Advertisement

Generalized Systems Governing Probability Density Functions

  • Aslak Tveito
  • Glenn T. Lines
Open Access
Chapter
Part of the Lecture Notes in Computational Science and Engineering book series (LNCSE, volume 111)

Abstract

So far we have considered one-dimensional and two-dimensional release processes. When the channel can take on two states—open or closed—we have seen that the associated probability density functions are governed by 2 × 2 systems of partial differential equations. When a drug is added to the Markov model, an extra state is introduced associated with either the open or the closed state and we obtain a model for the probability density functions phrased in terms of 3 × 3 systems of partial differential equations. In subsequent chapters, we will study situations involving many states and, to do so without drowning in cumbersome notation, we need mathematical formalism to present such models compactly. The compact form we use here is taken from Huertas and Smith [35]. We will introduce the more compact notation simply by providing a couple of examples. These will, hopefully, clarify how to formulate rather complex models in an expedient manner.

So far we have considered one-dimensional and two-dimensional release processes. When the channel can take on two states—open or closed—we have seen that the associated probability density functions are governed by 2 × 2 systems of partial differential equations. When a drug is added to the Markov model, an extra state is introduced associated with either the open or the closed state and we obtain a model for the probability density functions phrased in terms of 3 × 3 systems of partial differential equations. In subsequent chapters, we will study situations involving many states and, to do so without drowning in cumbersome notation, we need mathematical formalism to present such models compactly. The compact form we use here is taken from Huertas and Smith [35]. We will introduce the more compact notation simply by providing a couple of examples. These will, hopefully, clarify how to formulate rather complex models in an expedient manner.

7.1 Two-Dimensional Calcium Release Revisited

Let us start by recalling that the two-dimensional process of calcium release illustrated in Fig.  5.2 on page 92 can be modeled as
$$\displaystyle\begin{array}{rcl} \bar{x}^{{\prime}}(t) =\bar{\gamma } (t)v_{ r}\left (\bar{y} -\bar{ x}\right ) + v_{d}\left (c_{0} -\bar{ x}\right ),& &{}\end{array}$$
(7.1)
$$\displaystyle\begin{array}{rcl} \bar{y}^{{\prime}}(t) =\bar{\gamma } (t)v_{ r}\left (\bar{x} -\bar{ y}\right ) + v_{s}\left (c_{1} -\bar{ y}\right ),& &{}\end{array}$$
(7.2)
where \(\bar{\gamma }=\bar{\gamma } (t)\) is a stochastic variable governed by a Markov model represented by a reaction scheme of the form
$$\displaystyle{C\mathop{\mathop{ \leftrightarrows }\limits^{ k_{oc}}}\limits_{k_{co}}O.}$$
We have seen (see, e.g., page 102) that the probability density functions of the open state (ρ o ) and the closed state (ρ c ) are governed by the system
$$\displaystyle\begin{array}{rcl} \frac{\partial \rho _{o}} {\partial t} + \frac{\partial } {\partial x}\left (a_{o}^{x}\rho _{ o}\right ) + \frac{\partial } {\partial y}\left (a_{o}^{y}\rho _{ o}\right )& = k_{co}\rho _{c} - k_{oc}\rho _{o},&{}\end{array}$$
(7.3)
$$\displaystyle\begin{array}{rcl} \frac{\partial \rho _{c}} {\partial t} + \frac{\partial } {\partial x}\left (a_{c}^{x}\rho _{ c}\right ) + \frac{\partial } {\partial y}\left (a_{c}^{y}\rho _{ c}\right )& = k_{oc}\rho _{o} - k_{co}\rho _{c},&{}\end{array}$$
(7.4)
where
$$\displaystyle\begin{array}{rcl} a_{o}^{x}& =& v_{ r}\left (y - x\right ) + v_{d}\left (c_{0} - x\right ), \\ a_{o}^{y}& =& v_{ r}\left (x - y\right ) + v_{s}\left (c_{1} - y\right ), \\ a_{c}^{x}& =& v_{ d}\left (c_{0} - x\right ), \\ a_{c}^{y}& =& v_{ s}\left (c_{1} - y\right ). {}\end{array}$$
(7.5)
To prepare ourselves for more complex systems, we number the states in this simple system with i = 1, 2, where i = 1 is for the open state and i = 2 is for the closed state. The system can now be written in the form
$$\displaystyle{\frac{\partial \rho _{i}} {\partial t} + \frac{\partial } {\partial x}\left (a_{i}^{x}\rho _{ i}\right ) + \frac{\partial } {\partial y}\left (a_{i}^{y}\rho _{ i}\right ) = \left (K\rho \right )_{i},}$$
where \(\left (K\rho \right )_{i}\) denotes the ith component of the matrix vector product K ρ. Here the vector ρ is given by
$$\displaystyle{\rho = \left (\begin{array}{c} \rho _{1}\\ \rho _{ 2} \end{array} \right ) = \left (\begin{array}{c} \rho _{o}\\ \rho _{c } \end{array} \right )}$$
and the matrix is given by
$$\displaystyle{K = \left (\begin{array}{cc} - k_{12} & k_{21} \\ k_{12} & - k_{21} \end{array} \right ) = \left (\begin{array}{cc} - k_{oc}& k_{co} \\ k_{oc} & - k_{co} \end{array} \right ).}$$
Furthermore, we introduce the functions
$$\displaystyle\begin{array}{rcl} a_{i}^{x}& =& \gamma _{ i}v_{r}\left (y - x\right ) + v_{d}\left (c_{0} - x\right ), {}\\ a_{i}^{y}& =& \gamma _{ i}v_{r}\left (x - y\right ) + v_{s}\left (c_{1} - y\right ), {}\\ \end{array}$$
where γ i is one for the open state (i.e., i = 1) and zero for the closed state (i.e., i = 2).

