; the Reaction Volume Method and rvm-BD
are described in detail in the Master's Thesis of Paijmans [6], and
also briefly in [2-4].
Authors
========================
(alphabetical order)
Laurens Bossen
Kazunari Kaizu
Moriyoshi Koizumi
Thomas Miedema
Joris Paijmans
Thomas R. Sokolowski
Koichi Takahashi
Martijn Wehrens
License
========================
This package is distributed under the terms of GNU General Public License
version 2. See COPYING.
Building this package
========================
See INSTALL.
History of the Code
========================
Koichi Takahashi initially stated development of the code in 2005 to
implement his prototype of Greens Function Reaction Dynamics
simulation method invented by Jeroen van Zon and Pieter Rein ten Wolde
in AMOLF, Amsterdam [7]. He gave a brief invited talk about
performance evaluation and applicability of the method to yeast
pheromon response pathway (the Alpha pathway) using the prototype in
the Third Annual Alpha Project Research Symposium (June 16-27, 2005, at
UC Berkeley Art Museum).
Later, in December 2006, ten Wolde, Tanase-Nicola, and Takahashi
introduced the concept called first-passage processes [8,9] to Greens
Function Reaction Dynamics by putting protective domains around
particles to further boost the performance and accuracy of the method.
The new method was named eGFRD (enhanced Greens Function Reaction
Dynamics). By 2010, it was implemented for simulating particle-based
reaction-diffusion systems in 3D, and fully integrated with a BD
fallback system; the framework was applied to study details of
enzymatic reactions in the MAPK push-pull networks [1].
Starting from 2009, Sokolowski, Bossen, Miedema and ten Wolde began
to derive the Green's functions and implementing the new domains
needed for extending eGFRD to lower dimensions. Later Paijmans and
Wehrens joined these efforts. Since extension of RBD to lower dimensions
turned out to be difficult, Paijmans and ten Wolde devised rvm-BD, a new
BD simulation method that preserves detailed balance, based on the
Reaction Volume Method (described in more detail in [6, 2-4]).
All extensions to lower dimensions were assembled together by Bossen
and Sokolowski in 2012-2013. In 2013, Sokolowski created a first working
prototype comprising all new features mentioned above. In the following
years, this prototype was tested and improved, and some further new features,
such as the option of transiting particles between cylinders and planes,
were integrated. The prototype code then was used to simulate an idealized
model of Pom1 gradient formation, as described in [4].
Plans
=========================
Some features planned to be added are:
- outsourcing further core routines from Python to C++
- optimistic discrete-event scheduling for massive parallelization
- connections to non-spatial simulation methods such as ODE and Gillespie methods
- more efficient visualizers / visualization interfaces
References
=========================
[1] K. Takahashi, S. Tanase-Nicola and P.R. ten Wolde, PNAS doi:10.1073/pnas.0906885107 (2010).
[2] T.R. Sokolowski, "A Computational Study of Robust Formation of Spatial Protein Patterns",
PhD Thesis, VU Amsteram, ISBN: 978-90-77209-72-1 (2013).
[3] T.R. Sokolowski and P.R. ten Wolde, arXiv:1705.08669 [q-bio.MN] (2017).
[4] T.R. Sokolowski et al., arXiv:1708.09364 [q-bio.MN] (2017).
[5] M.J. Morelli and P.R. ten Wolde, J. Chem. Phys. 7;129(5):054112 (2008).
[6] J. Paijmans, Master's Thesis, University of Amsterdam (2012).
[7] van Zon and ten Wolde, Phys. Rev. Lett. 94 (2005).
[8] T. Opplestrup, V.V. Bulatov, G.H. Gilmer, M.H. Kalos, and B. Sadigh, Phys. Rev. Lett. 97 (2006).
[9] M.H. Kalos, D. Levesque and L. Verlet, Phys. Rev. A 9 (1974).