0% found this document useful (0 votes)

15 views86 pages

Bacterial Molecular Networks Methods and Protocols 1st Edition Jacques Van Helden instant download

The document provides a comprehensive overview of the book 'Bacterial Molecular Networks: Methods and Protocols', edited by Jacques van Helden, Ariane Toussaint, and Denis Thieffry. It focuses on experimental and computational methods for characterizing molecular interaction networks in bacteria, aimed at postgraduate students and researchers in microbiology. The book includes detailed protocols, reviews, and a variety of methods for analyzing and modeling bacterial networks.

Uploaded by

vlendyaboali28

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

15 views86 pages

Bacterial Molecular Networks Methods and Protocols 1st Edition Jacques Van Helden instant download

Uploaded by

vlendyaboali28

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 86

Bacterial Molecular Networks Methods and

Protocols 1st Edition Jacques Van Helden pdf

download

https://ebookgate.com/product/bacterial-molecular-networks-
methods-and-protocols-1st-edition-jacques-van-helden/

Get Instant Ebook Downloads – Browse at https://ebookgate.com

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

Membrane Protein Structure Determination Methods and

Protocols Methods in Molecular Biology Vol 654 1st
Edition. Edition Jean-Jacques Lacapere
https://ebookgate.com/product/membrane-protein-structure-
determination-methods-and-protocols-methods-in-molecular-biology-
vol-654-1st-edition-edition-jean-jacques-lacapere/
ebookgate.com

Hemoglobin Disorders Molecular Methods and Protocols

Methods in Molecular Medicine 1st Edition Ronald L. Nagel

https://ebookgate.com/product/hemoglobin-disorders-molecular-methods-
and-protocols-methods-in-molecular-medicine-1st-edition-ronald-l-
nagel/
ebookgate.com

Molecular Motors Methods and Protocols Methods in

Molecular Biology Vol 392 1st Edition Ann O. Sperry

https://ebookgate.com/product/molecular-motors-methods-and-protocols-
methods-in-molecular-biology-vol-392-1st-edition-ann-o-sperry/

ebookgate.com

Bacterial Secretion Systems Methods and Protocols 2nd ed.

2024 Edition Edition Laure Journet

https://ebookgate.com/product/bacterial-secretion-systems-methods-and-
protocols-2nd-ed-2024-edition-edition-laure-journet/

ebookgate.com
Claudins Methods and Protocols 1st Edition Christina M.
Van Itallie

https://ebookgate.com/product/claudins-methods-and-protocols-1st-
edition-christina-m-van-itallie/

ebookgate.com

Molecular Diagnosis of Cancer Methods and Protocols 2nd

Edition Methods in Molecular Medicine Joseph E. Roulston

https://ebookgate.com/product/molecular-diagnosis-of-cancer-methods-
and-protocols-2nd-edition-methods-in-molecular-medicine-joseph-e-
roulston/
ebookgate.com

Autoimmunity Methods and Protocols Methods in Molecular

Medicine 1st Edition. Edition Andras Perl

https://ebookgate.com/product/autoimmunity-methods-and-protocols-
methods-in-molecular-medicine-1st-edition-edition-andras-perl/

ebookgate.com

Yeast Genetic Networks Methods and Protocols 1st Edition

Julia Marín-Navarro

https://ebookgate.com/product/yeast-genetic-networks-methods-and-
protocols-1st-edition-julia-marin-navarro/

ebookgate.com

Cystic Fibrosis Methods and Protocols Methods in Molecular

Medicine 1st Edition Brian J. Nickoloff

https://ebookgate.com/product/cystic-fibrosis-methods-and-protocols-
methods-in-molecular-medicine-1st-edition-brian-j-nickoloff/

ebookgate.com
METHODS MOLECULAR BIOLOGY
TM
IN

Series Editor
John M. Walker
School of Life Sciences
University of Hertfordshire
Hatfield, Hertfordshire, AL10 9AB, UK

For further volumes:

http://www.springer.com/series/7651
.
Bacterial Molecular Networks

Methods and Protocols

Edited by

Jacques van Helden and Ariane Toussaint

Laboratoire de Bioinformatique des Génomes et des Réseaux (BiGRe),
Université Libre de Bruxelles, Bruxelles, Belgium

Denis Thieffry
Institut de Biologie de l'Ecole Normale Supérieure (IBENS), UMR ENS,
CNRS 8197, INSERM 1024, Paris, France
Editors
Jacques van Helden Ariane Toussaint
Laboratoire de Bioinformatique des Génomes Laboratoire de Bioinformatique des Génomes
et des Réseaux (BiGRe) et des Réseaux (BiGRe)
Université Libre de Bruxelles Université Libre de Bruxelles
Bruxelles, Belgium Bruxelles, Belgium
[email protected] [email protected]

Denis Thieffry
Institut de Biologie de l’Ecole
Normale Supérieure (IBENS)
UMR ENS
CNRS 8197
INSERM 1024
Paris, France
[email protected]

ISSN 1064-3745 e-ISSN 1940-6029

ISBN 978-1-61779-360-8 e-ISBN 978-1-61779-361-5
DOI 10.1007/978-1-61779-361-5
Springer New York Dordrecht Heidelberg London

Library of Congress Control Number: 2011938202

ª Springer Science+Business Media, LLC 2012

All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the
publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013,
USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of
information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified
as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.

