Get the full ebook with Bonus Features for a Better Reading Experience on ebookgate.

com

Stochastic modelling for systems biology Second

Edition Wilkinson

https://ebookgate.com/product/stochastic-modelling-for-
systems-biology-second-edition-wilkinson/

OR CLICK HERE

DOWLOAD NOW

Download more ebook instantly today at https://ebookgate.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Elements of Stochastic Modelling 2nd Edition Konstantin

Borovkov

https://ebookgate.com/product/elements-of-stochastic-modelling-2nd-
edition-konstantin-borovkov/

ebookgate.com

Stochastic Relations Foundations for Markov Transition

Systems 1st Edition Ernst-Erich Doberkat

https://ebookgate.com/product/stochastic-relations-foundations-for-
markov-transition-systems-1st-edition-ernst-erich-doberkat/

ebookgate.com

Applied Statistics for Network Biology Methods in Systems

Biology Quantitative and Network Biology VCH 1st Edition
Matthias Dehmer
https://ebookgate.com/product/applied-statistics-for-network-biology-
methods-in-systems-biology-quantitative-and-network-biology-vch-1st-
edition-matthias-dehmer/
ebookgate.com

Modelling in Medicine and Biology 1st Edition C. A.

Brebbia

https://ebookgate.com/product/modelling-in-medicine-and-biology-1st-
edition-c-a-brebbia/

ebookgate.com
Introduction to Stochastic Processes Second Edition
Gregory F. Lawler

https://ebookgate.com/product/introduction-to-stochastic-processes-
second-edition-gregory-f-lawler/

ebookgate.com

Thesaurus of Traditional English Metaphors Second Edition

P. R. Wilkinson

https://ebookgate.com/product/thesaurus-of-traditional-english-
metaphors-second-edition-p-r-wilkinson/

ebookgate.com

Modelling Photovoltaic Systems Using PSpice 1st Edition

Luis Castaner

https://ebookgate.com/product/modelling-photovoltaic-systems-using-
pspice-1st-edition-luis-castaner/

ebookgate.com

Systems Modelling Theory and Practice 1st Edition Michael

Pidd

https://ebookgate.com/product/systems-modelling-theory-and-
practice-1st-edition-michael-pidd/

ebookgate.com

Computational Methods for Modelling of Nonlinear Systems

1st Edition A. Torokhti And P. Howlett (Eds.)

https://ebookgate.com/product/computational-methods-for-modelling-of-
nonlinear-systems-1st-edition-a-torokhti-and-p-howlett-eds/

ebookgate.com
 Bioinformatics

Second
Edition

Praise for the First Edition

“…well suited as an in-depth introduction into stochastic chemical simulation, Stochastic Modelling

Stochastic Modelling for Systems Biology

both for self-study or as a course text…”
—Biomedical Engineering Online, December 2006

Since the first edition of Stochastic Modelling for Systems Biology, there have
for Systems Biology
been many interesting developments in the use of “likelihood-free” methods
of Bayesian inference for complex stochastic models. Re-written to reflect this
modern perspective, this second edition covers everything necessary for a good
SECOND EDITION
appreciation of stochastic kinetic modelling of biological networks in the systems
biology context.

Keeping with the spirit of the first edition, all of the new theory is presented in a
very informal and intuitive manner, keeping the text as accessible as possible to
the widest possible readership.

New in the Second Edition

• All examples have been updated to Systems Biology Markup Language
Level 3
• All code relating to simulation, analysis, and inference for stochastic kinetic
models has been rewritten and restructured in a more modular way
• An ancillary website provides links, resources, errata, and up-to-date
information on installation and use of the associated R package
• More background material on the theory of Markov processes and
stochastic differential equations, providing more substance for
mathematically inclined readers
• Discussion of some of the more advanced concepts relating to stochastic
kinetic models, such as random time change representations, Kolmogorov
equations, Fokker–Planck equations and the linear noise approximation
• Simple modelling of “extrinsic” and “intrinsic” noise

An effective introduction to the area of stochastic modelling in computational

systems biology, this new edition adds additional mathematical detail and
computational methods which will provide a stronger foundation for the
Wilkinson

development of more advanced courses in stochastic biological modelling.

K11715 Darren J. Wilkinson

K11715_Cover.indd 1 10/7/11 8:55 AM

 Stochastic Modelling
for Systems Biology
SECOND EDITION

K11715_FM.indd 1 10/3/11 10:33 AM

 CHAPMAN & HALL/CRC
Mathematical and Computational Biology Series

Aims and scope:

This series aims to capture new developments and summarize what is known
over the entire spectrum of mathematical and computational biology and
medicine. It seeks to encourage the integration of mathematical, statistical,
and computational methods into biology by publishing a broad range of
textbooks, reference works, and handbooks. The titles included in the
series are meant to appeal to students, researchers, and professionals in the
mathematical, statistical and computational sciences, fundamental biology
and bioengineering, as well as interdisciplinary researchers involved in the
field. The inclusion of concrete examples and applications, and programming
techniques and examples, is highly encouraged.

Series Editors

N. F. Britton
Department of Mathematical Sciences
University of Bath

Xihong Lin
Department of Biostatistics
Harvard University

Hershel M. Safer
School of Computer Science
Tel Aviv University

Maria Victoria Schneider

European Bioinformatics Institute

Mona Singh
Department of Computer Science
Princeton University

Anna Tramontano
Department of Biochemical Sciences
University of Rome La Sapienza

Proposals for the series should be submitted to one of the series editors above or directly to:
CRC Press, Taylor & Francis Group
4th, Floor, Albert House
1-4 Singer Street
London EC2A 4BQ
UK

K11715_FM.indd 2 10/3/11 10:33 AM

 Published Titles
Algorithms in Bioinformatics: A Practical Exactly Solvable Models of Biological
Introduction Invasion
Wing-Kin Sung Sergei V. Petrovskii and Bai-Lian Li
Bioinformatics: A Practical Approach Gene Expression Studies Using
Shui Qing Ye Affymetrix Microarrays
Biological Computation Hinrich Göhlmann and Willem Talloen
Ehud Lamm and Ron Unger Glycome Informatics: Methods and
Biological Sequence Analysis Using Applications
the SeqAn C++ Library Kiyoko F. Aoki-Kinoshita
Andreas Gogol-Döring and Knut Reinert Handbook of Hidden Markov Models
Cancer Modelling and Simulation in Bioinformatics
Luigi Preziosi Martin Gollery

Cancer Systems Biology Introduction to Bioinformatics

Edwin Wang Anna Tramontano

Cell Mechanics: From Single Scale- Introduction to Bio-Ontologies

Based Models to Multiscale Modeling Peter N. Robinson and Sebastian Bauer
Arnaud Chauvière, Luigi Preziosi, Introduction to Computational
and Claude Verdier Proteomics
Clustering in Bioinformatics and Drug Golan Yona
Discovery Introduction to Proteins: Structure,
John D. MacCuish and Norah E. MacCuish Function, and Motion
Combinatorial Pattern Matching Amit Kessel and Nir Ben-Tal
Algorithms in Computational Biology An Introduction to Systems Biology:
Using Perl and R Design Principles of Biological Circuits
Gabriel Valiente Uri Alon
Computational Biology: A Statistical Kinetic Modelling in Systems Biology
Mechanics Perspective Oleg Demin and Igor Goryanin
Ralf Blossey Knowledge Discovery in Proteomics
Computational Hydrodynamics of Igor Jurisica and Dennis Wigle
Capsules and Biological Cells Meta-analysis and Combining
C. Pozrikidis Information in Genetics and Genomics
Computational Neuroscience: Rudy Guerra and Darlene R. Goldstein
A Comprehensive Approach Methods in Medical Informatics:
Jianfeng Feng Fundamentals of Healthcare
Data Analysis Tools for DNA Microarrays Programming in Perl, Python, and Ruby
Sorin Draghici Jules J. Berman
Differential Equations and Mathematical Modeling and Simulation of Capsules
Biology, Second Edition and Biological Cells
D.S. Jones, M.J. Plank, and B.D. Sleeman C. Pozrikidis
Dynamics of Biological Systems Niche Modeling: Predictions from
Michael Small Statistical Distributions
Engineering Genetic Circuits David Stockwell
Chris J. Myers

K11715_FM.indd 3 10/3/11 10:33 AM

 Published Titles (continued)
Normal Mode Analysis: Theory and Statistics and Data Analysis for
Applications to Biological and Chemical Microarrays Using R and Bioconductor,
Systems Second Edition
Qiang Cui and Ivet Bahar Sorin Drăghici
Optimal Control Applied to Biological Stochastic Modelling for Systems
Models Biology, Second Edition
Suzanne Lenhart and John T. Workman Darren J. Wilkinson
Pattern Discovery in Bioinformatics: Structural Bioinformatics: An Algorithmic
Theory & Algorithms Approach
Laxmi Parida Forbes J. Burkowski
Python for Bioinformatics The Ten Most Wanted Solutions in
Sebastian Bassi Protein Bioinformatics
Spatial Ecology Anna Tramontano
Stephen Cantrell, Chris Cosner, and
Shigui Ruan
Spatiotemporal Patterns in Ecology
and Epidemiology: Theory, Models,
and Simulation
Horst Malchow, Sergei V. Petrovskii, and
Ezio Venturino

K11715_FM.indd 4 10/3/11 10:33 AM

 Stochastic Modelling
for Systems Biology
SECOND EDITION

Darren J. Wilkinson

K11715_FM.indd 5 10/3/11 10:33 AM

CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

© 2012 by Taylor & Francis Group, LLC

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

Version Date: 2011926

International Standard Book Number-13: 978-1-4398-3776-4 (eBook - PDF)

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts
have been made to publish reliable data and information, but the author and publisher cannot assume
responsibility for the validity of all materials or the consequences of their use. The authors and publishers
have attempted to trace the copyright holders of all material reproduced in this publication and apologize to
copyright holders if permission to publish in this form has not been obtained. If any copyright material has
not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmit-
ted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented,
including photocopying, microfilming, and recording, or in any information storage or retrieval system,
without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.
com (http://www.copyright.com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood
Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-for-profit organization that provides licenses and
registration for a variety of users. For organizations that have been granted a photocopy license by the CCC,
a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used
only for identification and explanation without intent to infringe.
Visit the Taylor & Francis Web site at
http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com
 Contents

List of tables xi

List of figures xiii

Author biography xix

Acknowledgements xxi

Preface to the second edition xxiii

Preface to the first edition xxv

I Modelling and networks 1

1 Introduction to biological modelling 3

1.1 What is modelling? 3
1.2 Aims of modelling 4
1.3 Why is stochastic modelling necessary? 4
1.4 Chemical reactions 9
1.5 Modelling genetic and biochemical networks 10
1.6 Modelling higher-level systems 18
1.7 Exercises 20
1.8 Further reading 20

2 Representation of biochemical networks 21

2.1 Coupled chemical reactions 21
2.2 Graphical representations 21
2.3 Petri nets 24
2.4 Stochastic process algebras 34
2.5 Systems Biology Markup Language (SBML) 36
2.6 SBML-shorthand 41
2.7 Exercises 47
2.8 Further reading 48

vii
viii CONTENTS
II Stochastic processes and simulation 49

3 Probability models 51
3.1 Probability 51
3.2 Discrete probability models 62
3.3 The discrete uniform distribution 70
3.4 The binomial distribution 71
3.5 The geometric distribution 72
3.6 The Poisson distribution 74
3.7 Continuous probability models 77
3.8 The uniform distribution 82
3.9 The exponential distribution 85
3.10 The normal/Gaussian distribution 89
3.11 The gamma distribution 93
3.12 Quantifying “noise” 96
3.13 Exercises 97
3.14 Further reading 98

4 Stochastic simulation 99
4.1 Introduction 99
4.2 Monte Carlo integration 99
4.3 Uniform random number generation 100
4.4 Transformation methods 101
4.5 Lookup methods 106
4.6 Rejection samplers 107
4.7 Importance resampling 110
4.8 The Poisson process 111
4.9 Using the statistical programming language, R 112
4.10 Analysis of simulation output 118
4.11 Exercises 120
4.12 Further reading 122

5 Markov processes 123

5.1 Introduction 123
5.2 Finite discrete time Markov chains 123
5.3 Markov chains with continuous state-space 130
5.4 Markov chains in continuous time 137
5.5 Diffusion processes 152
5.6 Exercises 166
5.7 Further reading 168

III Stochastic chemical kinetics 169

6 Chemical and biochemical kinetics 171

6.1 Classical continuous deterministic chemical kinetics 171
CONTENTS ix
6.2 Molecular approach to kinetics 178
6.3 Mass-action stochastic kinetics 180
6.4 The Gillespie algorithm 182
6.5 Stochastic Petri nets (SPNs) 183
6.6 Structuring stochastic simulation codes 186
6.7 Rate constant conversion 189
6.8 Kolmogorov’s equations and other analytic representations 194
6.9 Software for simulating stochastic kinetic networks 199
6.10 Exercises 200
6.11 Further reading 200

7 Case studies 203

7.1 Introduction 203
7.2 Dimerisation kinetics 203
7.3 Michaelis–Menten enzyme kinetics 208
7.4 An auto-regulatory genetic network 212
7.5 The lac operon 217
7.6 Exercises 219
7.7 Further reading 220

8 Beyond the Gillespie algorithm 221

8.1 Introduction 221
8.2 Exact simulation methods 221
8.3 Approximate simulation strategies 226
8.4 Hybrid simulation strategies 239
8.5 Exercises 245
8.6 Further reading 245

IV Bayesian inference 247

9 Bayesian inference and MCMC 249

9.1 Likelihood and Bayesian inference 249
9.2 The Gibbs sampler 254
9.3 The Metropolis–Hastings algorithm 264
9.4 Hybrid MCMC schemes 268
9.5 Metropolis–Hastings algorithms for Bayesian inference 269
9.6 Bayesian inference for latent variable models 270
9.7 Alternatives to MCMC 274
9.8 Exercises 275
9.9 Further reading 275

10 Inference for stochastic kinetic models 277

10.1 Introduction 277
10.2 Inference given complete data 278
10.3 Discrete-time observations of the system state 281
 10.4 Diffusion approximations for inference 288
10.5 Likelihood-free methods 292
10.6 Network inference and model comparison 308
10.7 Exercises 309
10.8 Further reading 310

11 Conclusions 311

Appendix A SBML Models 315

A.1 Auto-regulatory network 315
A.2 Lotka–Volterra reaction system 318
A.3 Dimerisation-kinetics model 319

References 323

Index 331

x
 List of tables

2.1 The auto-regulatory system displayed in tabular (matrix) form (zero

stoichiometries omitted for clarity) 27
2.2 Table representing the overall effect of each transition (reaction) on
the marking (state) of the network 28

xi
This page intentionally left blank
 List of figures

1.1 Five deterministic solutions of the linear birth–death process for

values of λ − µ given in the legend (x0 = 50). 6
1.2 Five realisations of a stochastic linear birth–death process together
with the continuous deterministic solution (x0 = 50, λ = 3, µ = 4). 7
1.3 Five realisations of a stochastic linear birth–death process together
with the continuous deterministic solution for four different (λ, µ)
combinations, each with λ − µ = −1 and x0 = 50. 8
1.4 Transcription of a single prokaryotic gene. 11
1.5 A simple illustrative model of the transcription process in eukaryotic
cells. 13
1.6 A simple prokaryotic transcription repression mechanism. 14
1.7 A very simple model of a prokaryotic auto-regulatory gene network. 17
1.8 Key mechanisms involving the lac operon. 18