7.2 Four-State Model

It useful to illustrate this compact notation for a slightly more complex model based on four states. Suppose that the Markov model governing the stochastic variable \(\bar{\gamma }\) in model (7.1) and (7.2) is based on four states: two open states O1 and O2 and two closed states C1 and C2, as shown in Fig. 7.1.
Fig. 7.1

Markov model including four possible states: two open states, O1 and O2, and two closed states, C1 and C2

The probability density system associated with the model (7.1) and (7.2) when the Markov model is given by Fig. 7.1 can now be written in the form

$$\displaystyle\begin{array}{rcl} \frac{\partial \rho _{o_{1}}} {\partial t} + \frac{\partial } {\partial x}\left (a_{o}^{x}\rho _{ o_{1}}\right ) + \frac{\partial } {\partial y}\left (a_{o}^{y}\rho _{ o_{1}}\right )& =& k_{c_{1}o_{1}}\rho _{c_{1}} -\left (k_{o_{1}c_{1}} + k_{o_{1}o_{2}}\right )\rho _{o_{1}} + k_{o_{2}o_{1}}\rho _{o_{2}}, \\ \frac{\partial \rho _{o_{2}}} {\partial t} + \frac{\partial } {\partial x}\left (a_{o}^{x}\rho _{ o_{2}}\right ) + \frac{\partial } {\partial y}\left (a_{o}^{y}\rho _{ o_{2}}\right )& =& k_{c_{2}o_{2}}\rho _{c_{2}} -\left (k_{o_{2}c_{2}} + k_{o_{2}o_{1}}\right )\rho _{o_{2}} + k_{o_{1}o_{2}}\rho _{o_{1}}, \\ \frac{\partial \rho _{c_{1}}} {\partial t} + \frac{\partial } {\partial x}\left (a_{c}^{x}\rho _{ c_{1}}\right ) + \frac{\partial } {\partial y}\left (a_{c}^{y}\rho _{ c_{1}}\right )& =& k_{o_{1}c_{1}}\rho _{o_{1}} -\left (k_{c_{1}o_{1}} + k_{c_{1}c_{2}}\right )\rho _{c_{1}} + k_{c_{2}c_{1}}\rho _{c_{2}}, \\ \frac{\partial \rho _{c_{2}}} {\partial t} + \frac{\partial } {\partial x}\left (a_{c}^{x}\rho _{ c_{2}}\right ) + \frac{\partial } {\partial y}\left (a_{c}^{y}\rho _{ c_{2}}\right )& =& k_{c_{1}c_{2}}\rho _{c_{1}} -\left (k_{c_{2}c_{1}} + k_{c_{2}o_{2}}\right )\rho _{c_{2}} + k_{o_{2}c_{2}}\rho _{o_{2}},{}\end{array}$$
(7.6)
where
$$\displaystyle\begin{array}{rcl} a_{o}^{x}& =& v_{ r}\left (y - x\right ) + v_{d}\left (c_{0} - x\right ), \\ a_{o}^{y}& =& v_{ r}\left (x - y\right ) + v_{s}\left (c_{1} - y\right ), \\ a_{c}^{x}& =& v_{ d}\left (c_{0} - x\right ), \\ a_{c}^{y}& =& v_{ s}\left (c_{1} - y\right ). {}\end{array}$$
(7.7)
By defining the states O1, O2, C1, and C2 to be the states 1, 2, 3, and 4, respectively, we can write the system (7.6) in the more compact form
$$\displaystyle{ \frac{\partial \rho _{i}} {\partial t} + \frac{\partial } {\partial x}\left (a_{i}^{x}\rho _{ i}\right ) + \frac{\partial } {\partial y}\left (a_{i}^{y}\rho _{ i}\right ) = \left (K\rho \right )_{i} }$$
(7.8)
for i = 1, 2, 3, 4, where
$$\displaystyle\begin{array}{rcl} a_{i}^{x}& =& \gamma _{ i}v_{r}\left (y - x\right ) + v_{d}\left (c_{0} - x\right ), {}\\ a_{i}^{y}& =& \gamma _{ i}v_{r}\left (x - y\right ) + v_{s}\left (c_{1} - y\right ), {}\\ \end{array}$$
and ρ = (ρ1, ρ2, ρ3, ρ4) T . Here γ i is one for the open states (i.e., i = 1 and i = 2) and zero for the closed states (i.e., i = 3 and i = 4). Furthermore, the matrix is given by
$$\displaystyle{K = \left (\begin{array}{cccc} -\left (k_{o_{1}c_{1}} + k_{o_{1}o_{2}}\right )& k_{o_{2}o_{1}} & k_{c_{1}o_{1}} & 0 \\ k_{o_{1}o_{2}} & -\left (k_{o_{2}c_{2}} + k_{o_{2}o_{1}}\right )& 0 & k_{c_{2}o_{2}} \\ k_{o_{1}c_{1}} & 0 & -\left (k_{c_{1}o_{1}} + k_{c_{1}c_{2}}\right )& k_{c_{2}c_{1}} \\ 0 & k_{o_{2}c_{2}} & k_{c_{1}c_{2}} & -\left (k_{c_{2}c_{1}} + k_{c_{2}o_{2}}\right ) \end{array} \right ),}$$
which in compact notation is
$$\displaystyle{K = \left (\begin{array}{cccc} -\left (k_{13} + k_{12}\right )& k_{21} & k_{31} & 0 \\ k_{12} & -\left (k_{24} + k_{21}\right )& 0 & k_{42} \\ k_{13} & 0 & -\left (k_{31} + k_{34}\right )& k_{43} \\ 0 & k_{24} & k_{34} & -\left (k_{43} + k_{42}\right ) \end{array} \right ).}$$

