Scribd Logo
Upload

Welcome to Scribd!

  • 38 views

    Modeling Cellular Signaling Systems: An Abstraction-Refinement Approach

    Uploaded by

    dhermit
    This document describes a modeling strategy for cellular signaling systems using a probabilistic temporal concurrent constraint (CCP) calculus. The approach models systems as abstract models that can be incrementally refined by adding more details. It was applied to model the interaction between G-protein coupled receptors and G-proteins. The model divides the system into environmental domains and uses stochastic equations to represent interactions between molecules. Simulating the model produced results consistent with previous studies, demonstrating the methodology.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    Modeling Cellular Signaling Systems: An Abstraction-Refinement Approach

    Uploaded by

    dhermit
    38 views5 pages
    This document describes a modeling strategy for cellular signaling systems using a probabilistic temporal concurrent constraint (CCP) calculus. The approach models systems as abstract models that can be incrementally refined by adding more details. It was applied to model the interaction between G-protein coupled receptors and G-proteins. The model divides the system into environmental domains and uses stochastic equations to represent interactions between molecules. Simulating the model produced results consistent with previous studies, demonstrating the methodology.

    Original Title

    paams11

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document describes a modeling strategy for cellular signaling systems using a probabilistic temporal concurrent constraint (CCP) calculus. The approach models systems as abstract models that can be incrementally refined by adding more details. It was applied to model the interaction between G-protein coupled receptors and G-proteins. The model divides the system into environmental domains and uses stochastic equations to represent interactions between molecules. Simulating the model produced results consistent with previous studies, demonstrating the methodology.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    38 views5 pages

    Modeling Cellular Signaling Systems: An Abstraction-Refinement Approach

    Uploaded by

    dhermit
    This document describes a modeling strategy for cellular signaling systems using a probabilistic temporal concurrent constraint (CCP) calculus. The approach models systems as abstract models that can be incrementally refined by adding more details. It was applied to model the interaction between G-protein coupled receptors and G-proteins. The model divides the system into environmental domains and uses stochastic equations to represent interactions between molecules. Simulating the model produced results consistent with previous studies, demonstrating the methodology.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 5

    Modeling Cellular Signaling Systems: An

    Abstraction-Refinement Approach

    Diana Hermith, Carlos Olarte, Camilo Rueda, and Frank D. Valencia

    Abstract The molecular mechanisms of cell communication with the environment


    involve many concurrent processes governing dynamically the cell function. This
    concurrent behavior makes traditional methods, as differential equations, unsatis-
    factory as modeling strategy since they do not scale well when a more detailed view
    of the system is required. We describe a modeling strategy based on a probabilistic
    temporal concurrent constraint (CCP) calculus. Starting from an abstract model, we
    build refinements adding further details coming from experimentation or abstract
    assumptions. The advantage of our approach are: Due to the notion of partial in-
    formation as constraints in CCP, the model can be straightforwardly extended when
    more information is available; qualitative and quantitative information can be repre-
    sented by means of probabilistic constructs of CCP; finally, the model is a runnable
    specification and can be executed. We apply this methodology to model the interac-
    tion of G-protein-coupled receptors with their respective G-proteins.

    1 Introduction

    Molecular biologists use information and computer technology to process, analyze,


    understand, compare and share scientific knowledge. The major effort is to scale
    up to system biology, taking under consideration spatio-temporal interactions of
    molecules. Complex biological processes are orchestrated by means of a precise
    dynamic regulation of cell behavior, achieved through an active dialog between cells
    and their environment controlled by cell-surface receptors. In response to specific

    Diana Hermith, Carlos Olarte, Camilo Rueda (e-mail: {carlosolarte,crueda,


    dhermith}@javerianacali.edu.co)
    Dept. of Natural Sciences and Mathematics, and Computer Science, Javeriana University Cali,
    Colombia.
    Frank D. Valencia (e-mail: fvalenci@lix.polytechnique.fr)
    CNRS LIX, Ecole Polytechnique, France.