2.1 A simple graph of the auto-regulatory reaction network. 22

2.2 A simple digraph. 23
2.3 A Petri net for the auto-regulatory reaction network. 24
2.4 A Petri net labelled with tokens. 25
2.5 A Petri net with new numbers of tokens after reactions have taken
place. 25

3.1 CDF for the sum of a pair of fair dice. 64

3.2 PMF and CDF for a Bin(8, 0.7) distribution. 71
3.3 PMF and CDF for a P o(5) distribution. 75
3.4 PDF and CDF for a U (0, 1) distribution. 83
3.5 PDF and CDF for an Exp(1) distribution. 85
3.6 PDF and CDF for a N (0, 1) distribution. 91
3.7 Graph of Γ(x) for small positive values of x. 94
3.8 PDF and CDF for a Ga(3, 1) distribution. 95

4.1 Density of Y = exp(X), where X ∼ N (2, 1). 120

4.2 Normal Q–Q plot for the samples resulting from the importance
resampling procedure, showing good agreement with the theoretical
distribution. 121

xiii
xiv LIST OF FIGURES
5.1 An R function to simulate a sample path of length n from a Markov
chain with transition matrix P and initial distribution pi0. 130
5.2 A sample R session to simulate and analyse the sample path of a
finite Markov chain. 131
5.3 SBML-shorthand for the simple gene activation process with
α = 0.5 and β = 1. 140
5.4 A simulated realisation of the simple gene activation process with
α = 0.5 and β = 1. 142
5.5 An R function to simulate a sample path with n events from a
continuous time Markov chain with transition rate matrix Q and
initial distribution pi0. 143
5.6 SBML-shorthand for the immigration-death process with λ = 1 and
µ = 0.1. 144
5.7 A single realisation of the immigration-death process with parame-
ters λ = 1 and µ = 0.1, initialised at X(0) = 0. 145
5.8 R function for discrete-event simulation of the immigration-death
process. 146
5.9 R function for simulation of a diffusion process using the Euler
method. 154
5.10 A single realisation of the diffusion approximation to the immigration-
death process with parameters λ = 1 and µ = 0.1, initialised at
X(0) = 0. 155
5.11 R code for simulating the diffusion approximation to the immigration-
death process. 155

6.1 Lotka–Volterra dynamics for [Y1 ](0) = 4, [Y2 ](0) = 10, k1 =

1, k2 = 0.1, k3 = 0.1. Note that the equilibrium solution for this
combination of rate parameters is [Y1 ] = 1, [Y2 ] = 10. 173
6.2 Lotka–Volterra dynamics in phase-space for rate parameters k1 =
1, k2 = 0.1, k3 = 0.1. 174
6.3 Dimerisation kinetics. 175
6.4 An R function to numerically integrate a system of coupled ODEs
using a simple first-order Euler method. 177
6.5 An R function to implement the Gillespie algorithm for a stochastic
Petri net representation of a coupled chemical reaction system. 184
6.6 Some R code to set up the LV system as a SPN and then simulate it
using the Gillespie algorithm. 185
6.7 A single realisation of a stochastic LV process. 186
6.8 A single realisation of a stochastic LV process in phase-space. 187
6.9 SBML-shorthand for the stochastic Lotka–Volterra system. 188
6.10 An R function to discretise the output of gillespie onto a regular
grid of time points. 189
6.11 An R function to implement the Gillespie algorithm for a SPN,
recording the state on a regular grid of time points. 190
LIST OF FIGURES xv
6.12 An R function which accepts as input an SPN, and returns as output
a function (closure) for advancing the state of the SPN using the
Gillespie algorithm. 191
6.13 An R function to simulate a process on a regular time grid using a
stepping function such as output by StepGillespie. 192
6.14 R code showing how to use the functions StepGillespie and
simTs together in order to simulate a realisation from a SPN. 192

7.1 SBML-shorthand for the dimerisation kinetics model (continuous

deterministic version). 204
7.2 Left: Simulated continuous deterministic dynamics of the dimerisa-
tion kinetics model. Right: A simulated realisation of the discrete
stochastic dynamics of the dimerisation kinetics model. 205
7.3 SBML-shorthand for the dimerisation kinetics model (discrete
stochastic version). 205
7.4 R code to build an SPN object representing the dimerisation kinetics
model. 206
7.5 Left: A simulated realisation of the discrete stochastic dynamics of
the dimerisation kinetics model plotted on a concentration scale.
Right: The trajectories for levels of P from 20 runs overlaid. 207
7.6 Left: The mean trajectory of P together with some approximate
(point-wise) “confidence bounds” based on 1,000 runs of the
simulator. Right: Density histogram of the simulated realisations of
P at time t = 10 based on 10,000 runs, giving an estimate of the
PMF for P (10). 207
7.7 SBML-shorthand for the Michaelis–Menten kinetics model (contin-
uous deterministic version). 210
7.8 Left: Simulated continuous deterministic dynamics of the Michaelis–
Menten kinetics model. Right: Simulated continuous deterministic
dynamics of the Michaelis–Menten kinetics model based on the
two-dimensional representation. 210
7.9 SBML-shorthand for the Michaelis–Menten kinetics model (discrete
stochastic version). 212
7.10 Left: A simulated realisation of the discrete stochastic dynamics of
the Michaelis–Menten kinetics model. Right: A simulated realisation
of the discrete stochastic dynamics of the reduced-dimension
Michaelis–Menten kinetics model. 212
7.11 SBML-shorthand for the reduced dimension Michaelis–Menten
kinetics model (discrete stochastic version). 213
7.12 Left: A simulated realisation of the discrete stochastic dynamics
of the prokaryotic genetic auto-regulatory network model, for a
period of 5,000 seconds. Right: A close-up on the first period of 250
seconds of the left plot. 214
xvi LIST OF FIGURES
7.13 Left: Close-up showing the time-evolution of the number of
molecules of P over a 10-second period. Right: Empirical PMF
for the number of molecules of P at time t = 10 seconds, based on
10,000 runs. 215
7.14 Left: Empirical PMF for the number of molecules of P at time
t = 10 seconds when k2 is changed from 0.01 to 0.02, based
on 10,000 runs. Right: Empirical PMF for the prior predictive
uncertainty regarding the observed value of P at time t = 10 based
on the prior distribution k2 ∼ U (0.005, 0.03). 216
7.15 SBML-shorthand for the lac-operon model (discrete stochastic
version). 218
7.16 A simulated realisation of the discrete stochastic dynamics of the
lac-operon model for a period of 50,000 seconds. 219

8.1 An R function to implement the first reaction method for a stochastic

Petri net representation of a coupled chemical reaction system. 223
8.2 An R function to implement the Poisson timestep method for a
stochastic Petri net representation of a coupled chemical reaction
system. 228
8.3 An R function to integrate the CLE using an Euler method for a
stochastic Petri net representation of a coupled chemical reaction
system. 232
8.4 An R function to integrate a multivariate diffusion process using a
simple Euler–Maruyama method. 235
8.5 Example showing how to use the function StepSDE for the SDE
given in (8.4). 236
8.6 Figure showing realisations of the SDE models for the immigration-
death process discussed in Section 8.3.4 incorporating different
combinations of intrinsic and extrinsic noise. 237

9.1 Plot showing the prior and posterior for the Poisson rate example. 252
9.2 An R function to implement a Gibbs sampler for the simple normal
random sample model. 259
9.3 Example R code illustrating the use of the function normgibbs
from Figure 9.2. 260
9.4 Figure showing the Gibbs sampler output resulting from running the
example code in Figure 9.3. 261
9.5 An R function to implement a Metropolis sampler for a standard
normal random quantity based on U (−α, α) innovations. 268
9.6 Output from the Metropolis sampler given in Figure 9.5. 269

10.1 An R function to create a function closure for marginal likelihood

estimation using a bootstrap particle filter. 296
10.2 An R session showing how to use the function pfMLLik from
Figure 10.1. 297
10.3 Simulated time series data set, LVnoise10, consisting of 16
equally spaced observations of a realisation of a stochastic kinetic
Lotka–Volterra model subject to Gaussian measurement error with a
standard deviation of 10. 298
10.4 R code implementing an MCMC sampler for fully Bayesian
inference for the stochastic Lotka–Volterra model using time course
data. 299
10.5 Marginal posterior distributions for the parameters of the Lotka–
Volterra model, based on the data given in Figure 10.3. 300
10.6 Marginal posterior distributions for the parameters of the Lotka–
Volterra model, based only on observations of prey species levels. 301
10.7 Marginal posterior distributions for the parameters of the Lotka–
Volterra model with unknown measurement error standard deviation. 303
10.8 Marginal posterior distributions for the log-parameters of the Lotka–
Volterra model with unknown measurement error standard deviation. 304

xvii
This page intentionally left blank
 Author biography

Darren Wilkinson is professor of stochastic modelling at Newcastle University in

the United Kingdom. He was educated at the nearby University of Durham, where
he took his first degree in mathematics followed by a PhD in Bayesian statistics
which he completed in 1995. He moved to a lectureship in statistics at Newcastle
University in 1996, where he has remained since, being promoted to his current post
in 2007. Professor Wilkinson is interested in computational statistics and Bayesian
inference and in the application of modern statistical technology to problems in sta-
tistical bioinformatics and systems biology. He is involved in a variety of systems
biology projects at Newcastle, including the Centre for Integrated Systems Biology
of Ageing and Nutrition (CISBAN). He currently holds a BBSRC Research Develop-
ment Fellowship on integrative modelling of stochasticity, noise, heterogeneity and
measurement error in the study of model biological systems.

xix
This page intentionally left blank
 Acknowledgements

I would like to acknowledge the support of everyone at Newcastle University who is

involved with systems biology research. Unfortunately, there are far too many people
to mention by name, but particular thanks are due to everyone involved in the BBRSC
CISBAN project, without whom this book would never have been written.
The production of the second edition of this book has been greatly facilitated by
funding from the Biotechnology and Biological Sciences Research Council, both
through their funding of CISBAN (grant number BBC0082001) and the award to
me of a BBSRC Research Development Fellowship (grant number BBF0235451).
In addition, a considerable amount of work on this second edition was carried out
during a visit I made to the Statistical and Applied Mathematical Sciences Institute
(SAMSI, www.samsi.info) in North Carolina during the spring of 2011, as part
of their research programme on the Analysis of Object-Oriented Data.
Particular thanks are also due to all of the students who have been involved in the
MSc in bioinformatics and computational systems biology programme at Newcastle,
and especially those who took my course on Stochastic Systems Biology, as it was
the teaching of that course which persuaded me that it was necessary to write this
book.
Last, but by no means least, I would like to thank my family for supporting me in
everything that I do.

xxi
This page intentionally left blank
 Preface to the second edition

I was keen to write a second edition of this book even before the first edition was pub-
lished in the spring of 2006. The first edition was written during the latter half of 2004
and the first half of 2005 when the use of stochastic modelling within computational
systems biology was still very much in its infancy. Based on an inter-disciplinary
Masters course I was teaching I saw an urgent need for an introductory textbook in
this area, and tried in the first edition to lay down all of the key ingredients needed
to get started. I think that I largely succeeded, but the emphasis there was very much
on the “bare essentials” and accessibility to non-mathematical readers, and my goal
was to get the book published in a timely fashion, in order to help advance the field.
I would like to think that the first edition of this text has played a small role in help-
ing to make stochastic modelling a much more mainstream part of computational
systems biology today. But naturally there were many limitations of the first edi-
tion. There were several places where I would have liked to have elaborated further,
providing additional details likely to be of interest to the more mathematically or sta-
tistically inclined reader. Also, the latter chapters on inference from data were rather
limited and lacking in concrete examples. This was partly due to the fact that the
whole area of inference for stochastic kinetic models was just developing, and so
it wasn’t possible to give a coherent overview of the problem from an introductory
viewpoint. Since publishing the first edition there have been many interesting devel-
opments in the use of “likelihood-free” methods of Bayesian inference for complex
stochastic models, and so the latter chapters have now been re-written to reflect this
more modern perspective, including a detailed case study accompanied by working
code examples.
Of course the whole field has moved on considerably since 2005, and so the sec-
ond edition is also an opportunity to revise and update, and to change the emphasis
of the text slightly. The Systems Biology Markup Language (SBML) has contin-
ued to evolve, and SBML Level 3 is now finalised. Consequently, I have updated
all of the examples to Level 3, which is likely to remain the standard encoding for
dynamic biological models for the foreseeable future. I have also taken the oppor-
tunity to revise and update the R code examples associated with the book, and to
bundle them all together as an R package (smfsb). This should make it much easier
for people to try out the examples given in the book. I have also re-written and re-
structured all of the code relating to simulation, analysis and inference for stochastic
kinetic models. The code is now structured in a more modular way (using a functional
programming style), making it easy to “bolt together” different models, simulation
algorithms, and analysis tools. I’ve created a new website specific to this second
edition (http://www.staff.ncl.ac.uk/d.j.wilkinson/smfsb/2e/), where

xxiii
xxiv PREFACE TO THE SECOND EDITION
I will keep links, resources, an errata, and up-to-date information on installation and
use of the associated R package.
The new edition contains more background material on the theory of Markov pro-
cesses and stochastic differential equations, providing more substance for mathemati-
cally inclined readers. This allows discussion of some of the more advanced concepts
relating to stochastic kinetic models, such as random time-change representations,
Kolmogorov equations, Fokker–Planck equations and the linear noise approxima-
tion. It also enables simple modelling of “extrinsic” in addition to “intrinsic” noise.
This should make the text suitable for use in a greater range of courses. Naturally, in
keeping with the spirit of the first edition, all of the new theory is presented in a very
informal and intuitive way, in order to keep the text accessible to the widest possible
readership. This is not a rigorous text on the theory of Markov processes (there are
plenty of other good texts in that vein) — the book is still intended for use in courses
for students with a life sciences background.
I’ve also updated the references, and provided new pointers to recent publications
in the literature where this is especially pertinent. However, it should be emphasised
that the book is not intended to provide a comprehensive survey of the stochastic
systems biology literature — I don’t think that is necessary (or even helpful) for an
introductory textbook, and I hope that people working in this area accept this if I fail
to cite their work.
So here it is, the second edition, completed at last. I hope that this text continues to
serve as an effective introduction to the area of stochastic modelling in computational
systems biology, and that this new edition adds additional mathematical detail and
computational methods which will provide a stronger foundation for the development
of more advanced courses in stochastic biological modelling.

