ViewVC Help
View File | Revision Log | Show Annotations | View Changeset | Root Listing
root/group/trunk/matt_papers/canidacy_paper/canidacy_paper.tex
Revision: 106
Committed: Wed Sep 11 03:26:01 2002 UTC (21 years, 9 months ago) by mmeineke
Content type: application/x-tex
File size: 15803 byte(s)
Log Message:
revisions to the first draft halfway complete.

completed the introduction.

File Contents

# User Rev Content
1 mmeineke 95 \documentclass[11pt]{article}
2    
3     \usepackage{graphicx}
4 mmeineke 106 \usepackage{color}
5 mmeineke 98 \usepackage{floatflt}
6 mmeineke 95 \usepackage{amsmath}
7     \usepackage{amssymb}
8     \usepackage[ref]{overcite}
9    
10    
11    
12     \pagestyle{plain}
13     \pagenumbering{arabic}
14     \oddsidemargin 0.0cm \evensidemargin 0.0cm
15     \topmargin -21pt \headsep 10pt
16     \textheight 9.0in \textwidth 6.5in
17     \brokenpenalty=10000
18     \renewcommand{\baselinestretch}{1.2}
19     \renewcommand\citemid{\ } % no comma in optional reference note
20    
21    
22     \begin{document}
23    
24 mmeineke 98
25 mmeineke 95 \title{A Mesoscale Model for Phospholipid Simulations}
26    
27     \author{Matthew A. Meineke\\
28     Department of Chemistry and Biochemistry\\
29     University of Notre Dame\\
30     Notre Dame, Indiana 46556}
31    
32     \date{\today}
33     \maketitle
34    
35     \section{Background and Research Goals}
36    
37 mmeineke 106 Simulations of phospholipid bilayers are, by necessity, quite
38     complex. The lipid molecules are large molecules containing many
39     atoms,and the head group of the lipid will typically contain charge
40     separated ions which set up a large dipole within the molecule. Adding
41     to the complexity are the number of water molecules needed to properly
42     solvate the lipid bilayer. Because of these factors, many current
43     simulations are limited in both length and time scale due to to the
44     sheer number of calculations performed at every time step and the
45     lifetime of the researcher. A typical
46     simulation\cite{saiz02,lindahl00,venable00,Marrink01} will have around
47     64 phospholipids forming a bilayer approximately 40~$\mbox{\AA}$ by
48     50~$\mbox{\AA}$ with roughly 25 waters for every lipid. This means
49     there are on the order of 8,000 atoms needed to simulate these systems
50     and the trajectories in turn are integrated for times up to 10 ns.
51    
52     These limitations make it difficult to study certain biologically
53     interesting phenomena that don't fit within the short time and length
54     scale requirements. One such phenomena is the existence of the ripple
55     phase ($P_{\beta'}$) of the bilayer between the gel phase
56     ($L_{\beta'}$) and the fluid phase ($L_{\alpha}$). The $P_{\beta'}$
57     phase has been shown to have a ripple period of
58     100-200~$\mbox{\AA}$.\cite{katsaras00,sengupta00} A simulation of this
59     length scale would require approximately 1,300 lipid molecules and
60     roughly 25 waters for every lipid to fully solvate the bilayer. With
61     the large number of atoms involved in a simulation of this magnitude,
62     steps \emph{must} be taken to simplify the system to the point where
63     these numbers are reasonable.
64    
65     Another system of interest would be drug molecule diffusion through
66     the membrane. Due to the fluid like properties of a lipid membrane,
67     not all diffusion takes place at membrane channels. It is of interest
68     to study certain molecules that may incorporate themselves directly
69     into the membrane. These molecules may then have an appreciable
70     waiting time (on the order of nanoseconds) within the
71     bilayer. Simulation of such a long time scale again requires
72     simplification of the system in order to lower the number of
73     calculations needed at each time step.
74    
75    
76 mmeineke 95 \section{Methodology}
77    
78 mmeineke 96 \subsection{Length and Time Scale Simplifications}
79 mmeineke 95
80 mmeineke 106 The length scale simplifications are aimed at increasing the number of
81     molecules that can be simulated without drastically increasing the
82 mmeineke 95 computational cost of the system. This is done by a combination of
83     substituting less expensive interactions for expensive ones and
84     decreasing the number of interaction sites per molecule. Namely,
85 mmeineke 106 point charge distributions are replaced with dipoles, and unified atoms are
86     used in place of water, phospholipid head groups, and alkyl groups.
