Automated kinetic simulation of molecular interaction networks [Elektronische Ressource] / von Jakub Pijewski

Sujets

• Biologie

Informations

• Publié par: goethe_universitat_frankfurt_am_main
• Publié le: 01 janvier 2007
• Nombre de lectures: 20
• Langue: English
• Poids de l'ouvrage: 2 Mo

Extrait

Automated kinetic simulation of
molecular interaction networks
Dissertation
zur Erlangung des Doktorgrades
der Naturwissenschaften
vorgelegt beim Fachbereich Biowissenschaften
der Johann Wolfgang Goethe Universit¨at
in Frankfurt am Main
von
Jakub Pijewski
aus Warschau
Frankfurt 2007
(D30)vom Fachbereich Biowissenschaften der
Johann Wolfgang Goethe-Universit¨at
als Dissertation angenommen
Dekan: Prof. Dr. Rud¨ iger Wittig
Erster Gutachter: Prof. Dr. Bernd Schu¨rmann
Zweiter Prof. Dr. Gisbert Schneider
Datum der Disputation:Contents
Contents i
1 Introduction 3
1.1 Features of MIN in the cell . . . . . . . . . . . . . . . . . . . 3
1.1.1 Definition of molecular interaction . . . . . . . . . . . 3
1.1.2 Detection of interactions and networks . . . . . . . . 4
1.1.3 Topological characteristics of MIN . . . . . . . . . . . 5
1.1.4 Combinatorial complexity in MIN . . . . . . . . . . . 7
1.2 Models of MIN . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.2.1 Overview of modeling strategies . . . . . . . . . . . . 8
1.2.2 Kinetic models . . . . . . . . . . . . . . . . . . . . . 9
1.2.3 Elementary and approximated mass action approach 11
1.3 Summary of modeling requirements and alternatives . . . . . 13
2 Formal description of molecular interaction networks 15
2.1 Conceptual framework . . . . . . . . . . . . . . . . . . . . . 15
2.1.1 Two-level network representation . . . . . . . . . . . 16
2.1.2 Components of the model . . . . . . . . . . . . . . . 17
2.1.3 Agent-like view of the network . . . . . . . . . . . . . 19
2.2 Representation of basic interactions . . . . . . . . . . . . . . 20
2.2.1 Representation of binding . . . . . . . . . . . . . . . 20
2.2.2ntation of enzymatic reactions . . . . . . . . 22
2.2.3 Representation of synthesis and degradation . . . . . 24
2.3 Composition of basic interactions into complex modules . . . 27
2.3.1 Multiple binding . . . . . . . . . . . . . . . . . . . . 27
2.3.2 Coupled forward and backward enzymatic reactions . 30
2.3.3 Multiple coupled enzymatic reactions . . . . . . . . . 31
2.4 Representation of regulation . . . . . . . . . . . . . . . . . . 38
iii CONTENTS
2.4.1 Regulation of binding and enzymatic reactions . . . . 38
2.4.2tion of protein synthesis . . . . . . . . . . . . 47
2.4.3 Regulation of degradation . . . . . . . . . . . . . . . 49
2.5 Representation of phenomena . . . . . . . . . . . . . . . . . 51
2.5.1 Species-to-phenomenon relationship. . . . . . . . . . 53
2.5.2 Phenomenon-to-phenomenon relationship. . . . . . . 53
2.5.3 Phenomenon-to-species relationship. . . . . . . . . . 54
3 Automated model construction and simulation 55
3.1 General algorithm . . . . . . . . . . . . . . . . . . . . . . . . 56
3.2 Structure of the job file . . . . . . . . . . . . . . . . . . . . . 56
3.2.1 Job file - Interactions . . . . . . . . . . . . . . . . . . 59
3.2.2 Job file - Regulation . . . . . . . . . . . . . . . . . . 60
3.2.3 Job file - Initial Values . . . . . . . . . . . . . . . . . 60
3.2.4 Job file - Phenomena . . . . . . . . . . . . . . . . . . 61
3.3 Definition of agents . . . . . . . . . . . . . . . . . . . . . . . 62
3.4 Communication between agents via message board. . . . . . 63
3.4.1 Message board - Interactions . . . . . . . . . . . . . . 64
3.4.2 Message board - sub-species . . . . . . . . . . . . . . 66
3.4.3 Message board - Conversions . . . . . . . . . . . . . . 67
3.4.4 Message board - Relations . . . . . . . . . . . . . . . 70
3.4.5 Message board - Phenomena . . . . . . . . . . . . . . 72
3.5 Simulation and analysis . . . . . . . . . . . . . . . . . . . . 73
3.5.1 Definition of dynamical variables . . . . . . . . . . . 73
3.5.2 Derivation of equations from message board . . . . . 73
3.5.3 Numerical integration. . . . . . . . . . . . . . . . . . 75
3.5.4 Steady state analysis . . . . . . . . . . . . . . . . . . 77
3.5.5 Parameter exploration and bifurcation diagrams . . . 77
3.5.6 Comparison with experimental data . . . . . . . . . . 78
3.5.7 Mutation and knock-out analysis . . . . . . . . . . . 78
4 Simulation of selected biological systems 79
4.1 Description of analyzed systems . . . . . . . . . . . . . . . . 79
4.1.1 Kinase-phosphatase motif . . . . . . . . . . . . . . . 80
4.1.2 Linear cascade . . . . . . . . . . . . . . . . . . . . . 81
4.1.3 Branched cascade and G2/M transition pathway. . . 82
4.2 Methods.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