Darren Wilkinson
Newcastle upon Tyne
 Preface to the first edition

Stochastic models for chemical and biochemical reactions have been around for a
long time. The standard algorithm for simulating the dynamics of such processes
on a computer (the “Gillespie algorithm”) was published nearly 30 years ago (and
most of the relevant theory was sorted out long before that). In the meantime there
have been dozens of papers published on stochastic kinetics, and several books on
stochastic processes in physics, chemistry and biology. Biological modelling and
biochemical kinetic modelling have been around even longer. These distinct subjects
have started to merge in recent years as technology has begun to give real insight
into intra-cellular processes. Improvements in experimental technology are enabling
quantitative real-time imaging of expression at the single-cell level, and improve-
ment in computing technology is allowing modelling and stochastic simulation of
such systems at levels of detail previously impossible. The message that keeps be-
ing repeated is that the kinetics of biological processes at the intra-cellular level are
stochastic, and that cellular function cannot be properly understood without build-
ing that stochasticity into in silico models. It was this message that first interested
me in systems biology and in the many challenging statistical problems that follow
naturally from this observation.
It was only when I came to try and teach this interesting view of computational
systems biology to graduate students that I realised there was no satisfactory text on
which to base the course. The papers assumed far too much background knowledge,
the standard biological texts didn’t cover stochastic modelling, and the stochastic
processes texts were too far removed from the practical applications of systems biol-
ogy, paying little attention to stochastic biochemical kinetics. Where stochastic mod-
els do crop up in the mainstream systems biology literature, they tend to be treated as
an add-on or after-thought, in a slightly superficial way. As a statistician I see this as
problematic. The stochastic processes formalism provides a beautiful, elegant, and
coherent foundation for chemical kinetics, and there is a wealth of associated theory
every bit as powerful and elegant as that for conventional continuous deterministic
models. Given the increasing importance of stochastic models in systems biology, I
thought it would be particularly appropriate to write an introductory text in this area
from this perspective.
This book assumes a basic familiarity with what might be termed high school
mathematics. That is, a basic familiarity with algebra and calculus. It is also helpful
to have had some exposure to linear algebra and matrix theory, but not vital. Since
the teaching of probability and statistics at school is fairly patchy, essentially noth-
ing will be assumed, though obviously a good background in this area will be very
helpful. Starting from here, the book covers everything that is necessary for a good

xxv
xxvi PREFACE TO THE FIRST EDITION
appreciation of stochastic kinetic modelling of biological networks in the systems
biology context. There is an emphasis on the necessary probabilistic and stochastic
methods, but the theory is rooted in the intended application, and no time is wasted
covering interesting theory that is not necessary for stochastic kinetic modelling. On
the other hand, more-or-less everything that is necessary is covered, and the text (at
least up to Chapter 8) is intended to be self-contained. The final chapters are nec-
essarily a little more technical in nature, as they concern the difficult problem of
inference for stochastic kinetic models from experimental data. This is still an active
research area, and so the main aim here is to give pointers to the existing literature
and provide enough background information to render that literature more accessible
to the non-specialist.
The decision to make the book practically oriented necessitated some technologi-
cal choices that will not suit everyone. The two key technologies chosen for illustrat-
ing the theory in this book are SBML and R. I hope that the choice of the Systems
Biology Markup Language (SBML) for model representation is not too controversial.
It is the closest thing to a standard that exists in the systems biology area, and there
are dozens of software tools that support it. Of course, most people using SBML are
using it to encode continuous deterministic models. However, SBML Level 2 and
beyond are perfectly capable of encoding discrete stochastic models, and so one of
the reasons for using it in this text is to provide some working examples of SBML
models constructed with discrete stochastic simulation in mind.
The other technological choice was the use of the statistical programming lan-
guage, R. This is likely to be more controversial, as there are plenty of other lan-
guages that could have been used. It seemed to me that in the context of a textbook,
using a very high-level language was most appropriate. In the context of stochastic
modelling, a language with good built-in mathematical and statistical support also
seemed highly desirable. R stood out as being the best choice in terms of built-in lan-
guage support for stochastic simulation and statistical analysis. It also has the great
advantage over some of the other possible choices of being completely free open-
source software, and therefore available to anyone reading the book. In addition, R
is being used increasingly in bioinformatics and other areas of systems biology, so
hopefully for this reason too it will be regarded as a positive choice.
This book is intended for a variety of audiences (advanced undergraduates, grad-
uate students, postdocs, and academics from a variety of backgrounds), and exactly
how it is read will depend on the context. The book is certainly suitable for a va-
riety of graduate programs in computational biology. I will be using it as a text for
a second-semester course on a masters in bioinformatics programme that will cover
much of Chapters 1, 2, 4, 5, 6, and 7 (most of the material from Chapter 3 will be
covered in a first-semester course). However, the book has also been written with
self-study in mind, and here it is intended that the entire book be read in sequence,
with some chapters skipped depending on background knowledge. It is intended to be
suitable for computational systems biologists from a continuous deterministic back-
ground who would like to know more about the stochastic approach, as well as for
statisticians who are interested in learning more about systems biology (though statis-
ticians will probably want to skip most of Chapters 3, 4, and 5). It is worth pointing
PREFACE TO THE FIRST EDITION xxvii
out that Chapters 9 and 10 will be easier to read with a reasonable background in
probability and statistics. Though it should be possible to read these chapters with-
out such a background and still appreciate the key concepts and ideas, some of the
technical details may be difficult to understand fully from an elementary viewpoint.
Writing this book has been more effort than I anticipated, and there were one
or two moments when I doubted the wisdom of taking the project on. On the whole
however, it has been an interesting and rewarding experience for me, and I am pleased
with the result. I know it is a book that I will find useful and will refer to often,
as it integrates a fairly diverse literature into a single convenient and notationally
consistent source. I can only hope that others share the same view, and that the book
will help make a stochastic approach to computational systems biology more widely
appreciated.

Darren Wilkinson
Newcastle upon Tyne, October 2005
This page intentionally left blank
 PART I

Modelling and networks

This page intentionally left blank
 CHAPTER 1

Introduction to biological modelling

1.1 What is modelling?

Modelling is an attempt to describe, in a precise way, an understanding of the ele-
ments of a system of interest, their states, and their interactions with other elements.
The model should be sufficiently detailed and precise so that it can in principle be
used to simulate the behaviour of the system on a computer. In the context of molec-
ular cell biology, a model may describe (some of) the mechanisms involved in tran-
scription, translation, gene regulation, cellular signalling, DNA damage and repair
processes, homeostatic processes, the cell cycle, or apoptosis. Indeed any biochemi-
cal mechanism of interest can, in principle, be modelled. At a higher level, modelling
may be used to describe the functioning of a tissue, organ, or even an entire organ-
ism. At still higher levels, models can be used to describe the behaviour and time
evolution of populations of individual organisms.
The first issue to confront when embarking on a modelling project is to decide on
exactly which features to include in the model, and in particular, the level of detail
the model is intended to capture. So, a model of an entire organism is unlikely to
describe the detailed functioning of every individual cell, but a model of a cell is
likely to include a variety of very detailed descriptions of key cellular processes.
Even then, however, a model of a cell is unlikely to contain details of every single
gene and protein.
Fortunately, biologists are used to thinking about processes at different scales and
different levels of detail. Consider, for example, the process of photosynthesis. When
studying photosynthesis for the first time at school, it is typically summarised by a
single chemical reaction mixing water with carbon dioxide to get glucose and oxygen
(catalysed by sunlight). This could be written very simply as
Sunlight
Water + Carbon Dioxide −→ Glucose + Oxygen,
or more formally by replacing the molecules by their chemical formulae and balanc-
ing to get
6H2 O + 6CO2 −→ C6 H12 O6 + 6O2 .
Of course, further study reveals that photosynthesis consists of many reactions, and
that the single reaction was simply a summary of the overall effect of the process.
However, it is important to understand that the above equation is not really wrong, it
just represents the overall process at a higher level than the more detailed description
that biologists often prefer to work with. Whether a single overall equation or a full
breakdown into component reactions is necessary depends on whether intermediaries
such as ADP and ATP are elements of interest to the modeller. Indeed, really accurate

3
4 INTRODUCTION TO BIOLOGICAL MODELLING
modelling of the process would require a model far more detailed and complex than
most biologists would be comfortable with, using molecular dynamic simulations
that explicitly manage the position and momentum of every molecule in the system.
The “art” of building a good model is to capture the essential features of the biol-
ogy without burdening the model with non-essential details. Every model is to some
extent a simplification of the biology, but models are valuable because they take ideas
that might have been expressed verbally or diagrammatically and make them more
explicit, so that they can begin to be understood in a quantitative rather than purely
qualitative way.

1.2 Aims of modelling

The features of a model depend very much on the aims of the modelling exercise. We
therefore need to consider why people model and what they hope to achieve by so
doing. Often the most basic aim is to make clear the current state of knowledge re-
garding a particular system, by attempting to be precise about the elements involved
and the interactions between them. Doing this can be a particularly effective way of
highlighting gaps in understanding. In addition, having a detailed model of a system
allows people to test that their understanding of a system is correct, by seeing if the
implications of their models are consistent with observed experimental data. In prac-
tice, this model validation stage is central to the systems biology approach. However,
this work will often represent only the initial stage of the modelling process. Once
people have a model they are happy with, they often want to use their models predic-
tively, by conducting “virtual experiments” that might be difficult, time-consuming,
or impossible to do in the lab. Such experiments may uncover important indirect re-
lationships between model components that would be hard to predict otherwise. An
additional goal of modern biological modelling is to pool a number of small models
of well-understood mechanisms into a large model in order to investigate the effect
of interactions between the model components. Models can also be extremely useful
for informing the design and analysis of complex biological experiments.
In summary, modelling and computer simulation are becoming increasingly im-
portant in post-genomic biology for integrating knowledge and experimental data
and making testable predictions about the behaviour of complex biological systems.

1.3 Why is stochastic modelling necessary?

Ignoring quantum mechanical effects, current scientific wisdom views biological
systems as essentially deterministic in character, with dynamics entirely predictable
given sufficient knowledge of the state of the system (together with complete knowl-
edge of the physics and chemistry of interacting biomolecules). At first this perhaps
suggests that a deterministic approach to the modelling of biological systems is likely
to be successful. However, despite the rapid advancements in computing technology,
we are still a very long way away from a situation where we might expect to be able
to model biological systems of realistic size and complexity over interesting time
scales using such a molecular dynamic approach. We must therefore use models that
WHY IS STOCHASTIC MODELLING NECESSARY? 5
leave out many details of the “state” of a system (such as the position, orientation, and
momentum of every single molecule under consideration), in favour of a higher-level
view. Viewed at this higher level, the dynamics of the system are not deterministic,
but intrinsically stochastic, and consideration of statistical physics is necessary to un-
cover the precise nature of the stochastic processes governing the system dynamics.
A more detailed discussion of this issue will have to be deferred until much later in
the book, once the appropriate concepts and terminology have been established. In
the meantime, it is helpful to highlight the issues using a very simple example that
illustrates the importance of stochastic modelling, both for simulation and inference.
The example we will consider is known as the linear birth–death process. In the
first instance, it is perhaps helpful to view this as a model for the number of bacteria
in a bacterial colony. It is assumed that each bacterium in the colony gives rise to new
individuals at rate λ (that is, on average, each bacterium will produce λ offspring per
unit time). Similarly, each bacterium dies at rate µ (that is, on average, the proportion
of bacteria that die per unit time is µ). These definitions are not quite right, but we
will define such things much more precisely later. Let the number of bacteria in the
colony at time t be denoted X(t). Assume that the number of bacteria in the colony
at time zero is known to be x0 . Viewed in a continuous deterministic manner, this
description of the system leads directly to the ordinary differential equation
dX(t)
= (λ − µ)X(t),
dt
which can be solved analytically to give the complete dynamics of the system as
X(t) = x0 exp{(λ − µ)t}.
So, predictably, in the case λ > µ the population size will increase exponentially as
t −→ ∞, and will decrease in size exponentially if λ < µ. Similarly, it will remain
at constant size x0 if λ = µ. Five such solutions are given in Figure 1.1. There are
other things worth noting about this solution. In particular, the solution clearly only
depends on λ − µ and not on the particular values that λ and µ take (so, for example,
λ = 0.5, µ = 0 will lead to exactly the same solution as λ = 1, µ = 0.5). In some
sense, therefore, λ − µ (together with x0 ) is a “sufficient” description of the system
dynamics. At first this might sound like a good thing, but it is clear that there is a flip-
side: namely that studying experimental data on bacteria numbers can only provide
information about λ−µ, and not on the particular values of λ and µ separately (as the
data can only provide information about the “shape” of the curve, and the shape of
the curve is determined by λ − µ). Of course, this is not a problem if the continuous
deterministic model is really appropriate, as then λ − µ is the only thing one needs
to know and the precise values of λ and µ are not important for predicting system
behaviour. Note, however, that the lack of identifiability of λ and µ has implications
for network inference, as well as inference for rate constants. It is clear that in this
model we cannot know from experimental data if we have a pure birth or death
process, or a process involving both births and deaths, as it is not possible to know if
λ or µ is zero.∗
∗ This also illustrates another point that is not widely appreciated — the fact that reliable network in-
6 INTRODUCTION TO BIOLOGICAL MODELLING

100 0.2
0.1
80

0
−0.3
−1
60
X

40
20
0

0 1 2 3 4 5

Figure 1.1 Five deterministic solutions of the linear birth–death process for values of λ − µ
given in the legend (x0 = 50).