87 mmeineke 95
88     The replacement of charge distributions with dipoles allows us to
89 mmeineke 106 replace an interaction that has a relatively long range,
90     $(\frac{1}{r})$, for the coulomb potential, with that of a relatively
91     short range, $(\frac{1}{r^{3}})$, for dipole - dipole
92     potentials. Combined with a computational simplification algorithm
93     such as a Verlet neighbor list,\cite{allen87:csl} this should give
94     computational scaling by $N$. This is in comparison to the Ewald
95     sum\cite{leach01:mm} needed to compute the coulomb interactions, which
96     scales at best by $N \ln N$.
97 mmeineke 95
98 mmeineke 106 The second step taken to simplify the number of calculations is to
99 mmeineke 95 incorporate unified models for groups of atoms. In the case of water,
100     we use the soft sticky dipole (SSD) model developed by
101     Ichiye\cite{Liu96} (Section~\ref{sec:ssdModel}). For the phospholipids, a
102     unified head atom with a dipole will replace the atoms in the head
103     group, while unified $\text{CH}_2$ and $\text{CH}_3$ atoms will
104 mmeineke 106 replace the alkyl units in the tails (Section~\ref{sec:lipidModel}).
105 mmeineke 95
106 mmeineke 106 The time scale simplifications are introduced so that we can take
107     longer time steps. By increasing the size of the time steps taken by
108     the simulation, we are able to integrate the simulation trajectory
109     with fewer calculations. However, care must be taken that any
110     simplifications used, still conserve the total energy of the
111     simulation. In practice, this means taking steps small enough to
112     resolve all motion in the system without accidently moving an object
113     too far along a repulsive energy surface before it feels the effect of
114     the surface.
115 mmeineke 95
116 mmeineke 96 In our simulation we have chosen to constrain all bonds to be of fixed
117     length. This means the bonds are no longer allowed to vibrate about
118 mmeineke 106 their equilibrium positions. Bond vibrations are typically the fastest
119     periodic motion in a dynamics simulation. By taking this action, we
120     are able to take time steps of 3 fs while still maintaining constant
121     energy. This is in contrast to the 1 fs time step typically needed to
122     conserve energy when bonds lengths are allowed to oscillate.
123 mmeineke 95
124     \subsection{The Soft Sticky Water Model}
125     \label{sec:ssdModel}
126    
127 mmeineke 106 %\begin{floatingfigure}{55mm}
128     %\includegraphics[width=45mm]{ssd.epsi}
129     %\caption{The SSD model with the oxygen and hydrogen atoms drawn in for reference. \vspace{5mm}}
130     %\label{fig:ssdModel}
131     %\end{floatingfigure}
132 mmeineke 96
133 mmeineke 106 The water model used in our simulations is a modified soft
134     Stockmayer-sphere model.\cite{stevens95} Like the Stockmayer-sphere, the SSD
135 mmeineke 98 model\cite{Liu96} consists of a Lennard-Jones interaction site and a
136     dipole both located at the water's center of mass (Figure
137     \ref{fig:ssdModel}). However, the SSD model extends this by adding a
138     tetrahedral potential to correct for hydrogen bonding.
139    
140 mmeineke 106 The SSD water potential is then given by the following equation:
141 mmeineke 96 \begin{equation}
142 mmeineke 106 V_{\text{SSD}} = V_{\text{LJ}}(r_{i\!j}) + V_{\text{dp}}(\mathbf{r}_{i\!j},
143 mmeineke 96 \boldsymbol{\Omega}_{i},\boldsymbol{\Omega}_{j})
144     + V_{\text{sp}}(\mathbf{r}_{i\!j},\boldsymbol{\Omega}_{i},
145     \boldsymbol{\Omega}_{j})
146 mmeineke 98 \label{eq:ssdTotPot}
147 mmeineke 96 \end{equation}
148 mmeineke 106 $V_{\text{LJ}}$ is the Lennard-Jones potential:
149     \begin{equation}
150     V_{\text{LJ}} =
151     4\epsilon_{ij} \biggl[
152     \biggl(\frac{\sigma_{ij}}{r_{ij}}\biggr)^{12}
153     - \biggl(\frac{\sigma_{ij}}{r_{ij}}\biggr)^{6}
154     \biggr]
155     \label{eq:lennardJonesPot}
156     \end{equation}
157     where $r_{ij}$ is the distance between two $ij$ pairs, $\sigma_{ij}$
158     scales the length of the iteraction, and $\epsilon_{ij}$ scales the
159     energy of the potential. For SSD, $\sigma_{\text{SSD}} = 3.051 \mbox{
160     \AA}$ and $\epsilon_{\text{SSD}} = 0.152\text{ kcal/mol}$.