    1
    2 Diana Hermith, Carlos Olarte, Camilo Rueda, and Frank D. Valencia

    ligands, these translate the environmental cues into specific intracellular signaling
    reactions to achieve an appropriate response [1].
    Formal computational methods can be useful in this setting to develop reason-
    ing skills and to establish conceptual frameworks to handily represent biological
    behavior. This contributes not only to theoretical biology, but also to experimental
    biologists by offering a fertile substrate to think and redesign experiments.
    This paper strives to contribute in the modeling of biological phenomena by using
    a compositional and scalable representation of them. For this end, we shall use a
    probabilistic temporal concurrent constraint language that allows for the modeling
    of reactive systems where: 1) the environment reacts continuously with the system;
    2) the system evolves in discrete time units; 3) there is no a complete description of
    the some components (partial information); and 4) the components react accordingly
    to stochastic laws. Our approach allows for building abstract models of the system
    that are incrementally refined by adding new information. Furthermore, the model
    can be directly executed in a simulation tool. This is a salient feature for biologists
    since they can observe the reaction of the system when parameters are adjusted.
    We shall apply this method to model Guanine proteins (G-proteins) and Gua-
    nine nucleotide-binding protein-coupled receptors (GPCRs). These components are
    a crucial family of signal transduction molecules that govern a variety of physiolog-
    ical functions. They have been (and continue to be) a major exploitable drug target
    giving rise to a plethora of clinically relevant molecules. Compositional and extensi-
    ble modeling tools as the one proposed here may help to understand the fundamental
    properties of these systems, thus contributing to the future of drug discovery.

    2 The modeling language

    Nowadays concurrent systems are ubiquitous in several domains as biological sys-


    tems, security protocols, mobile systems, etc. In general, concurrent systems exhibit
    complex forms of interaction, not only among their internal components, but also
    with the surrounding environment. In computer science, process calculi have arisen
    as mathematical formalisms to model and reason about concurrent systems.
    We shall use as modeling language Concurrent Constraint Programming (CCP)
    [9], a model for concurrency that combines the traditional operational view of pro-
    cess calculi with a declarative one based upon logic. CCP has successfully been
    used in the modelling and verification of several concurrent scenarios, e.g., timed,
    reactive and stochastic systems [9, 6].
    Agents in CCP interact with each other by telling and asking constraints in a
    global store. Constraints (e.g, x > 42) represent (partial) information about the vari-
    ables of the system. The basic constructs in CCP are tell(c) adding the constraint c
    to the store, thus making it available to the other processes; and the ask when c do P
    querying if the current store can entail the guard c; if so, it behaves like P. Otherwise
    it remains blocked until more information is added. This way, ask processes define
    a synchronization mechanism based on entailment of constraints.
    Modeling Cellular Signaling Systems: An Abstraction-Refinement Approach 3

    (a) Process model (b) Simulation results

    Fig. 1: Modeling of the interaction of G-protein-coupled receptors (GPCRs) with Heterotrimeric


    G proteins

    CCP features also constructs for declaring local variables as in (local x) P and for
    executing processes in parallel as in P k Q. Furthermore, temporal and probabilistic
    extensions of CCP have been proposed to deal with the notion of discrete time [8]
    and probabilistic behavior [4]. For instance, it is possible to delay one time unit the
    execution of P as in next P and to choose with a probability ρ (resp. 1 − ρ) the
    execution of P (resp. Q) with the construct P +ρ Q.

    3 The Modeling Strategy and Preliminary Results

    In this section we describe the modeling methodology and some findings in the use
    of it to model signaling systems of G-proteins and GPCRs. We tame the complexity
    of the modeling task through different abstraction levels. This allows us to focus on
    particular principles that helps to understand the behavior of the system.
    The most simple picture of the system is the cell-surface receptor, the ligand, the
    G-proteins components, and other supporting molecules interacting in three envi-
    ronmental domains (Figure 1a). The extracellular domain (ED) is the model of the
    signaling of G Protein. The transmembrane domain (TD) is the model of signal-
    ing of the GPCRs including G Protein activation and receptor desensitization. The
    intracellular domain (ID) is the model for the cycle of the heterotrimeric G Protein.
    Each environmental domain is modeled by a set of stoichiometric equations of
    the form a1 X1 + ...an Xn 99K b1Y1 ... + bmYm where X1 ...Xn are reactants that inter-
    act (and are consumed) yielding to the products Y1 ...Ym . Each type of molecule is
    represented as a variable (e.g., Xi ) and equations as CCP processes of the form:
    eq-proc = when X1 ≥ a1 ∧ ... ∧ Xn ≥ an do next (t) tell(Y1 = Y10 + b1 ∧ ... ∧Ym = Ym0 + bm )
    k next tell(X1 = X10 − a1 ∧ ... ∧ Xn = Xn0 − an )
    4 Diana Hermith, Carlos Olarte, Camilo Rueda, and Frank D. Valencia

    Roughly, when the reactants are available, they are consumed and the right hand
    components are produced t time units later. The parameter t allows us to model sys-
    tems where the speed of reactions may vary (see [2] for the complete CCP model).
    The set of equations leads to a simple description of the system by means of
    stoichiometric analysis. We consider equations to describe binding, dissociation,
    complex formation, and transfer of molecule groups [3, 5, 7, 10]. Besides, when
    rates for a reaction to occur are known, they can be added to the model by choosing
    probabilistically (+ρ ) the reaction to be applied.
    The novelty of the modeling design is the subdivision of the macroscopic view
    of the signaling system by partial descriptions conditioned on a minimal set of sup-
    positions (constraints) widely applicable as possible. This abstraction-refinement
    approach is certainly more difficult to achieve if we one uses models based on ordi-
    nary differential equations that contain a large number of parameters.