The problem, of course, is that bacteria don’t vary in number continuously and
deterministically. They vary discretely and stochastically. Using the techniques that
will be developed later in this book, it is straightforward to understand the stochastic
process associated with this model as a Markov jump process, and to simulate it on a
computer. By their very nature, such stochastic processes are random, and each time
they are simulated they will look different. In order to understand the behaviour of
such processes it is therefore necessary (in general) to study many realisations of the
process. Five realisations are given in Figure 1.2, together with the corresponding
deterministic solution.
It is immediately clear that the stochastic realisations exhibit much more inter-
esting behaviour and match much better with the kind of experimental data one is
likely to encounter. They also allow one to ask questions and get answers to issues
that can’t be addressed using a continuous deterministic model. For example, ac-
cording to the deterministic model, the population size at time t = 2 is given by
X(2) = 50/e2 ≃ 6.77. Even leaving aside the fact that this is not an integer, we see
from the stochastic realisations that there is considerable uncertainty for the value of
X(2), and stochastic simulation allows us to construct, inter alia, a likely range of
values for X(2). Another quantity of considerable practical interest is the “time to
extinction” (the time, t, at which X(t) first becomes zero). Under the deterministic
model, X(t) never reaches zero, but simply tends to zero as t −→ ∞. We see from
the stochastic realisations that these do go extinct, and that there is considerable ran-

ference is necessarily more difficult than rate-parameter inference, as determining the existence of a
reaction is equivalent to deciding whether the rate of that reaction is zero.
WHY IS STOCHASTIC MODELLING NECESSARY? 7

60
50
40
30
X

20
10
0

0 1 2 3 4 5

Figure 1.2 Five realisations of a stochastic linear birth–death process together with the con-
tinuous deterministic solution (x0 = 50, λ = 3, µ = 4).

domness associated with the time that this occurs. Again, stochastic simulation will
allow us to understand the distribution associated with the time to extinction, some-
thing that simply isn’t possible using a deterministic framework.
Another particularly noteworthy feature of the stochastic process representation is
that it depends explicitly on both λ and µ, and not just on λ − µ. This is illustrated in
Figure 1.3. It is clear that although λ−µ controls the essential “shape” of the process,
λ + µ controls the degree of “noise” or “volatility” in the system. This is a critically
important point to understand — it tells us that if stochastic effects are present in
the system, we cannot properly understand the system dynamics unless we know
both λ and µ. Consequently, we cannot simply fit a deterministic model to available
experimental data and then use the inferred rate constants in a stochastic simulation,
as it is not possible to infer the stochastic rate constants using a deterministic model.
This has important implications for the use of stochastic models for inference from
experimental data. It suggests that given some data on the variation in colony size
over time, it ought to be possible to get information about λ − µ from the overall
shape of the data, and information about λ + µ from the volatility of the data. If we
know both λ − µ and λ + µ, we can easily determine both λ and µ separately. Once
we know both λ and µ, we can accurately simulate the dynamics of the system we
are interested in (as well as inferring network structure, as we could also test to see
if either λ or µ is zero). However, it is only possible to make satisfactory inferences
for both λ and µ if the stochastic nature of the system is taken into account at the
inference stage of the process.
Although we have here considered a trivial example, the implications are broad.
In particular, they apply to the genetic and biochemical network models that much of
8 INTRODUCTION TO BIOLOGICAL MODELLING

λ = 0, µ = 1 λ = 3, µ = 4

60

60
40

40
X

X
20

20
0

0
0 1 2 3 4 5 0 1 2 3 4 5

t t

λ = 7, µ = 8 λ = 10, µ = 11
60

60
40

40
X

X
20

20
0

0 1 2 3 4 5 0 0 1 2 3 4 5

t t

Figure 1.3 Five realisations of a stochastic linear birth–death process together with the con-
tinuous deterministic solution for four different (λ, µ) combinations, each with λ − µ = −1
and x0 = 50.

this book will be concerned with. This is because genetic and biochemical networks
involve the interaction of integer numbers of molecules that react when they collide
after random times, driven by Brownian motion. Although it is now becoming in-
creasingly accepted that stochastic mechanisms are important in many (if not most)
genetic and biochemical networks, routine use of stochastic simulation in order to
understand system dynamics is still not as ubiquitous as it might be.† This could be
because inference methods regularly used in practice work by fitting continuous de-
terministic models to experimental data. We have just seen that such methods cannot
in general give us reliable information about all of the parameters important for de-
termining the stochastic dynamics of a system, and so stochastic simulation cannot
be done reliably until we have good methods of inference for stochastic models. It
turns out that it is possible to formalise the problem of inference for stochastic ki-
netic models from time-course experimental data, and this is the subject matter of
the latter chapters. However, it should be pointed out at the outset that inference for
stochastic models is an order of magnitude more difficult than inference for deter-
ministic models (in terms of the mathematics required, algorithmic complexity, and
computation time), and is still the subject of a great deal of ongoing research.

† But it is a lot more widespread than when the first edition of this book was written.
CHEMICAL REACTIONS 9
1.4 Chemical reactions
There are a number of ways one could represent a model of a biological system. Bi-
ologists have traditionally favoured diagrammatic schemas coupled with verbal ex-
planations in order to convey qualitative information regarding mechanisms. At the
other extreme, applied mathematicians traditionally prefer to work with systems of
ordinary or partial differential equations (ODEs or PDEs). These have the advantage
of being more precise and fully quantitative, but also have a number of disadvan-
tages. In some sense differential equation models are too low level a description, as
they not only encode the essential features of the model, but also a wealth of accom-
panying baggage associated with a particular interpretation of chemical kinetics that
is not always well suited to application in the molecular biology context. Somewhere
between these two extremes, the biochemist will tend to view systems as networks of
coupled chemical reactions, and it appears that most of the best ways of representing
biochemical mechanisms exist at this level of detail, though there are many ways
of representing networks of this type. Networks of coupled chemical reactions are
sufficiently general that they can be simulated in different ways using different algo-
rithms depending on assumptions made about the underlying kinetics. On the other
hand, they are sufficiently detailed and precise that once the kinetics have been spec-
ified, they can be used directly to construct full dynamic simulations of the system
behaviour on a computer.
A general chemical reaction takes the form
m1 R1 + m2 R2 + · · · + mr Rr −→ n1 P1 + n2 P2 + · · · + np Pp ,
where r is the number of reactants and p is the number of products. Ri represents
the ith reactant molecule and Pj is the jth product molecule. mi is the number
of molecules of Ri consumed in a single reaction step, and nj is the number of
molecules of Pj produced in a single reaction step. The coefficients mi and nj are
known as stoichiometries. The stoichiometries are usually (though not always) as-
sumed to be integers, and in this case it is assumed that there is no common factor
of the stoichiometries. That is, it is assumed that there is no integer greater than one
which exactly divides each stoichiometry on both the left and right sides. There is
no assumption that the Ri and Pj are distinct, and it is perfectly reasonable for a
given molecule to be both consumed and produced by a single reaction.‡ The reac-
tion equation describes precisely which chemical species§ react together, and in what
proportions, along with what is produced.
In order to make things more concrete, consider the dimerisation of a protein P .
This is normally written
2P −→ P2 ,
as two molecules of P react together to produce a single molecule of P2 . Here P
‡ Note that a chemical species that occurs on both the left- and right-hand sides with the same stoichiom-
etry is somewhat special, and is sometimes referred to as a modifier. Clearly the reaction will have no
effect on the amount of this species. Such a species is usually included in the reaction because the rate
at which the reaction proceeds depends on the level of this species.
§ The use of the term “species” to refer to a particular type of molecule will be explained later in the
chapter.
10 INTRODUCTION TO BIOLOGICAL MODELLING
has a stoichiometry of 2 and P2 has a stoichiometry of 1. Stoichiometries of 1 are
not usually written explicitly. Similarly, the reaction for the dissociation of the dimer
would be written
P2 −→ 2P.
A reaction that can happen in both directions is known as reversible. Reversible re-
actions are quite common in biology and tend not to be written as two separate reac-
tions. They can be written with a double-headed arrow such as
2P ⇋ P2 or 2P ←→ P2 or 2P ⇐⇒ P2 .
If one direction predominates over the other, this is sometimes emphasised in the
notation. So if the above protein prefers the dimerised state, this may be written
something like
2P ⇀ P2 or 2P −→ P2 .
↽ ←
It is important to remember that the notation for a reversible reaction is simply a
convenient shorthand for the two separate reaction processes taking place. In the
context of the discrete stochastic models to be studied in this book, it will not usually
be acceptable to replace the two separate reactions by a single reaction proceeding
at some kind of overall combined rate.

1.5 Modelling genetic and biochemical networks

Before moving on to look at different ways of representing and working with sys-
tems of coupled chemical reactions in the next chapter, it will be helpful to end this
chapter by looking in detail at some basic biochemical mechanisms and how their
essential features can be captured with fairly simple systems of coupled chemical
reactions. Although biological modelling can be applied to biological systems at a
variety of different scales, it turns out that stochastic effects are particularly impor-
tant and prevalent at the scale of genetic and biochemical networks, and these will
therefore provide the main body of examples for this book.

1.5.1 Transcription (prokaryotes)

Transcription is a key cellular process, and control of transcription is a fundamen-
tal regulation mechanism. As a result, virtually any model of genetic regulation is
likely to require some modelling of the transcription process. This process is much
simpler in prokaryotic organisms, so it will be helpful to consider this in the first
instance. Here, typically, a promoter region exists just upstream of the gene of inter-
est. RNA-polymerase (RNAP) is able to bind to this promoter region and initiate the
transcription process, which ultimately results in the production of an mRNA tran-
script and the release of RNAP back into the cell. The transcription process itself is
complex, but whether it will be necessary to model this explicitly will depend very
much on the modelling goals. If the modeller is primarily interested in control and
the downstream effects of the transcription process, it may not be necessary to model
transcription itself in detail.
MODELLING GENETIC AND BIOCHEMICAL NETWORKS 11

Figure 1.4 Transcription of a single prokaryotic gene.

The process is illustrated diagrammatically in Figure 1.4. Here, g is the gene of

interest, p is the upstream promoter region, and r is the mRNA transcript of g. A very
simple representation of this process as a system of coupled chemical reactions can
be written as follows:

p + RNAP −→ p · RNAP
p · RNAP −→ p + RNAP + r.

As discussed, the second reaction is really the end result of a very large number of
reactions. It is also worth emphasising that the reactions do not represent a closed
system, as r appears to be produced out of nothing. In reality, it is created from other
chemical species within the cell, but we have chosen here not to model at such a
fine level of detail. One detail not included here that may be worth considering is
the reversible nature of the binding of RNAP to the promoter region. It is also worth
noting that these two reactions form a simple linear chain, whereby the product of
the first reaction is the reactant for the second. Indeed, we could write the pair of
reactions as

p + RNAP −→ p · RNAP −→ p + RNAP + r.

It is therefore tempting to summarise this chain of reactions by the single reaction

p + RNAP −→ p + RNAP + r,

and this is indeed possible, but is likely to be inadequate for any model of regulation
or control where the intermediary compound p · RNAP is important, such as any
model for competitive binding of RNAP and a repressor in the promoter region.
If modelling the production of the entire RNA molecule in a single step is felt to be
an oversimplification, it is relatively straightforward to model the explicit elongation
of the molecule. As a first attempt, consider the following model for the transcription
12 INTRODUCTION TO BIOLOGICAL MODELLING
of an RNA molecule consisting of n nucleotides.

p + RNAP −→ p · RNAP
p · RNAP −→ p · RNAP · r1
p · RNAP · r1 −→ p · RNAP · r2
.. ..
. −→ .
p · RNAP · rn−1 −→ p · RNAP · rn
p · RNAP · rn −→ p + RNAP + r.

This still does not model the termination process in detail; see Arkin, Ross, & McAd-
ams (1998) for details of how this could be achieved. One problem with the above
model is that the gene is blocked in the first reaction and is not free for additional
RNAP binding until it is released again after the last reaction. This prevents concur-
rent transcription from occurring. An alternative would be to model the process as
follows:

p + RNAP −→ p · RNAP
p · RNAP −→ p + RNAP · r1
RNAP · r1 −→ RNAP · r2
.. ..
. −→ .
RNAP · rn−1 −→ RNAP · rn
RNAP · rn −→ RNAP + r.

This model frees the gene for further transcription as soon as the transcription process
starts. In fact, it is probably more realistic to free the gene once a certain number
of nucleotides have been transcribed, and this is easily incorporated into the above
model. Another slightly undesirable feature is that it does not prevent one RNAP
from “overtaking” another during concurrent transcription (but this is not usually
particularly important, nor is it very easy to fix).

1.5.2 Eukaryotic transcription (a very simple case)

The transcription process in eukaryotic cells is rather more complex than in prokary-
otes. This book is not an appropriate place to explore the many and varied mecha-
nisms for control and regulation of eukaryotic transcription, so we will focus on a
simple illustrative example, shown in Figure 1.5. In this model, there are two tran-
scription factor (TF) binding sites upstream of a gene, g. Transcription factor TF1
reversibly binds to site tf1, and TF2 reversibly binds to tf2, but is only able to bind
if TF1 is already in place. Also TF1 cannot dissociate if TF2 is in place. The tran-
scription process cannot initiate (starting with RNAP binding) unless both TFs are in
place.
MODELLING GENETIC AND BIOCHEMICAL NETWORKS 13

Figure 1.5 A simple illustrative model of the transcription process in eukaryotic cells.

We can model this as follows:

g + TF1 ⇋ TF1 · g
TF1 · g + TF2 ⇋ TF2 · TF1 · g
RNAP + TF2 · TF1 · g ⇋ RNAP · TF2 · TF1 · g
RNAP · TF2 · TF1 · g −→ TF2 · TF1 · g + RNAP + r.
Note that we have not explicitly included tf1, tf2, and g separately in the model, as
they are all linked on a DNA strand and hence are a single entity from a modelling
perspective. Instead we use g to represent the gene of interest together with its reg-
ulatory region (including tf1 and tf2). Note that this system, like the previous, also
forms a linear progression of ordered reactions and does not involve a “feedback”
loop of any sort.

1.5.3 Gene regulation (prokaryotes)

Regulation and control are fundamental to biological systems. These necessarily in-
volve feedback and a move away from a simple ordered set of reactions (hence the
term biochemical network). Sometimes such systems are large and complex, but
feedback, and its associated non-linearity, can be found in small and apparently
simple systems. We will look here at a simple control mechanism (repression of a
prokaryotic gene) and see how this can be embedded into a regulatory feedback sys-
tem in a later example.
Figure 1.6 illustrates a model where a repressor protein R can bind to regulatory
site q, downstream of the RNAP binding site p but upstream of the gene g, thus
preventing transcription of g. We can formulate a set of reactions for this process in
the following way:
g+R⇋g·R
g + RNAP ⇋ g · RNAP
g · RNAP −→ g + RNAP + r.
14 INTRODUCTION TO BIOLOGICAL MODELLING

Figure 1.6 A simple prokaryotic transcription repression mechanism.

This set of equations no longer has a natural ordering and hence cannot be read from
top to bottom to go from reactants to products in an obvious way. Each reaction
represents a possible direction for the system to move in. Also note that there are
actually five reactions represented here, as two of the three listed are reversible. The
crucial thing to appreciate here is that from a modelling perspective, g·R is a different
species from g, and so the fact that RNAP can bind to g does not suggest that RNAP
can bind to g · R. Thus, this set of reactions precisely captures the mechanism of
interest; namely that RNAP can bind to g when it is free but not when it is repressed
by R.
At this point it is worth mentioning another critically important genetic regulation
mechanism often used in prokaryotes such as bacteria. Sigma factors (often written
σ-factors) are special proteins which bind to RNAP and allow the RNAP to recognise
the (sigma-factor-specific) promoter region upstream of the gene. Different sigma
factors have different promoter sequences, and hence the regulation of sigma factors
is a key control mechanism. Typically the number of distinct sigma factor proteins is
very small relative to the total number of genes, and therefore sigma factor regulation
is used to turn on and off large numbers of genes which share the same sigma factor.
As a concrete example, the sigma factor SigD in the Gram positive bacterium Bacillus
subtilis is used to turn on a large number of genes relating to cell motility. A simple
model for transcription which includes a sigma factor can be constructed as follows:

SigF + RNAP ⇋ SigF · RNAP

g + SigF · RNAP ⇋ g · SigF · RNAP
g · SigF · RNAP −→ g + SigF · RNAP + r.