7.3 Nine-State Model

We have seen how to formulate probability density systems for two-state and four-state Markov models. For even larger Markov models, it is useful to introduce two-dimensional numbering. This will be illustrated using the nine-state model given in Fig. 7.2.
Fig. 7.2

Markov model including nine possible states

 Here S ij , i, j = 1, 2, 3, denotes the states of the Markov model and K ij mn denotes1 the reaction rate from the state S ij to the state S mn . The system governing the probability density functions of these states can be written in the form
$$\displaystyle{ \frac{\partial \rho _{ij}} {\partial t} + \frac{\partial } {\partial x}\left (a_{ij}^{x}\rho _{ ij}\right ) + \frac{\partial } {\partial y}\left (a_{ij}^{y}\rho _{ ij}\right ) = R_{ij}, }$$
(7.9)
where
$$\displaystyle\begin{array}{rcl} R_{ij}& =& K_{i,j+1}^{i,j}\rho _{ i,j+1} + K_{i+1,j}^{i,j}\rho _{ i+1,j} + K_{i,j-1}^{i,j}\rho _{ i,j-1} + K_{i-1,j}^{i,j}\rho _{ i-1,j} {}\\ & -& \left (K_{i,j}^{i,j+1} + K_{ i,j}^{i+1,j} + K_{ i,j}^{i,j-1} + K_{ i,j}^{i-1,j}\right )\rho _{ i,j}. {}\\ \end{array}$$
Here ρ ij denotes the probability density function of the state S ij and we use the convention that K ij mn  = 0 for \(i,j,m,n\notin \left \{1,2,3\right \}.\) We also have
$$\displaystyle\begin{array}{rcl} a_{ij}^{x}& =& \gamma _{ ij}v_{r}\left (y - x\right ) + v_{d}\left (c_{0} - x\right ), {}\\ a_{ij}^{y}& =& \gamma _{ ij}v_{r}\left (x - y\right ) + v_{s}\left (c_{1} - y\right ), {}\\ \end{array}$$
where γ ij  = 1 when the state S ij represents an open state and γ ij  = 0 when S ij represents a closed state.

Footnotes

  1. 1.

    We use K ij as shorthand for Ki, j, but we use the comma when an index of the form j + 1 is needed, that is we write Ki, j+1. 

References

  1. 35.
    M.A. Huertas, G.D. Smith, The dynamics of luminal depletion and the stochastic gating of Ca2+-activated Ca2+ channels and release sites. J. Theor. Biol. 246(2), 332–354 (2007)MathSciNetCrossRefGoogle Scholar

Copyright information

© The Author(s) 2016

Open Access This chapter is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits use, duplication, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, a link is provided to the Creative Commons license and any changes made are indicated.

The images or other third party material in this chapter are included in the work’s Creative Commons license, unless indicated otherwise in the credit line; if such material is not included in the work’s Creative Commons license and the respective action is not permitted by statutory regulation, users will need to obtain permission from the license holder to duplicate, adapt or reproduce the material.

Authors and Affiliations

  • Aslak Tveito
    • 1
  • Glenn T. Lines
    • 1
  1. 1.Center for Biomedical ComputingSimula Research LaboratoryLysakerNorway

Personalised recommendations