    Simulating the Model. CCP processes can be seen as runnable specifications of


    a system: The model can be directly simulated in tools as BioWays [2]. We imple-
    mented the model above and simulated a million of time units under different kinetic
    parameters and concentrations of system molecules (see [2] for details).
    The Figure 1b shows some results similar to those reported in [11]. For this case,
    there are six species in the graph. GProtein (turquoise) represents that the α unit
    and the β γ complex are bound and the α unit is bound to GDP in the ID. RL (in
    green for TD and purple for ED) represents that the receptor (GPCRs) and ligand
    are bound. GProtein (red for TD and yellow for ID) represents that the α unit is
    separated from the β γ complex and is bound to GT P (G Protein activation). The
    curves of GαGTP production considering the TD and ED, decrease oppositely to
    the increase of G Protein complex in ID. RL in TD decreases below G-protein in the
    ID-viewpoint. GαGTP is keeping relatively constant in the ID.
    This cross viewpoint allows some interpretations of the system to the same stim-
    uli, that can be explored systematically. Although our target molecule GαGTP re-
    mains relatively constant after activation, the interaction between receptor and lig-
    and, changes when we look inside each domain. The recognition and interpreta-
    tion of this behaviors could provide some insights into the relationship between
    G-Proteins, GPCRs and ligands.

    Ongoing Work. Our model focuses on the characteristic qualitative patterns (sup-
    ported by quantitative information) of the time evolution of the key components. If
    the molecules concentrations and rate constants were available (fully experimental
    sources), the model parameters could be re-estimated in an easy and modular way to
    fit experimental data, thus obtaining a predictive model. Here CCP may be helpful
    since each sub-system can be altered locally without modifying other components.
    We are currently working on models of the control system of intracellular
    metabolic processes for the signal transduction pathway of glycogen breakdown,
    towards find some principles of the G-Protein cycle activation in a complete con-
    text.
    Modeling Cellular Signaling Systems: An Abstraction-Refinement Approach 5

    References

    1. Anand R. Asthagiri1 and Douglas A. Lauffenburger. Bioengineering models of cell signaling.


    Annual Review of Biomedical Engineering, 2:31–53, 2000.
    2. Davide Chiarugi, Diana Hermith, Moreno Falaschi, and Carlos Olarte. Bioways: A ccp-based
    tool for simulating biological systems. Available at http://www.dianahermith.com/
    bioways/.
    3. David Csercsika, Katalin M. Hangosa, and Gyrgy M. Nagy. A simple reaction kinetic model
    of rapid (g protein dependent) and slow (betha-arrestin dependent) transmission. Journal of
    Theoretical Biology, 255(1):119–128, 2008.
    4. Vineet Gupta, Radha Jagadeesan, and Vijay A. Saraswat. Probabilistic concurrent constraint
    programming. In Proc. of CONCUR 97, London, UK, 1997. Springer-Verlag.
    5. Vladimir L. Katanaev and Matey Chornomorets. Kinetic diversity in g-protein-coupled recep-
    tor signalling. Biochem J., 401(2):485–495, 2006.
    6. M. Nielsen, C. Palamidessi, and F.D. Valencia. Temporal concurrent constraint programming:
    Denotation, logic and applications. Nordic Journal of Computing, 9(1), 2002.
    7. Todd A. Riccobene, Geneva M. Omann, and Jennifer J. Linderman. Modeling activation and
    desensitization of g-protein coupled receptors provides insight into ligand efficacy. Journal of
    Theoretical Biology, 200(2):207–222, 1999.
    8. Vijay Saraswat, Radha Jagadeesan, and Vineet Gupta. Foundations of timed concurrent con-
    straint programming. In Proc. of LICS’94. IEEE CS, 1994.
    9. Vijay A. Saraswat. Concurrent Constraint Programming. MIT Press, 1993.
    10. Eduardo D. Sontag. Molecular systems biology and control. In Control 11:396-435. of
    Boolean networks 15, 2005.
    11. Tau-Mu Yi, Hiroaki Kitano, and Melvin I. Simon. A quantitative characterization of the yeast
    heterotrimeric g protein cycle. PNAS, 100(19):10764–10769, 2003.

    You might also like