It is straightforward to combine such a promoter model with a model for repression

to obtain a combined model of prokaryotic regulation.
MODELLING GENETIC AND BIOCHEMICAL NETWORKS 15
1.5.4 Translation
Translation (like transcription) is a complex process involving several hundred reac-
tions to produce a single protein from a single mRNA transcript. Again, however, it
will not always be necessary to model every aspect of the translation process — just
those features pertinent to system features of interest. The really key stages of the
translation process are the binding of a ribosome (Rib) to the mRNA, the translation
of the mRNA, and the folding of the resulting polypeptide chain into a functional
protein. These stages are easily coded as a set of reactions:
r + Rib ⇋ r · Rib
r · Rib −→ r + Rib + Pu
Pu −→ P.
Here, Pu denotes unfolded protein and P denotes the folded protein. In some sit-
uations, it will also be necessary to model various post-translational modifications
such as phosphorylation. Clearly the second and third reactions are gross simplifica-
tions of the full translation process. Elongation can be modelled in more detail using
an approach similar to that adopted for transcription; see Arkin, Ross & McAdams
(1998) for further details. Folding could also be modelled similarly if necessary.

1.5.5 Degradation
The simplest model for degradation is just
r −→ ∅,
where ∅ is the “empty set” symbol, meaning that r is transformed to nothing (as far
as the model is concerned). A more appropriate model for RNA degradation would
be
r + RNase −→ r · RNase
r · RNase −→ RNase,
where RNase denotes ribonuclease (an RNA-degrading enzyme). Modelling in this
way is probably only important if there is limited RNase availability, but is interesting
in conjunction with a translation model involving Rib, as it will then capture the
competitive binding of Rib and RNase to r.
Models for protein degradation can be handled similarly. Here one would typically
model the tagging of the protein with a cell signalling molecule (such as ubiquitin),
t, and then subsequent degradation in a separate reaction. A minimal model would
therefore look like the following:
P + t −→ P · t
P · t −→ t.
In fact, cell protein degradation machinery is rather complex; see Proctor et al. (2005)
for a more detailed treatment of this problem.
16 INTRODUCTION TO BIOLOGICAL MODELLING
1.5.6 Transport

In eukaryotes, mRNA is transported out of the cell nucleus before the translation
process can begin. Often, the modelling of this process will be unnecessary, but could
be important if the transportation process itself is of interest, if the delay associated
with transportation is relevant, or if the number of available transportation points is
limited. A model for this could be as simple as
rn −→ rc ,
where rn denotes the mRNA pool within the nucleus, and rc the corresponding pool
in the cytoplasm. However, this would not take into account the limited number of
transport points. A more realistic model would therefore be
rn + N −→ rn · N
rn · N −→ rc + N,
where N denotes the set of available mRNA transport points embedded in the outer
shell of the cell nucleus. In fact, this system is very closely related to the Michaelis–
Menten enzyme kinetic system that will be examined in more detail later in the book.
Here, the transport points behave like an enzyme whose abundance limits the flow
from rn to rc .

1.5.7 Prokaryotic auto-regulation

Now that we have seen how to generate very simple models of key processes involved
in gene expression and regulation, we can put them together in the form of a simple
prokaryotic auto-regulatory network.
Figure 1.7 illustrates a simple gene expression auto-regulation mechanism often
present in prokaryotic gene networks. Here dimers of the protein P coded by the
gene g repress their own transcription by binding to a (repressive) regulatory region
upstream of g,

g + P2 ⇋ g · P2 Repression
g −→ g + r Transcription
r −→ r + P Translation
2P ⇋ P2 Dimerisation
r −→ ∅ mRNA degradation
P −→ ∅ Protein degradation.

Notice that this model is minimal in terms of the level of detail included. In particular,
the transcription part ignores sigma factor and RNAP binding, the translation/mRNA
degradation parts ignore Rib/RNase competitive binding, and so on. However, as
we will see later, this model contains many of the interesting features of an auto-
regulatory feedback network. See Bundschuh, Hayot & Jayaprakash (2003) for dis-
cussion of this dimer-autoregulation model in the context of the λ repressor protein
cI of phage-λ in E. coli.
MODELLING GENETIC AND BIOCHEMICAL NETWORKS 17

Figure 1.7 A very simple model of a prokaryotic auto-regulatory gene network. Here dimers
of a protein P coded for by a gene g repress their own transcription by binding to a regulatory
region q upstream of g and downstream of the promoter p.

1.5.8 lac operon

We will finish this section by looking briefly at a classic example of prokaryotic

gene regulation — probably the first well-understood genetic regulatory network.
The genes in the operon code for enzymes required for the respiration of lactose
(Figure 1.8). That is, the enzymes convert lactose to glucose, which is then used as
the “fuel” for respiration in the usual way. These enzymes are only required if there
is a shortage of glucose and an abundance of lactose, and so there is a transcription
control mechanism regulating their production. Upstream of the lac operon there is
a gene coding for a protein which represses transcription of the operon by binding
to the DNA just downstream of the RNAP binding site. Under normal conditions
(absence of lactose), transcription of the lac operon is turned off. However, in the
presence of lactose, the inhibitor protein preferentially binds to lactose, and in the
bound state can no longer bind to the DNA. Consequently, the repression of tran-
scription is removed, and production of the required enzymes can take place. We can
represent this with the following simple set of reactions:

i −→ i + rI
rI −→ rI + I
I + Lactose ⇋ I · Lactose
I +o⇋I ·o
18 INTRODUCTION TO BIOLOGICAL MODELLING

Figure 1.8 Key mechanisms involving the lac operon. Here an inhibitor protein I can repress
transcription of the lac operon by binding to the operator o. However, in the presence of
lactose, the inhibitor preferentially binds to it, and in the bound state can no longer bind to
the operator, thereby allowing transcription to proceed.

o + RNAP ⇋ RNAP · o
RNAP · o −→ o + RNAP + r
r −→ r + A + Y + Z
Lactose + Z −→ Z.
Here i represents the gene for the inhibitor protein, rI the associated mRNA, and
I the inhibitor protein itself. The lac operon is denoted o and is treated as a single
entity from a modelling viewpoint. The mRNA transcript from the operon is denoted
by r, and this codes for all three lac proteins. The final reaction represents the trans-
formation of lactose to something not directly relevant to the regulation mechanism.
Again this system is fairly minimal in terms of the detail included, and all of the
degradation reactions have been omitted, along with what happens to lactose once it
has been acted on by β-galactosidase (Z). In fact, there is also another mechanism we
have not considered here that ensures that transcription of the operon will only occur
when there is a shortage of glucose (as respiration of glucose is always preferred).

1.6 Modelling higher-level systems

We have concentrated so far on fairly low-level biochemical models where the con-
cept of modelling with “chemical reactions” is perhaps most natural. However, it is
important to recognise that we use the notation of chemical reactions simply to de-
scribe things that combine and the things that they produce, and that this framework
can be used to model higher-level phenomena in a similar way. In Section 1.3 the lin-
MODELLING HIGHER-LEVEL SYSTEMS 19
ear birth–death process was introduced as a model for the number of bacteria present
in a colony. We can use our chemical reaction notation to capture the qualitative
structure of this model:
X −→ 2X
X −→ ∅.
The first equation represents “birth” of new bacteria and the second “death”. There
are many possible extensions of this simple model, including the introduction of im-
migration of new bacteria from another source and emigration of bacteria to another
source.
The model represents a “population” of individuals (here the individuals are bac-
teria), and it is possible to extend such models to populations involving more than
one “species”. Consider the Lotka–Volterra predator prey model for two interacting
species:
Y1 −→ 2Y1
Y1 + Y2 −→ 2Y2
Y2 −→ ∅.
Again this is not a real reaction system in the strictest sense, but it is interesting and
useful, as it is the simplest model exhibiting the kind of non-linear auto-regulatory
feedback behaviour considered earlier. Also, as it only involves two species and three
reactions, it is relatively easy to work with without getting lost in detail. Here, Y1
represents a “prey” species (such as rabbits) and Y2 represents a “predator” species
(such as foxes).¶ The first reaction is a simple representation of prey reproduction.
The second reaction is an attempt to capture predator-prey interaction (consumption
of prey by predator, in turn influencing predator reproduction rate). The third reaction
represents death of predators due to natural causes. We will revisit this model in
greater detail in later chapters.
Another widely studied individual level model is the so-called SIR model for dis-
ease epidemiology, where the initials stand for Susceptible, Infected, and Recovered.
The idea is that individuals are initially susceptible to catching a disease from an
infected person. Should they contract the disease, they will make the transition from
susceptible to infected, where they will have the possibility of infecting susceptibles.
Eventually the infected individual will make the transition to the “recovered” cate-
gory, when they will no longer be able to infect susceptibles, but will have immunity
to the disease, and hence will not be themselves any longer susceptible to infection.
Of course for some diseases, this “recovered” category will include individuals who
are in fact dead! In this case, the “R” category is sometimes used to stand for Re-
moved. The simplest variant of this model can be summarised with just two reactions
as
S −→ I −→ R.
¶ Note that the use of reactions to model the interaction of “species” in a population dynamics context
explains the use of the term “species” to refer to a particular type of chemical molecule in a set of
coupled chemical reactions.
20 INTRODUCTION TO BIOLOGICAL MODELLING
There are obviously many variants on this basic model. For example, some individ-
uals may develop immunity without ever becoming infectious (S −→ R) and some
recovered individuals may lose their immunity (R −→ S), etc. Another commonly
studied variant is the SEIR model which introduces an additional category, Exposed,
representing individuals who have been infected with the disease but are not yet
themselves infectious:
S −→ E −→ I −→ R.

1.7 Exercises
1. Write out a more detailed and realistic model for the simple auto-regulatory net-
work considered in Section 1.5.7. Include a sigma factor, RNAP binding, Rib/R-
Nase competitive binding, and so on.
2. Consider the lac operon model from Section 1.5.8.
(a) First add more detail to the model, as in the previous exercise.
(b) Look up the β-galactosidase pathway and add detail from this to the model.
(c) Find details of the additional regulation mechanism mentioned, which ensures
lactose is only respired in an absence of glucose, and try to incorporate that
into the model.

1.8 Further reading

See Wilkinson (2009) for a review of stochastic modelling approaches with applica-
tions to systems biology, which provides a more comprehensive survey of the recent
literature than this text. See Bower & Bolouri (2000) for more detailed informa-
tion on modelling, and the different possible approaches to modelling genetic and
biochemical networks. Kitano (2001) gives a more general overview of biological
modelling and systems biology. McAdams & Arkin (1997) and Arkin et al. (1998)
explore biological modelling in the context of the discrete stochastic models we
will consider later. The lac operon is discussed in many biochemistry texts, includ-
ing Stryer (1988), and is modelled in Santillán, Mackey & Zeron (2007). Elowitz,
Levine, Siggia & Swain (2002) and Swain, Elowitz & Siggia (2002) were key papers
investigating the sources of stochasticity in gene expression at the single-cell level.
Latchman (2002) is the classic text on eukaryotic gene regulation. The original ref-
erences for the Lotka–Volterra predator-prey models are Lotka (1925) and Volterra
(1926).
The website associated with this (edition of the) book∗ contains a range of links
to on-line information of relevance to the various chapters of the book. I will also
include an errata for any typos which are notified to me. Now would probably be
a good time to have a quick look at it and “bookmark” it for future reference. The
links for this chapter contain pointers to various pertinent Wikipedia pages for further
reading.

∗ URL: http://www.staff.ncl.ac.uk/d.j.wilkinson/smfsb/2e/
 CHAPTER 2

Representation of biochemical networks

2.1 Coupled chemical reactions

As was illustrated in the first chapter, a powerful and flexible way to specify a model
is to simply write down a list of reactions corresponding to the system of interest.
Note, however, that the reactions themselves specify only the qualitative structure
of a model and must be augmented with additional information before they can be
used to carry out a dynamic simulation on a computer. The model is completed by
specifying the rate of every reaction, together with initial amounts of each reacting
species.
Reconsider the auto-regulation example from Section 1.5.7:
g + P2 ⇋ g · P2
g −→ g + r
r −→ r + P
2P ⇋ P2
r −→ ∅, P −→ ∅.
Although only six reactions are listed, there are actually eight, as two are reversible.
Each of those eight reactions must have a rate law associated with it. We will defer
a complete discussion of rate laws until Chapter 6. For now, it is sufficient to know
that the rate laws quantify the propensity of particular reactions to take place and are
likely to depend on the current amounts of available reactants. In addition there must
be an initial amount for each of the five chemical species involved: g ·P2 , g, r, P , and
P2 . Given the reactions, the rate laws, and the initial amounts (together with some as-
sumptions regarding the underlying kinetics, which are generally not regarded as part
of the model), the model is specified and can in principle be simulated dynamically
on a computer.
The problem is that even this short list of reactions is hard to understand on its
own, whereas the simple biologist’s diagram (Figure 1.7) is not sufficiently detailed
and explicit to completely define the model. What is needed is something between
the biologist’s diagram and the list of reactions.