161     $V_{\text{dp}}$ is the dipole potential:
162     \begin{equation}
163     V_{\text{dp}}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},
164     \boldsymbol{\Omega}_{j}) = \frac{1}{4\pi\epsilon_{0}} \biggl[
165     \frac{\boldsymbol{\mu}_{i} \cdot \boldsymbol{\mu}_{j}}{r^{3}_{ij}}
166     -
167     \frac{3(\boldsymbol{\mu}_i \cdot \mathbf{r}_{ij}) %
168     (\boldsymbol{\mu}_j \cdot \mathbf{r}_{ij}) }
169     {r^{5}_{ij}} \biggr]
170     \label{eq:dipolePot}
171     \end{equation}
172     where $\mathbf{r}_{ij}$ is the vector between $i$ and $j$,
173     $\boldsymbol{\Omega}$ is the orientation of the species, and
174     $\boldsymbol{\mu}$ is the dipole vector. The SSD model specifies a dipole
175     magnitude of 2.35~D for water.
176 mmeineke 96
177 mmeineke 101 The hydrogen bonding of the model is governed by the $V_{\text{sp}}$
178 mmeineke 106 term of the potential. Its form is as follows:
179 mmeineke 96 \begin{equation}
180     V_{\text{sp}}(\mathbf{r}_{i\!j},\boldsymbol{\Omega}_{i},
181     \boldsymbol{\Omega}_{j}) =
182     v^{\circ}[s(r_{ij})w_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},
183     \boldsymbol{\Omega}_{j})
184     +
185     s'(r_{ij})w^{x}_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},
186     \boldsymbol{\Omega}_{j})]
187 mmeineke 98 \label{eq:spPot}
188 mmeineke 96 \end{equation}
189 mmeineke 106 Where $v^\circ$ scales strength of the interaction.
190 mmeineke 98 $w_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},\boldsymbol{\Omega}_{j})$
191     and
192     $w^{x}_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},\boldsymbol{\Omega}_{j})$
193     are responsible for the tetrahedral potential and a correction to the
194     tetrahedral potential respectively. They are,
195 mmeineke 96 \begin{equation}
196     w_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},\boldsymbol{\Omega}_{j}) =
197     \sin\theta_{ij} \sin 2\theta_{ij} \cos 2\phi_{ij}
198     + \sin \theta_{ji} \sin 2\theta_{ji} \cos 2\phi_{ji}
199 mmeineke 106 \label{eq:spPot2}
200 mmeineke 96 \end{equation}
201 mmeineke 98 and
202 mmeineke 96 \begin{equation}
203     \begin{split}
204 mmeineke 98 w^{x}_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},\boldsymbol{\Omega}_{j}) =
205     &(\cos\theta_{ij}-0.6)^2(\cos\theta_{ij} + 0.8)^2 \\
206     &+ (\cos\theta_{ji}-0.6)^2(\cos\theta_{ji} + 0.8)^2 - 2w^{\circ}
207 mmeineke 96 \end{split}
208 mmeineke 98 \label{eq:spCorrection}
209 mmeineke 96 \end{equation}
210 mmeineke 106 The angles $\theta_{ij}$ and $\phi_{ij}$ are defined by the spherical
211     polar coordinates of the position of sphere $j$ in the reference frame
212     fixed on sphere $i$ with the z-axis aligned with the dipole moment.
213     The correction
214 mmeineke 98 $w^{x}_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},\boldsymbol{\Omega}_{j})$
215     is needed because
216     $w_{ij}(\mathbf{r}_{ij},\boldsymbol{\Omega}_{i},\boldsymbol{\Omega}_{j})$
217 mmeineke 106 vanishes when $\theta_{ij}$ is $0^\circ$ or $180^\circ$.