4.2.1 Definition of response . . . . . . . . . . . . . . . . . . 86
4.2.2 of parameters . . . . . . . . . . . . . . . . 86iii
4.2.3 Calculations. . . . . . . . . . . . . . . . . . . . . . . 87
4.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
4.3.1 Simulation of a kinase-phosphatase motif . . . . . . . 88
4.3.2 Simulation of a linear cascade . . . . . . . . . . . . . 91
4.3.3 SimulationofabranchedcascadeandtheG2/Mtran-
sition pathway. . . . . . . . . . . . . . . . . . . . . . 93
5 Discussion 103
5.1 Formal description . . . . . . . . . . . . . . . . . . . . . . . 103
5.1.1 Graphical description of MIN . . . . . . . . . . . . . 103
5.1.2 ODE-based description of MIN . . . . . . . . . . . . 108
5.2 MIN simulation software . . . . . . . . . . . . . . . . . . . . 110
5.2.1 Existing MIN simulation software . . . . . . . . . . . 111
5.2.2 Automation and scope of the simulation process . . . 115
5.2.3 Descriptive scope . . . . . . . . . . . . . . . . . . . . 120
5.3 Simulation results . . . . . . . . . . . . . . . . . . . . . . . . 122
5.3.1 Behavior of different systems . . . . . . . . . . . . . . 122
5.3.2 Effect of combinatorial complexity and EMA . . . . . 125
5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
6 Zusammenfassung 129
6.1 Einfuhrung¨ . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
6.2 Methode . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
6.3 Implementierung . . . . . . . . . . . . . . . . . . . . . . . . 132
6.4 Simulation biologischer Systeme . . . . . . . . . . . . . . . . 133
Bibliography 135Acknowledgements
I am grateful to many persons who have contributed to making this work a
success. Especially, I would like to thank:
Prof. Dr. Bernd Shurma¨ nn for supervising and evaluating this work and
for his generous support.
Prof. Dr. Gisbert Schneider for his commitment and evaluation of this
work.
Prof. Dr. Horst St¨ocker for his generous support.
Special thanks to Dr. Martin Stetter for his constant support, supervi-
sion, inspiration and enthusiasm.
Dr. Michael Meyer-Hermann and his group at FIAS and to Dr. Martin
Stetter and his group at Siemens for insightful discussions, very pleasant
working atmosphere and good time.
Last but not least, to my family and friends, especially to Andy for his
heart-and-soul engagement and to Tatiana for the greatest PhD 101 ever.
1Chapter 1
Introduction
The biological complexity comprises of several levels, starting with single
molecules, going through biochemical reactions, cells, tissues, organisms
and ending up at ecosystems [121]. All of these levels are at the interest of
theoretical research in the domain of Systems Biology and Computational
Biology and can be modeled mathematically [49], [68], [158] and simulated
with computational methods [76], [135], [201] in order to unveil the mecha-
nisms determining the organization at each level and support experimental
research [122] [123].
The level of molecular interaction networks (MIN), is a good example of
howthetheoreticalresearchcancontributetounderstandingvitalbiological
and medical problems, such as regulation of gene expression [25], intracel-
lular signaling [97], regulation of metabolism [211] or drug discovery [139].
Thus, it is of general interest to develop efficient approaches for modeling
of MIN [45] [128], [135], [190]. This work presents one such approach.
1.1 Features of MIN in the cell
1.1.1 Definition of molecular interaction
Living cells are filled with a number of molecule types, proteins in par-
ticular, that can potentially interact with each other [24], [145]. Pairwise
interactionscanbecombinedintopathwayswhichgeneratecomplexcellular
responses [166], [167].
There are 3 established standards for interaction data exchange and
modeling: Proteomics Standards Initiative Molecular Interaction XML for-
34 CHAPTER 1. INTRODUCTION
mat(PSIMI)[104][105],SystemsBiologyMarkupLanguage(SBML)[83],[109]
and Biological Pathways Exchange format (BioPAX) [28]. All these stan-
dards consider interactions to be always based on simple physical contact
of molecules [200].
Furthermore, the mentioned standards differentiate between several ba-
sic types of physical interaction, such as binding or enzymatic modification.
ThePSIMIdefinesonlyonetypeofphysicalinteraction-aggregation(bind-
ing), formalized description of other interaction types, such as enzymatic
modification is under development [162]. The SBML is able to describe
binding, transformation and transport, which can be related to kinetic rate
laws. The BioPAX offers the broadest descriptive scope with such cate-
gories as: complex assembly, catalysis, modulation, transport (see [200] for
review).
Finally, the basic interaction types can be further classified into nu-