2.2 Graphical representations

2.2.1 Introduction
One way to begin to understand a reaction network is to display it as a pathway
diagram of some description. The diagram in Figure 2.1 is similar to that used by

21
Random documents with unrelated
content Scribd suggests to you:
 CURIOUS THEATRE CUSTOMS IN
PARIS.
The visitor to Paris may witness a kind of theatrical performance
which is strikingly different from any that can be seen in Great
Britain. We refer to the Théâtre des Menus Plaisirs, in the Boulevard
de Strasbourg. Part of the entertainment here consists in certain of
the actors and actresses criticising the performances which are
proceeding upon the stage, from seats in various parts of the house
—pit, circle, and gallery—which they have quietly got into
unobserved by the audience. They assume the rôle of ordinary
spectators who find themselves compelled in the interests of
literature and art to remonstrate in a rather extraordinary manner
against what they see and hear upon the stage; and the surprise of
the uninitiated when the ball is set rolling is considerable.
The manager comes upon the stage and begins a modest speech
upon past successes and future prospects; but he has not far
advanced in his speech when a gentleman rises in the stalls, with
hat in hand, and in the most respectful manner corrects him with
regard to a word which he declares to be ill chosen and misleading,
at the same time obliging the manager with the correct word. Here
another gentleman introduces himself into the dispute, and
complicates matters by a new suggestion, which involves the subject
in inextricable confusion and absurdity. Both gentlemen are
extremely polite, but firm in denying the right of the manager to that
word; and the latter is driven frantic, and retires from the stage
glaring at his antagonists.
Silence for a few seconds succeeds this scene, when suddenly a man
in the front seat of the gallery starts up from his seat with a wild cry,
throws one leg over the gallery, hangs forward suspended from the
railing, and gazes towards the pit entrance of the theatre. He sees
something of absorbing interest, and with another cry he is about to
throw himself over the gallery. The people scream; and then he finds
he has been mistaken; he resumes a normal position, and looking
round upon the audience with a kindly smile, which strangely
contrasts with his late look of anxiety, he asks pardon for
unnecessarily disturbing their composure, and resumes his seat. A
tenor singer now comes upon the stage and commences a song; but
the two critics in the stalls are particular, and take exception to his
style; they do so with manifest regret, but the principles of art must
be attended to. With profuse apologies, and an expressed hope that
he will proceed with his song in the corrected form, the critics
resume their seats. The tenor, at first exasperated, becomes
mollified by the courteous manners of the gentlemen, and begins his
song again; but almost immediately a lady sitting in the front seat of
the circle tells him that he is in danger of dropping his moustache.
This last is the final ‘straw’ on the back of the vocalist, and he retires
in high dudgeon.
By the side of the lady in the circle there sits a meek-looking old
gentleman, who being naturally shocked at the conduct of his wife,
puts on his hat as if to leave the theatre; but the better-half is equal
to the occasion, and knocks his hat over the meek old gentleman’s
eyes, and the meek old gentleman himself back into his seat.
Presently several actresses appear upon the stage, and one of them
commences to sing, with probably a pleasing sympathetic voice; but
such is not the opinion of the lady, who holds the singer up to
ridicule. The vocalist then stops, and engages in a verbal and violent
encounter with her persecutor, who from her place in the ‘circle’
returns the badinage with interest, so that soon the other retires
from the stage vanquished. The victor is now asked herself to sing, a
request with which she readily complies, singing with abundant
action and in good voice an exceedingly catching song, and at the
chorus, giving a royal wave of the hands towards the gallery to join
with her at that point.
The stranger will be surprised to learn that this disturbing element in
the audience, in reality comes from behind the scenes; the lady who
has just sung is the leading member of the company, and the
gentlemen critics are well-known and highly appreciated comedians.
And though the stranger may think that all this is an impromptu
disturbance, it is quite certain that all is rehearsed as carefully as
any play that is put upon the stage. How long such a performance
would secure the favour of a London audience, is doubtful; here,
however, it is an abiding success, is received with immense applause
—the claqueurs or professional applauders being apparently
altogether dispensed with—and the audience is kept in continual
hilarity by the humorous attack and by the instant and witty reply.
Within the Parisian theatres the visitor may derive some amusement
from observing the operations of the claqueurs, who are employed
at the principal establishments to augment the enthusiasm of the
audience. The men who compose this body of professional
applauders appear to belong to the artisan class; they number from
forty to fifty, that is they are about a hundred hands all told. They
occupy the front row of seats in the second or third gallery, so that
to observe them and their movements it is necessary to occupy a
place in one of the galleries. Their leader sits in their midst, ever
ready at the points marked for him by author or manager to give the
signal which ‘brings down the house.’ As the moment arrives when
the bon-mot shall be uttered, the chef breathes upon his hands,
then stretches them slightly upwards, while he at the same time
looks right and left along his ranks. This is equivalent to ‘Attention’
or ‘Prepare to fire a volley.’ Each man is now at the ‘ready,’ and waits
anxiously upon the chef. When the mot is uttered, he brings his
hands together with a frantic wave, and the others simultaneously
with him make a very respectable, even enthusiastic show of
applause. At the end of a song the leader starts the cry Ploo, ploo
(plus, signifying more), in which all join; this, which is equivalent to
our ‘Encore,’ sounds in the stranger’s ears more like hooting than
aught else; but it is no doubt as welcome to the French actor as a
good British cheer is to an English one.
This little army, like all others, has its awkward squad. One evening
at the ‘Renaissance’ we observed the chef to become very uneasy on
account of one who was exceedingly remiss in his duty; not only was
the amount of applause when given small in volume, but once when
the signal was given he entirely neglected to comply with it. This
was gall and wormwood to the leader, who really seemed a very
earnest hard-working man in his profession; so after finishing the
round of applause, he ‘went for’ that awkward man, remonstrated
with him, and even gave him on the spur of the moment, a lesson
on the correct method of clapping hands. After this the pupil shewed
marked improvement, and by the end of the play performed his duty
in such a satisfactory manner as promised well for his future
advancement in this handy profession. The effect of this pernicious
system upon the audience is very different, we should think, from
what was anticipated when it was first organised; for finding that the
applause is supplied by the establishment, just as it supplies
programmes or turns on the gas, the audience feel that they are
relieved from all obligations in the matter, and unless stirred by an
irresistible influence, seldom dream of applauding at all.
 THE RIVAL LAIRDS.
In a recent article on Curling we endeavoured to give a sketch of the
history of this popular Scottish pastime, together with a brief outline
of the mode in which the game is usually played. The following story
of a match between two rival parishes, supposed to have been
played about the beginning of the present century, may give the
reader a further idea of the enthusiasm evoked on the ice whenever
and wherever curlers forgather. Let the non-initiated imagine himself
standing beside a frozen sheet of water, upon which are assembled a
company of men of various ranks from peer to peasant, each striving
to do his best to support the prowess and honour of his rink. The
rink let it be understood is a certain portion of ice, from thirty to
forty yards in length, apportioned off to the players. The players
consist usually of four on each side, and whereas in the well-known
game of grass-bowls, each player is provided with two wooden
bowls which he drives towards a small white ball called the Jack,
each player on the ice has two curling-stones shaped much like a
Gouda cheese—with a handle atop—which he propels or hurls
towards a certain marked spot at each end of the rink, called the
tee; and round each tee is scratched a series of concentric rings
ranging from two to ten or twelve feet in diameter. Standing at one
end of the rink the man whose turn it is to play, waits the bidding of
his director or ‘skip’ who stands at the other end, and then
endeavours to act according to the directions that may be given by
that important personage. Each of the four players on one side plays
alternately against his antagonist, the main object being to send the
stone gliding up the ice so that it may eventually lie within the rings
and as near the tee as possible. Thus, when the ‘end’ is finished, the
side whose stones lie nearest the tee scores so many towards the
game.
Sometimes when the ice is partially thawed the players have
difficulty in hurling their stones all the way to the tee; and
sometimes they fail to get them beyond a transverse mark called the
‘hog-score,’ two-thirds down the rink—in which case the lagging
stone is put off the ice and cannot count for that ‘end.’ Besoms,
however, with which each man is armed, are here of great account,
the laws of the game permitting each player to sweep the ice in
front of an approaching stone belonging to his side, so as to
accelerate its progress, if necessary. The shouts of ‘Sweep, sweep!’
or rather ‘Soop, soop!’ are of continual recurrence, and are
exceedingly amusing to strangers. The skip on each side first directs
his three men and then lastly plays himself. On his generalship in
skipping much depends, his efforts being mainly directed first to get
as many stones as possible near the tee, and then to get his men to
‘guard’ them from being driven off by those of the opposite side. Or
he may direct a player to aim at a certain stone already lying, with a
view to take an angle, or ‘wick’ as it is termed, and so land his own
stone near the tee. This wicking is a very pretty part of the game
and requires great delicacy of play.
The anxiety of the opposing skips is very amusing to watch, and the
enthusiasm of the several players when an unusually good shot is
made, is boundless. A good ‘lead’ or first player, though he is
necessarily debarred from the niceties of the game which fall to the
lot of the subsequent players, is a very important man in the game if
he can place his stones within the circles that surround the tee, or in
familiar parlance, ‘lie within the house.’ Second player’s post is not so
important; but ‘third stone’ is a position given usually to an
experienced player, as he has frequently to either drive off some
dangerous stone belonging to the other side, and himself take its
place; or has to guard a stone of his own side, which though in a
good position may lie open to the enemy. Thus proceeds with
varying fortune this ‘roaring game’ of give and take, stone after
stone being driven along the icy plain, till the skips themselves come
to play and so finish the ‘end.’
With these preliminary remarks we proceed to our tale.
Snow had fallen long and silently over all the high-lying districts of
the south of Scotland. It was an unusually bad year for the sheep-
farmers, whose stock was suffering severely from the protracted
storm and the snow which enveloped both hill and low-lying
pasturage. But while sheep-farmers were thus kept anxiously waiting
for fresh weather, curlers were in their glory, as day after day they
forgathered on the ice and followed up the ‘roaring game.’
The century was young, and the particular year of our story was that
known and spoken of for long afterwards as the ‘bad year.’ In these
days, there was no free-trade to keep down the price of corn or
beef, which during years of bad harvest in Great Britain, or long
periods of frost and snow, rose to famine prices, and were all but
unprocurable by the poorer classes. Oatmeal at half-a-crown a peck
told a sad tale in many a household, and especially on the helpless
children—the bairns.
As we have said, curling had been enjoyed to the full; perhaps there
had even been a surfeit of it, if the real truth were told. Match after
match had been played by parish against parish, and county against
county. Rival rinks of choice players belonging to counties such as
Peebles had challenged those of the neighbouring counties of
Selkirkshire, or even Midlothian. Prizes, consisting of medals or
money, had been gained by various enthusiasts; and last though not
least, matches for suppers of beef and greens—the true curlers’ fare,
had been contested, the reckoning to be paid by the losing rinks.
The benedicts too had played the bachelors, and had as usual,
beaten them.
Country squires had given prizes to be played for by their tenantry
versus adjoining tenantry, and had brought their fur-clad wives and
daughters to the ice to congratulate them on success, or condole
with them on defeat. In short, the sole occupation of the majority of
the adult male rural population of the south of Scotland in the year
of which we speak, seemed to be—curling.
Amongst other matches in the county of Peeblesshire there was one
that yet remained to come off, namely between the parishes of
Tweedsmuir and Broughton. In a series of matches—or bonspiels as
they were termed—between parish and parish, these two had stood
unbeaten. It therefore remained to be seen which parish should beat
the other, and thereby achieve the envied position of champion of
the county.
When the honour of a parish is at stake on the ice, the choice of the
men who are to play, is a matter of very grave import. In a friendly
match between two rinks, a little unskilfulness on the part of one or
more of the players is a very common affair and is comparatively
unheeded: but in a bonspiel between the two best parishes in a
celebrated curling county, the failure or even the occasional
uncertainty of any one man may be fraught with direst
consequences.
Foremost among the promoters of the forthcoming match which was
to decide matters, were Robert Scott laird of Tweedsmuir, and
Andrew Murray laird of Broughton. These worthies had long been
rivals on other than ice-fields, and though on friendly enough terms
at kirk or market were each keenly alive to his own honour and
prowess. Any game, therefore, in which these rival lairds engaged,
was sure to be closely contested; and the result was at all times as
eagerly watched by interested spectators as it was keenly fought by
the rival parties. It is even said that the lairds had been rivals in love
as well as in other sports, the result of which was that Murray had
carried off the lady and Scott had remained a bachelor, with an old
housekeeper named Betty to take charge of him. But as the story of
the love-match was but the ‘clash’ of the country, it may be taken for
what it is worth.
On the morning of the day fixed for the match (which was to come
off at Broughton and to consist of four men on each side), the laird
of Tweedsmuir was early astir, in order to see that the cart which
was to convey his own curling-stones and those of his men to
Broughton—a distance of some half-dozen miles—was ready, and
that the men themselves were prepared to accompany it. The cart
having been duly despatched with the schoolmaster of the parish,
who was to be one of the players, and the shepherd from Talla
Linns, who was to be another, Laird Scott ordered out his gig and
himself prepared to start.
‘Now Betty,’ cried the laird to his old housekeeper, as he proceeded
to envelop himself in his plaid, ‘you’ll see and have plenty of beef
and greens ready by six o’clock, and a spare bed or two; for besides
our own men it’s likely enough I may bring back one or two of the
beaten lads to stop all night.’
‘’Deed laird, tak ye care the Broughton folk dinna get the better o’
you, and beat ye after a’: they tell me they’re grand curlers.’
‘Well Betty, I’m not afraid of them, with Andrew Denholm on my
side.’
Thus assured, the stalwart laird seized the reins and took the road
for Broughton. On his way down the valley of the Tweed he called at
the humble cottage of the said Andrew Denholm, who usually played
the critical part of ‘third stone,’ and was one of his best supporters;
and whose employment, that of a mason, was for the nonce at a
stand-still.
‘What! not ready yet Andrew?’ exclaimed the laird in a tone of
disappointment. ‘Bestir yourself man, or we’ll not be on the ice by
ten o’clock.’
‘I’m no’ gaun’ to the curlin’ the day sir,’ replied Andrew with an air of
dejection.
‘And what for no’?’ inquired the laird with uneasy apprehension. ‘You
know Andrew, my man, the game canna’ go on without you. The
honour of Tweedsmuir at stake too! there’s not another man I would
risk in your place on the ice this day.’
‘Get Wattie Laidlaw the weaver to tak’ my place laird; he’s a grand
curler, and can play up a stane as well as ony man in the parish; the
fact is sir, just now I have na’ the heart even to curl. Gang yer ways
yersell laird, and skip against the laird o’ Broughton, and there’s nae
fear o’ the result: and Wattie can play third stane instead o’ me.’
‘Wattie will play nae third stane for me: come yourself Andrew, and
we’ll try to cheer you up; and you’ll take your beef and greens up
bye wi’ the rink callants and me in the afternoon.’
Denholm was considered one of the best curlers in that part of the
county, and was usually one of the first to be on the ice; to see him,
therefore, thus cast down and listless, filled the laird’s warm heart
with sorrow. He saw there was something wrong. He must rally the
dejected mason.
‘Do you think,’ continued the laird, ‘that I would trust Wattie to play
in your place; a poor silly body that can barely get to the hog-score,
let alone the tee? Na, na Andrew; rather let the match be off than