218 mmeineke 96
219 mmeineke 106 Finally, the sticky potential is scaled by a cutoff function,
220 mmeineke 101 $s(r_{ij})$ that scales smoothly between 0 and 1. It is represented
221     by:
222 mmeineke 96 \begin{equation}
223     s(r_{ij}) =
224     \begin{cases}
225     1& \text{if $r_{ij} < r_{L}$}, \\
226     \frac{(r_{U} - r_{ij})^2 (r_{U} + 2r_{ij}
227     - 3r_{L})}{(r_{U}-r_{L})^3}&
228     \text{if $r_{L} \leq r_{ij} \leq r_{U}$},\\
229     0& \text{if $r_{ij} \geq r_{U}$}.
230     \end{cases}
231 mmeineke 98 \label{eq:spCutoff}
232 mmeineke 96 \end{equation}
233    
234 mmeineke 106 Despite the apparent complexity of Equation \ref{eq:spPot}, the SSD
235     model is still computationaly inexpensive. This is due to Equation
236     \ref{eq:spCutoff}. With $r_{L}$ being 2.75~$\mbox\AA}$ and $r_{U}$
237     being equal to either 3.35~$\mbox{\AA}$ for $s(r_{ij})$ or
238     4.0~$\mbox{\AA}$ for $s'(r_{ij})$, the sticky potential is only active
239     over an extremly short range, and then only with other SSD
240     molecules. Therefore, it's predominant interaction is through it's
241     point dipole and Lennard-Jones sphere.
242 mmeineke 98
243 mmeineke 95 \subsection{The Phospholipid Model}
244     \label{sec:lipidModel}
245    
246 mmeineke 99 \begin{floatingfigure}{90mm}
247     \includegraphics[angle=-90,width=80mm]{lipidModel.epsi}
248     \caption{A representation of the lipid model. $\phi$ is the torsion angle, $\theta$ is the bend angle, $\mu$ is the dipole moment of the head group, and n is the chain length. \vspace{5mm}}
249     \label{fig:lipidModel}
250     \end{floatingfigure}
251 mmeineke 95
252 mmeineke 99 The lipid molecules in our simulations are unified atom models. Figure
253     \ref{fig:lipidModel} shows a template drawing for one of our
254     lipids. The Head group of the phospholipid is replaced by a single
255     Lennard-Jones sphere with a freely oriented dipole placed at it's
256     center. The magnitude of it's dipole moment is 20.6 D. The tail atoms
257 mmeineke 106 are unified $\text{CH}_2$ and $\text{CH}_3$ atoms and are also modeled
258 mmeineke 99 as Lennard-Jones spheres. The total potential for the lipid is
259     represented by Equation \ref{eq:lipidModelPot}.
260    
261     \begin{equation}
262     V_{\mbox{lipid}} = \overbrace{%
263     V_{\text{bend}}(\theta_{ijk}) +%
264     V_{\text{tors.}}(\phi_{ijkl})}^{bonded}
265     + \overbrace{%
266     V_{\text{LJ}}(r_{i\!j}) +
267     V_{\text{dp}}(r_{i\!j},\Omega_{i},\Omega_{j})%
268     }^{non-bonded}
269     \label{eq:lipidModelPot}
270     \end{equation}
271    
272 mmeineke 106 The non-bonded interactions, $V_{\text{LJ}}$ and $V_{\text{dp}}$, are
273 mmeineke 99 the Lennard-Jones and dipole-dipole interactions respectively. For the
274     non-bonded potentials, only the bend and the torsional potentials are
275     calculated. The bond potential is not calculated, and the bond lengths
276     are constrained via RATTLE.\cite{leach01:mm} The bend potential is of
277     the form:
278     \begin{equation}
279     V_{\text{bend}}(\theta_{ijk}) = k_{\theta}\frac{(\theta_{ijk} - \theta_0)^2}{2}
280     \label{eq:bendPot}
281     \end{equation}
282 mmeineke 100 Where $k_{\theta}$ sets the stiffness of the bend potential, and $\theta_0$
283     sets the equilibrium bend angle. The torsion potential was given by:
284     \begin{equation}
285 mmeineke 106 V_{\text{tors.}}(\phi_{ijkl})= c_1[1+\cos\phi_{ijkl}]
286     + c_2 [1 - \cos(2\phi_{ijkl})] + c_3[1 + \cos(3\phi_{ijkl})]
287 mmeineke 100 \label{eq:torsPot}
288     \end{equation}
289 mmeineke 102 All parameters for bonded and non-bonded potentials in the tail atoms
290     were taken from TraPPE.\cite{Siepmann1998} The bonded interactions for
291     the head atom were also taken from TraPPE, however it's dipole moment
292 mmeineke 106 and mass were based on the properties of the phosphatidylcholine head
293     group. The Lennard-Jones parameter for the head group was chosen such
294     that it was roughly twice the size of a $\text{CH}_3$ atom, and it's
295     well depth was set to be approximately equal to that of $\text{CH}_3$.