be beaten in that way.’
Seeing the laird thus determined to carry off his ‘third man’ to the
scene of the approaching conflict, the poor mason endeavoured still
further to remonstrate by a recital of his grievances.
‘Ye ken sir,’ he began, ‘what a long storm it has been. Six weeks
since I’ve had a day at my trade, though I have made a shilling or
two now and again up-bye at the homestead yonder. But wi’ the
price o’ meal at half-a-crown the peck, and no’ very good after a’;
and nineteenpence for a loaf of bread, we’ve had a sair time of it.
But we wadna’ vex oorsels about that, Maggie and me, if we had
meal eneugh to keep the bairns fed. Five o’ them dwining away
before our eyes; it’s been an unco job I assure you, laird. Indeed if it
hadna been for Mag’s sister that’s married upon the grieve o’
Drummelzier, dear knows what would have become of us, wi’ whiles
no a handfu’ o’ meal left in the girnel. Even wi’ the siller to pay for it,
it’s no’ aye to be gotten; and,’ faltered the poor fellow in conclusion,
‘there’s just meal eneugh in the house to-day to last till the morn.’
‘Well, cheer up my man!’ cried the laird; ‘the longest day has an end,
and this storm cannot last much longer. In fact there’s a thaw
coming on or I’m far cheated. There’s a crown to Maggie to
replenish the meal-ark, and get maybe a sup o’ something better for
the bairns. And there’s cheese an’ bread in the gig here that will
serve you and me Andrew, till the beef and greens are ready for us
up-bye in the afternoon. Meanwhile, a tastin’ o’ the flask will no be
amiss, and then for Broughton.’
Thus invigorated and reassured, the mason took his seat beside the
laird, and amid blessings from the gudewife and well-wishings from
the bairns, the two sped on their journey.
Arrived at the pond, they found tees marked, distances measured,
and all in readiness for the play to begin. The usual salutations
ensued. Broughton and Tweedsmuir shook hands all round with
much apparent warmth; and the two sides, of four each, took their
places in the following order:
BROUGHTON. TWEEDSMUIR.
Mr Henderson, schoolmaster,
Wil. Elliot, shoemaker, lead;
lead;
Wattie Dalgleish, shepherd, 2d
Rev. Isaac Stevenson, 2d stone;
stone;
Tam Johnston, blacksmith, 3d Andrew Denholm, mason, 3d
stone; stone;
Laird Murray, skip. Laird Scott, skip.
The play was begun and continued with varying fortune: sometimes
one side scored, sometimes the other. The match was to consist of
thirty-one points; and at one o’clock when a halt was called for
refreshments, the scoring was tolerably even. The frost was
beginning to shew a slight tendency to give way, but this only
nerved the players to further exertions in sweeping up the stones on
the somewhat dulled ice. The scene in the forenoon had been a very
lively one: but as the afternoon approached and the game was
nearing an end, the liveliness was tempered with anxiety, which
amounted almost to pain, as shot after shot was ‘put in’ by one side,
only to be cleverly ‘taken’ by the other. ‘Soop! soop!’ was the
incessant cry of the skips as from their point of vantage they
descried a lagging stone; or ‘Haud up! I tell ye; haud up!’ when from
that same point they beheld one of their players’ stones approaching
with sufficient velocity to do all that was wanted. Anxiety was
nearing a crisis. At half-past three the game stood: Broughton thirty,
Tweedsmuir twenty-nine. The game was anybody’s. Coats had been
cast as needless encumbrances; besoms were clutched with
determined firmness: the skips slightly pale with the terrible
excitement of the occasion, and the stake that was as it were
hanging in the balance: want of nerve on their part to direct, or on
the part of any one man to play, might decide the fate of the day.
The last end had come to be played, and Broughton having won the
previous end, was to lead. The shoemaker’s stone is played, and lies
well over the hog-score in good line with the tee, and on the road to
promotion. Tweedsmuir’s leading man, the schoolmaster, passes the
souter’s stone and lies in ‘the house.’ ‘Well played dominie!’ cries
Laird Scott to his lead. And so proceeds the ‘end’ till it comes to our
friend the mason’s turn to play; the blacksmith having just played his
first stone with but indifferent effect.
‘What do ye see o’ that stane Andrew?’ roars Laird Scott from the
tee, pointing at the same time to the winning stone of the other
side, which, however, was partially ‘guarded.’
‘I see the half o’ t.’
‘Then,’ says the laird, ‘make sure of it: tak it awa’, and if you rub off
the guard there’s no harm done.’
For a moment the mason steadies himself, settles his foot in the
crampet, and with a straight delivered shot shaves the guard and
wicks out the rival stone, himself lying in close to the tee, and
guarded both at the side and in front by stones belonging to his
side.
The effect of such a shot as this, at so critical a period of the game,
was electric, and is not easily to be described. Enthusiasm on the
part of Tweedsmuir, dismay on that of Broughton. But there are yet
several stones to come: the order may again be reversed, and
Andrew’s deftly played shot may be yet taken. We shall see. The
blacksmith, the third player on the Broughton side, follows with his
second stone, and though by adhering to the direction of his skip he
might have knocked off the guard and so laid open Andrew’s winner,
over-anxiety causes him to miss the guard and miss everything.
Thus is his second and last stone unfortunately played for
Broughton.
The mason has his second stone still to play for Tweedsmuir, and
before doing so Laird Scott thus accosts him: ‘Andrew my man, we
are lying shot now; we want but another to be game; and for the
honour o’ Tweedsmuir I am going to give you the shot that will give
it to us: do ye see this port?’ pointing to an open part of the ice (in
curling phraseology a port) to the left of the tee, with a stone on
each side.
‘I see the port sir.’
‘Well then,’ continued the laird, ‘I want you to fill that port; lay a
stone there Andrew, and there’s a lade o’ meal at your door to-
morrow morning.’
The stone is raised just for one instant with an easy backward sweep
of hand and arm, and delivered with a twist that curls it on and on
by degrees towards the spot required. Not just with sufficient
strength perhaps, but aligned to the point. In an instant the skip is
master of the situation. ‘Soop lads! O soop! soop her up—s-o-o-o-p
—there now; let her lie!’ as the stone curls into the ‘port,’ and lies a
provoking impediment to the opposite players. The pressure on
players of both sides is now too great to admit of many outward
demonstrations. Stern rigour of muscle stiffens every face as the two
skips themselves now leave the tee and take their places at the
other end. The silence bodes a something that no one cares to
explain away, so great is the strain of half-hope half-fear that
animates every breast.
Laird Murray is directed by his adviser at the tee (the blacksmith) to
break-off the guard in front, but misses. Scott his antagonist, by a
skilfully played stone, puts on another guard still, in order to avoid
danger from Laird Murray’s second and last stone. One chance only
now apparently remains for the laird of Broughton, who requires but
one shot to reverse the order of things and retrieve the game, and
he tries it. It is one of those very difficult shots known amongst
curlers as an outwick. A stone of his side has lain considerably to the
right of the tee short of it, which if touched on the outer side might
be driven in towards the centre and perhaps lie shot. The inwick
would be easier, but that the stone is unfortunately guarded for that
attempt. He knows that Denholm’s first stone still lies the shot, and
is guarded both in front and at the side; and that with another,
Tweedsmuir will be thirty-one and game. The shot is risked—after
other contingencies have been duly weighed—but without the
desired effect: the outlying stone is certainly touched, which in itself
was a good shot, but is not sufficiently taken on the side to produce
the desired effect. The laird of Broughton pales visibly as the shot is
missed, and mutters something between his clenched teeth anything
but complimentary to things in general.
The last stone now lies by the foot of our Tweedsmuir laird, who
calmly awaits the word of direction from Andrew at the other end.
‘Laird!’ shouts the anxious mason, ‘there’s but the one thing for it,
and I’ve seen ye play a dafter-like shot. What would ye say to try an
inwick aff my last stane and lift this ane a foot?’ pointing to a stone
of his side which lay near, though still not counting; ‘that would give
us another shot, and the game!’
‘Well Andrew, that’s why I asked you to fill the port, for I saw what
they didna see, that a wick and curl-in would be left: I think it may
be done. At any rate I can but try.’
Silence reigns o’er the rink: the sweepers on each side stand in
breathless suspense: the wick taken, as given by Andrew in advice
to the Laird, may proclaim Broughton beaten and Tweedsmuir the
champion parish of the county!
‘Stand back from behind, and shew me the stone with your besom,
Andrew; there.’
The suspense is soon broken, the last stone has sped on its mission,
the wick has been taken, a stone on Laird Scott’s side that was lying
farther from the tee than one of the opponents’, is ‘lifted’ into second
place, which with the mason’s winner makes exactly the magic score
of thirty-one! Like the thaw which after this long-continued storm
will be welcomed by man and beast alike, so does the thaw now
melt the frozen tongues of the players. Hats fly up in frenzy of
delight, and the phenomenon is witnessed (only to be witnessed on
ice) of a Scottish laird and his humble tenant in ecstatic embrace.
Flasks are produced, hands shaken by rivals as well as by friends—
though chiefly by friends: preparations are made to carry home the
paraphernalia of the roaring game: and while Betty congratulates
the laird and his guests on their victory, there is happiness in store
for Andrew Denholm, whose prowess so notably contributed to
secure the honour of Tweedsmuir.
 AN IRISH COUNTRY FUNERAL.
The difference between English and Irish as regards the funeral
customs of the peasantry in both countries is great. To have a large
assemblage at the ‘berrin’ is among the latter an object of ambition
and pride to the family; and the concourse of neighbours, friends,
and acquaintances who flock from all parts to the funeral is often
immense. Even strangers will swell the funeral cortège, and will
account for doing so by saying: ‘Sure, won’t it come to our turn
some day, and isn’t a big following—to do us credit at our latter end
—what we’d all like? So why shouldn’t we do what is dacent and
neighbourly by one another?’
What a contrast there is between a quiet interment in an English
country parish, attended only by the household of the departed, and
the well-remembered scenes in the churchyard of Kilkeedy, County
Limerick!
Here, in days gone by, a funeral was a picturesque and touching
sight. There was something very weird and solemn in the sound of
the ‘keen,’ as it came, mournful and wild upon the ear, rising and
falling with the windings of the road along which the vast procession
moved. In the centre was the coffin, borne on the shoulders of
relatives or friends, and followed by the next of kin. Outside the
churchyard gate, where was a large open space, there was a halt.
The coffin was laid reverently on the ground, the immediate relatives
of the dead kneeling round it.
And now on bended knees all in that vast assemblage sink down.
Every head is bowed in prayer—the men devoutly uncovered—every
lip moves; the wail of the keeners is hushed; you could hear a pin
drop among the silent crowds. It is a solemn and impressive pause.
After a few minutes the bearers again take up their burden and carry
it into the churchyard, when after being three times borne round the
church, it is committed to its final resting-place.
Years have passed since these scenes were witnessed by the writer
of these pages. The old familiar church has been pulled down (a
new one built on a neighbouring site), and nought of it remains but
the ivy-clad tower and graceful spire left standing—that ‘ivy-mantled
tower,’ where the sparrow had found her a house and the swallow a
nest; whose green depths in the still eventide were made vocal by
the chirpings and chatterings of its feathered inhabitants—the
sparrows fluttering fussily in and out, and after the manner of their
kind, closing the day in noisy gossip before subsiding into rest and
silence. Here too were to be found owls, curiously light—soft masses
of feathers with apparently no bodies to speak of, who captured by
the workmen while clipping the ivy, were brought up, all dazed-
looking and sleepy, to be admired and wondered at by the rectory
children, and finally restored tenderly to their ‘secret bower!’
A funeral scene similar to that just described forms the subject of
one of the illustrations in Lady Chatterton’s Rambles in the South of
Ireland, sketched by herself. She had stopped to make a drawing of
the beautiful ruins of Quin Abbey in the County Clare, when the wail
of an approaching funeral came floating on the breeze, and the
melancholy cadence was soon followed by the appearance of the
usual concourse of country people. Their figures scattered about in
groups, and the coffin in the foreground, enter with very picturesque
effect into the sketch.
When the funeral is over, those who have attended it disperse
through the churchyard; and any having friends buried there betake
themselves to their graves to pray and weep over them. The wild
bursts of grief and vehement sobbing, even over moss-grown graves
whose time-stained headstones bear witness to the length of time
their occupants have slept beneath, would surprise those who are
unfamiliar with the impulsive and demonstrative Irish nature.
An old man sitting beside a grave was rocking himself to and fro,
and wiping his eyes with a blue cotton handkerchief, while, rosary in
hand, he prayed with extraordinary fervour.
‘It’s my poor old wife is lying here,’ he said; ‘the heavens be her bed!
God rest her soul this day! Many’s the long year since she wint from
me, poor Norry, and left me sore and lonesome! She was well on in
years then, though the childer were young; for we were married a
long time before there was any. The neighbours were all at me to
marry again, if it was only for one to wash the shirt or knit the
stocking for me, or to keep the weenochs from running wild about
the roads while I was away at my work earning their bit. But I
couldn’t give in to the notion. I was used to my poor Norry, and the
thoughts of a stranger on the floor was bitter to my heart. Ah, it’s a
sore loss to a man in years when his old wife is took from him! The
old comrade he’s had so long; that understands every turn of him,
and knows his humours and his fancies; and fits him as easy and
comfortable as an old shoe. A man might get a new one—and
maybe more sightly to look at than the one that’s gone—but dear
knows, ’twould be at his peril! As likely as not, she’d fret him and
heart-scald him, and make him oneasy day and night, just blistering
like new leather! The old wife is like the shoe he’s used to, that will
lie into his foot. Stretching here and giving there, and coming, by
constant wearing, to fit, as easy and souple as the skin itself, into th’
exactness of every bump and contrairy spot! For there’s none of us,’
continued the old man, who seemed to be a bit of a moralist, ‘that
hasn’t our tendher places and our corns and oddities in body and
mind, God help us! Some more and some less, according. And
there’s no one can know where them raw spots lie, or how to save
’em from being hurt, like the loving crathur that’s been next us
through the long years, in rain and shine. So yer honours,’ he added,
getting up with a last sorrowful look at his wife’s grave, ‘I wouldn’t
hearken to the neighbours, and take a strange comrade. And after a
while a widow sister o’ mine came to live with me and to care my
poor orphans; but my heart is still with my poor Norry here in the
clay!’
There was another loving couple in the same neighbourhood, whose
apparently impending separation by death caused much sympathy
among their friends. The man was a farmer, and owing to his
industry and good conduct, he and his young wife were in
comfortable circumstances and well to do. They were devoted to
each other. When he was attacked with the severe illness that
threatened his life, she nursed him night and day until she was
wasted to a shadow, and looked from anxiety and want of sleep
almost as corpse-like as he did. Her misery when the doctors
pronounced the case hopeless was dreadful to witness. The poor
fellow’s strength was, they said, nearly exhausted, his illness had
lasted so long; so that his holding out was considered impossible.
Things were in this state, and the sufferer’s death daily expected,
when we were called away from the place, to pay a distant visit. On
our return home after some weeks’ absence, one of the first persons
we saw was young Mrs D—— dressed in the deepest widow’s weeds
—a moving mass of crape.
It was on a Sunday morning going to church; she was walking along
the road before us, stepping out with wonderful briskness, we
thought, considering her very recent bereavement. We had to
quicken our pace to come up with her, and said when we did so: ‘We
are so sorry for you, so very sorry! You have lost your husband.’
‘Thank you kindly; you were always good,’ she said, lifting up her
heavy crape veil from off a face radiant with smiles. ‘He isn’t dead at
all, glory be to God! an’ ’tis recovering beautiful he is. The doctor
says if he goes on gettin’ up his strength as he’s doing the last
fortnight, he’ll soon be finely; out and about in no time.—Oh, the
clothes, is it? Sure ’twas himself, the dear man, bought them for me!
When he was that bad there wasn’t a spark of hope, he calls me
over to him, an’ “Katie my heart,” sez he, “I’m going from you. The
doctors have gave me up, and you’ll be a lone widow before long,
my poor child. And when I’m gone, jewel, and you’re left without a
head or provider, there’ll be no one in the wide world to give you a
stitch of clothes or anything conformable. So I’ll order them home
now, darlin’, the best that can be got for money; for I’d like to leave
you dacent and respectable behind me.” And your honours,’ she
went on, ‘so he did. Two golden guineas he gev for the bonnet; and
as for the gownd, ladies dear, only feel the stuff that’s in it, and ye