296 mmeineke 99
297 mmeineke 102 \section{Initial Simulation: 25 lipids in water}
298     \label{sec:5x5}
299 mmeineke 101
300 mmeineke 106 \subsection{Starting Configuration and Parameters}
301 mmeineke 102 \label{sec:5x5Start}
302 mmeineke 101
303 mmeineke 102 Our first simulation was an array of 25 single chained lipids in a sea
304     of water (Figure \ref{fig:5x5Start}). The total number of water
305     molecules was 1386, giving a final of water concentration of 70\%
306     wt. The simulation box measured 34.5~$\mbox{\AA}$ x 39.4~$\mbox{\AA}$
307 mmeineke 106 x 39.4~$\mbox{\AA}$ with periodic boundary conditions invoked. The
308 mmeineke 102 system was simulated in the micro-canonical (NVE) ensemble with an
309     average temperature of 300~K.
310    
311     \subsection{Results}
312     \label{sec:5x5Results}
313    
314     Figure \ref{fig:5x5Final} shows a snapshot of the system at
315 mmeineke 106 3.6~ns. Here it is exciting to note that the system has spontaneously
316     self assembled into a bilayer. Discussion of the length scales of the
317     bilayer will follow in this section. However, it is interesting to
318     note several qualitative properties of the system revealed by this
319     snapshot. First, is how the tail chains are tilted to the bilayer
320     normal. This is usually indicative of the gel ($L_{\beta'}$)
321     phase. Likely, the box size is too small for the bilayer to relax to
322     the fluid ($P_{\alpha}$ phase. Showing the need for a constant
323     pressure simulation versus the constant volume of the current
324     ensemble.
325 mmeineke 102
326 mmeineke 106 The trajectory of the simulation was analyzed using the pair-wise
327     radial distribution function, $g(r)$, which has the form:
328     \begin{equation}
329     g(r) = \frac{V}{N_{\text{pairs}}}\langle \sum_{i} \sum_{j \neq i}
330     \delta(|\mathbf{r} - \mathbf{r}_{ij}|) \rangle
331     \label{eq:gofr}
332     \end{equation}
333     Equation \ref{eq:gofr} gives us information about the spacing of two
334     species as a function of radius. Essentially, if the observer were
335     located at atom $i$ and were looking out in all directions, $g(r)$
336     shows the relative density of atom $j$ at any given radius, $r$,
337     normalized by the expected density of atom $j$ at $r$. In a
338     homogeneously mixed fluid, $g(r)$ will approach 1 at large $r$, as a
339     fluid contains no long range structure to contribute peaks in the
340     number density.
341 mmeineke 102
342 mmeineke 106 For the species containing dipoles, a second pair wise distribution
343     function was used, $g_{\gamma}(r)$. It is of the form:
344     \begin{equation}
345     g_{\gamma}(r) = foobar
346     \label{eq:gammaofr}
347     \end{equation}
348     Where $\gamma_{ij}$ is the angle between the dipole of atom $j$ with
349     respect to the dipole of atom $i$. This correlation will vary between
350     $+1$ and $-1$ when the two dipoles are perfectly aligned and
351     anti-aligned respectively. This then gives us information about how
352     directional species are aligned with each other as a function of
353     distance.
354 mmeineke 102
355 mmeineke 106 Figure \ref{fig:5x5HHCorr} shows the two self correlation functions
356     for the Head groups of the lipids. The first peak at 4.03~$\mbox{\AA}$ is the
357     nearest neighbor separation of the heads of two lipids.
358 mmeineke 102
359    
360    
361    
362     \section{Second Simulation: 50 randomly oriented lipids in water}
363    
364     the second simulation
365    
366 mmeineke 101 \section{Future Directions}
367    
368    
369 mmeineke 100 \pagebreak
370     \bibliographystyle{achemso}
371 mmeineke 99 \bibliography{canidacy_paper} \end{document}