may guess what that cost. And beautiful crape, no end of a price!—
every whole thing the hoight of good quality—top lot of the shop,
and no stint.—Well,’ she continued, ‘there they all were in the chest.
And sure when himself got well we thought it a sin and a shame to
let lovely clothes like these lie by without wearing ’em—to be ruined
entirely and feed the moths—after they costing such a sight of
money too. So he made me put them on; and a proud man himself
was this morning, and a happy, seeing me go out the door so grand
and iligant—the best of everything upon me!’
There was something absurd, almost grotesque, in the self-
conscious complacent way in which the young woman gazed
admiringly down on her lugubrious finery; tripping off exulting and
triumphant, her manner in curious contrast with the sore woe
associated with those garments—the saddest in which mortal can be
clad.
 MR ASLATT’S WARD.
IN FOUR CHAPTERS.—CHAPTER IV.
I will pass over the misery of the days that followed; days stretched
by anxiety and suspense to double their ordinary length. The woman
succeeded only too well in proving the truth of her story; and
knowing how useless it would be, Mr Hammond did not attempt to
deny that she was his wife. Nor did he endeavour to justify his
conduct, which was truly inexcusable. Yet in after-years, when our
indignation had cooled, and we were able calmly to reflect upon the
history thus revealed, we could not help pitying the unfortunate
young man. He had not been much past twenty when, on a visit to
Wiesbaden, he had made the acquaintance of a woman several
years older than himself, whose brilliant beauty and fascinating
address had fairly bewitched him. She was a gay adventuress, who,
living by the chances of the gaming-table, and tired of such a
precarious livelihood, had fostered the young man’s passion, and
then condescended to marry him.
Alas! Frederick Hammond had not been long married before he
bitterly regretted the step he had taken. His wife proved the bane of
his life. She had contracted the habit of drinking to excess, and her
intemperance destroyed all hope of happiness in domestic life. Her
husband’s love changed to hatred, and unable to control her vicious
propensities, he deserted her. In one place after another he took
refuge, hoping to elude her search; but again and again she
succeeded in tracking him to his place of concealment, though she
was willing to leave him to himself when he had satisfied her
demand for money. But at last for a long time he heard nothing of
her; and as the months passed into years, the hope sprang up
within him that his wife was either dead, or else had lost all clue to
his whereabouts. Weary of residing abroad, he returned to England,
and finding it difficult to obtain other employment, was glad to
accept the post of village schoolmaster, for he thought the little
country village might prove a secure hiding-place. And here
becoming acquainted with Miss Sinclair, he basely yielded to the
temptation to act as though the hope he cherished that his wife was
dead were already a realised fact. He dared not openly ask Rose’s
hand of her guardian; but he sought by all the means in his power
to win her love, and did not rest till he had won from her a response
to his avowed affection, and gained her consent to a secret
engagement. It was a cruel selfish proceeding, for which his past
misfortunes offered no excuse; and thankful indeed were we that his
scheme of eloping with Rose had been frustrated.
But poor Rose! Bitter indeed was her distress when she found we
had no comfort to give her. The shock was too great for her physical
strength, and ere many hours had elapsed it was evident that a
severe illness would be the consequence. For days she lay tossing in
feverish delirium; whilst we kept anxious watch by her bedside,
much fearing what the issue might be. But our fears were mercifully
disappointed; the fever turned, and soon the much-loved patient
was pronounced out of danger. But the improvement was very
gradual, and after a while almost imperceptible. Extreme exhaustion
was accompanied in Rose’s case by an apathetic indifference to
everything around her, which formed the chief barrier to her
recovery. She felt no desire to get strong again, now that life had no
longer any great attraction for her.
‘If we could only rouse her to take an interest in anything, she would
soon be well,’ the doctor said to me one day.
A possibility of doing so occurred to me at that moment, and I
resolved to try, though I could scarcely hope to succeed. In the
evening, when I was sitting by Rose’s couch, and knew that Mr
Aslatt had gone out, and would not be back for an hour or two, I
said to her gently: ‘I think you feel a little stronger to-day; do you
not, darling?’
A heavy sigh was the only response to my question.
I knelt by her side, and gently drew her head upon my shoulder as I
whispered: ‘I wish you could unburden your heart to me, dear Rose.
Would it not be a relief to tell me the sad thoughts that occupy your
mind?’
No answer but by tears, which I was glad to see, for I knew they
would relieve her heavy heart. After a while, words followed. She
told me how little she cared to get well again; what a dreary blank
life appeared to her, now that he whom she had so loved and
trusted had proved unworthy; how it seemed to her she was of no
use in the world, and the sooner she were out of it the better for
herself and every one else. And a great deal more in the same
strain.
I reminded her of her guardian’s love for her, and his great anxiety
for her recovery, and urged her to try to get well for his sake. But
she only shook her head despondingly. ‘I have never been anything
but a trouble to him,’ she said; ‘he would be happier without me. If I
were out of the way, I daresay he would marry. I used to make plans
for his future as well as for my own, you know; but now everything
will be different.’
‘I do not think Mr Aslatt would have married,’ I ventured to say.
‘Why not?’ asked Rose.
I was silent, and she did not repeat the question.
‘I have a story to tell you, Rose, which I think you may like to hear,’ I
said presently.
‘A story!’ she said in surprise.
‘Yes, darling, a story.’
‘Many years ago, a gentleman was passing through the streets of
Vienna. He was a man about thirty years of age, but he looked older,
for he had known sorrow and disappointment, and life appeared to
him then nought but vanity and vexation of spirit. Yet many would
have envied his position, for he possessed much of what the world
most values. He was walking listlessly along, when his attention was
attracted by a group of musicians, who were performing at the
corner of a square. In the centre of the band stood a pretty little
fair-haired girl about six years old. She was poorly clad. Her tiny feet
were bare, and bleeding from contact with the sharp stones with
which the roads were strewn; and tears were in her large blue eyes
as, in her childish voice, she joined in the song. Her pretty yet
sorrowful face and the plaintive tone in which she sang touched the
stranger’s kind heart. He stood still to watch the group, and when
the song was ended went forward to place some money in the
child’s upturned palm. “Is this your little girl?” he asked the man by
whose side she was standing. He replied in the negative. The little
girl was an orphan, the child of an Englishman, who had formerly
belonged to the band, but who had died some months before,
leaving his little daughter entirely dependent on the good-will of his
late comrades.
‘Well, darling, you must know that they did not object to keeping her
with them, as her appearance was calculated to call forth pity, and
thus increase their earnings. But it was a rough life for the child, and
she suffered from the exposure to all weathers which it entailed. Her
father, who it was believed had seen better days, had never allowed
her to go out with the troop, and had done his utmost to shield her
from hardships. But now there was no help for it; she could not be
kept in idleness. Moved with pity for the child’s hapless lot, the
gentleman inquired where the musicians resided, and returned to his
hotel to consider how he might best serve the little orphan. After
much reflection his resolution was taken. He was a lonely man, with
no near relative to claim his love. His heart yearned with pity for the
desolate child, whose pleading blue eyes and plaintive voice kept
appealing to his compassion, to the exclusion of all other
considerations. He determined to adopt her, and provide for her for
the rest of her life. With this intention he sought the street musicians
on the following day, and easily induced them to commit the child to
his care. After handsomely rewarding the musicians, he took her
away with him that very day, and ever since she has had the first
place in his heart. His loving care for the orphan child brought its
own reward, for in striving to promote the happiness of little Rose he
found his own.’
I was interrupted by a cry from my companion. ‘Rose!’ she exclaimed
excitedly. ‘What are you saying, Miss Bygrave? Tell me—was I—am I
that little child?’
‘You are, darling; and now you know how truly you are the light of
Mr Aslatt’s life. He has no one to care for but you, and you alone can
make him happy.’
‘And I have really no claim upon him, am in no way related to him,
as I thought! I knew I owed him much, but I had no idea to what
extent I was indebted to him. But for his goodness, what should I be
now? Oh, if I had only known this before! How ungrateful I have
been to him, how wayward and perverse! Oh, Miss Bygrave, I
cannot bear to think of it!’
‘Do not trouble about that, dear,’ I said, trying to soothe her, for her
agitation alarmed me; ‘it is all forgiven and forgotten by Mr Aslatt.’
‘But I shall never forgive myself,’ she exclaimed passionately. ‘To
think that I have been receiving everything from him for years, living
upon his bounty, and yet making no return, evincing no gratitude,
taking all his kindness as a matter of course, just because I
imagined I was dear to him for my parents’ sake!’
‘Nay; you are too hard upon yourself, dear Rose,’ I said gently. ‘To a
certain extent you have been grateful to him; you have again and
again acknowledged to me your sense of his goodness; and now
that you know all, you will clearly prove your gratitude, I have no
doubt.’
‘But how?’ exclaimed Rose. ‘How can I express—how can I shew my
deep sense of all that I owe him?’
‘In the first place, by getting well as soon as possible, and by letting
him see that you once more take an interest in life. For his sake, I
know you will strive to bear bravely a trial, the bitterness of which he
fully appreciates. And Rose, I must beg you not to attempt to
express to Mr Aslatt your sense of indebtedness. He feels a morbid
shrinking from hearing such words from your lips, and has implored
me—in case I ever revealed to you the secret of your early life, as I
have been led to do this evening—to assure you that you are under
no great obligation to him, for he considers that he has been fully
repaid for what he has done for you, by the happiness your
companionship has given him.’
‘But I cannot bear to go on receiving so much from him, and yet
give no expression to my gratitude,’ said Rose.
‘You cannot do otherwise,’ I replied; ‘unless you wish to make him
very unhappy, and that would be a poor return for all his goodness.
Do all you can to please him; be as bright and cheerful as possible;
but do not, I beseech you, let him see that you labour under a sense
of painful obligation to him.’
‘I will act as you desire,’ said Rose. ‘But is there really no other way
in which I can prove my gratitude?’
‘Not at present,’ I replied. ‘But perhaps at some future time you may
be able to give him what he will consider worth far more than all he
has ever bestowed upon you; but it would not be acceptable to him
if it proceeded only from the promptings of gratitude.’
‘I do not understand you,’ said Rose, though her cheek flushed.
‘Perhaps you may some day,’ I answered. ‘But now, darling, you
must be still, and not talk any more, else I am afraid you will not be
so well to-morrow.’
I had hard work to persuade her to be quiet, and though after a
time she refrained from talking in obedience to my repeated
injunctions, I could see her thoughts were dwelling on the
communication I had made to her. Only good results, however,
followed from the excitement of that evening. There was a tinge of
pink on Rose’s delicate cheek the next day; her countenance was
brighter, and her manner more animated than we had seen it for
some time. Mr Aslatt was delighted at the change, and encouraged
by it, he began to talk to Rose of the plans he had formed for taking
her to Italy as soon as she felt strong enough to travel. He was
overjoyed to find that she made no objection to his proposal, but
even entered cheerfully into his plans, and declared that she should
be quite ready to start in the course of a few weeks. And so it
proved, for she gained strength with a rapidity which shewed the
truth of the doctor’s words, that she only needed to be roused in
order to get well.
We started for the continent at the end of October. It was thought
that residence abroad during the winter months would promote
Rose’s restoration to health, and afford that diversion of mind which
was so desirable after the trying experience she had passed through.
The result was most satisfactory. There was no return of the
apathetic melancholy which had been so distressing to witness; and
her enjoyment of the various entertainments her kind friend
provided for her was unassumed. I began to hope that, after all, her
attachment to Mr Hammond had not been very deep, but merely a
romantic fancy, kindled by the thought of his misfortunes, and
fanned into a flame by the breath of opposition. A thousand little
incidents strengthened this conviction of mine. Every day it became
evident that Rose was learning to appreciate her guardian’s
character more highly than she had done before. She took a growing
delight in his society, and indeed never seemed quite at ease if he
were absent.
When in the spring we returned to England, Rose’s health and spirits
had so completely returned, that she appeared little different from
the radiant girl whose loveliness had charmed me when I first looked
at her, save that her manner was gentler, being marked by a winning
humility and patience which her former bearing had lacked.
I did not long remain at Westwood Hall in the capacity of Rose’s
companion, though I have frequently visited it since as her friend.
One day soon after our return from Italy, she came to me with a
bright and blushing countenance, and whispered that she had a
secret to tell me. I had little doubt what the secret was, and could
therefore help Rose out with her confession, that Mr Aslatt had
asked her to be his wife, and that she had consented, though with
some reluctance, caused by a sense of her unworthiness.
‘I could not do otherwise,’ she said, ‘when he told me that the
happiness of his future life depended upon my answer; though I
know how little I deserve the love he bestows upon me.’
‘But Rose,’ I said, anxious to be relieved of a painful doubt, ‘you
have not, I trust, been led to a decision contrary to the dictates of
your heart? You know nothing would be further from Mr Aslatt’s
desire than that you should sacrifice your own inclinations from a
mistaken notion of his claims upon you. He would not be happy if he
thought you had only consented that you might not make him
unhappy, and not because your own happiness would be promoted
by the union.’
‘I know that,’ murmured Rose, as her cheek took a deeper tint; ‘but
it is not so. I feel very differently towards Mr Aslatt from what I did
when you first knew me. I think him the best and noblest of men,
and I shall be proud and happy to be his wife; only I wish I were
more worthy of him. O Miss Bygrave! I cannot tell you how ashamed
I feel, when I think of the infatuation which led me to deceive so
kind a friend, or how intensely thankful I am that you saved me from
a wicked act which would have caused unspeakable misery for us
both! I pity poor Mr Hammond, and forgive him for the injury he so
nearly inflicted upon me; but I must confess to you that I never
really had such confidence in him or cared for him, as I now care for
and trust the one whose love I have slighted and undervalued so
long.’
It only remains to add that shortly after that terrible scene at the
Priory, Mr Hammond disappeared, and it was thought, went abroad;
but of him and his wretched wife not a scrap of intelligence has ever
reached us.
 THE MONTH:
SCIENCE AND ARTS.
In a lecture at the Royal Institution, Dr Tyndall has made known the
results of a long series of experiments on fog-signals, all involving
more or less of noise, and demonstrating that the noisiest are the
best. Mariners in a fog are helpless: no lights, no cliffs, no towers
can be seen, and they must be warned off the land through their
ears. So in conjunction with the Trinity House and the authorities at
Woolwich, the Professor fired guns of various kinds and sizes, and
very soon found that a short five-and-a-half-inch howitzer with a
three-pound charge of powder produced a louder report than an
eighteen-pounder with the same weight of charge. Thereupon guns
of different forms were constructed, and one among them which had
a parabolic muzzle proved to be the best, that is in throwing the
sound over the sea, and not wasting it to rearward over the land.
Then it was ascertained that fine-grained powder produces a louder
report than coarse-grained; the shock imparted to the air being
more rapid in the one case than in the other.
Experiments made with gun-cotton shewed conclusively that the
cotton was ‘loudest of all;’ and ‘fired in the focus of the reflector, the
gun-cotton clearly dominated over all the other sound-producers.’
The reports were heard at distances varying from two to thirteen
miles and a half.
When the fog clears off, the noisy signals are laid aside and bright
lights all round the coast guide the seaman on his way. Some years
ago the old oil light was superseded by the magneto-electric light,
and this in turn has given place to the dynamo-electric light, which
excels all in brilliance and intensity. In this machine the required
movements are effected by steam or water power; and when the
electric current is thereby generated, it is conducted by wires to a
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about books and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebookgate.com