Printed on acid-free paper

Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface

Network-based representations have become pervasive in most fields in biology. Focusing

on networks applied to bacteria, this volume provides authoritative descriptions of various
experimental and computational methods enabling the characterization and analysis of
molecular interaction networks.
Intended primarily for postgraduate students and researchers working in the field of
experimental and computational microbiology, this volume combines up-to-date reviews
along with detailed protocols written by the developers of bioinformatics resources (data-
bases and software tools). Each protocol emphasizes the crucial steps and the way to set up
the parameters in order to obtain the best results.
The first section provides an extensive coverage of various experimental and in silico
approaches aiming at the characterization of network components. The second section is
devoted to the presentation of computational approaches to analyze the topology of
molecular networks. The third and last section further introduces a variety of methods
and tools enabling to generate qualitative or quantitative dynamical models of molecular
processes in bacteria.
Altogether, the volume constitutes a practical guide of methods and tools to charac-
terize, retrieve, visualize, analyze, and manipulate molecular networks.

Bruxelles, Belgium Jacques van Helden

Bruxelles, Belgium Ariane Toussaint
Paris, France Denis Thieffry

,
Gijs J.L. Wuite

v
Contents

Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

1 Bacterial Molecular Networks: Bridging the Gap Between Functional

Genomics and Dynamical Modelling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Jacques van Helden, Ariane Toussaint, and Denis Thieffry

PART I EXPERIMENTAL AND IN SILICO APPROACHES TO UNRAVEL

NETWORK COMPONENTS

2 Bacterial Interactomes: From Interactions to Networks. . . . . . . . . . . . . . . . . . . . . . . . 15

Emmanuelle Bouveret and Christine Brun
3 From Bacterial to Microbial Ecosystems (Metagenomics) . . . . . . . . . . . . . . . . . . . . . . 35
Shannon J. Williamson and Shibu Yooseph
4 Prokaryote Genome Fluidity: Toward a System Approach of the Mobilome . . . . . . 57
Ariane Toussaint and Mick Chandler
5 Reticulate Classification of Mosaic Microbial Genomes Using NeAT Website . . . . . 81
Gipsi Lima-Mendez
6 From Metabolic Reactions to Networks and Pathways. . . . . . . . . . . . . . . . . . . . . . . . . 93
Masanori Arita
7 Predicting Metabolic Pathways by Sub-network Extraction . . . . . . . . . . . . . . . . . . . . 107
Karoline Faust and Jacques van Helden
8 Directed Module Detection in a Large-Scale Expression Compendium . . . . . . . . . . 131
Qiang Fu, Karen Lemmens, Aminael Sanchez-Rodriguez, Inge M. Thijs,
Pieter Meysman, Hong Sun, Ana Carolina Fierro, Kristof Engelen,
and Kathleen Marchal
9 Using Phylogenetic Profiles to Predict Functional Relationships . . . . . . . . . . . . . . . . 167
Matteo Pellegrini
10 Extracting Regulatory Networks of Escherichia coli from RegulonDB . . . . . . . . . . . 179
Heladia Salgado, Irma Martı́nez-Flores, Alejandra López-Fuentes,
Jair Santiago Garcı́a-Sotelo, Liliana Porrón-Sotelo, Hilda Solano,
Luis Muñiz-Rascado, and Julio Collado-Vides
11 Browsing Metabolic and Regulatory Networks with BioCyc . . . . . . . . . . . . . . . . . . . 197
Mario Latendresse, Suzanne Paley, and Peter D. Karp

PART II TOPOLOGICAL ANALYSIS OF BACTERIAL NETWORKS

12 Algorithms for Systematic Identification of Small Subgraphs . . . . . . . . . . . . . . . . . . . 219
Joseph Geraci, Geoffrey Liu, and Igor Jurisica
13 The Degree Distribution of Networks: Statistical Model Selection . . . . . . . . . . . . . . 245
William P. Kelly, Piers J. Ingram, and Michael P.H. Stumpf

vii
viii Contents

14 MAVisto: A Tool for Biological Network Motif Analysis . . . . . . . . . . . . . . . . . . . . . . . 263

Henning Schwöbbermeyer and Röbbe W€ u nschiers
15 Using MCL to Extract Clusters from Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
Stijn van Dongen and Cei Abreu-Goodger
16 Protein Complex Prediction with RNSC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
Andrew D. King, Nataša Pržulj, and Igor Jurisica
17 Network Analysis and Protein Function Prediction
with the PRODISTIN Web Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
Anaı̈s Baudot, Ouissem Souiai, and Christine Brun
18 Using the NeAT Toolbox to Compare Networks to Networks,
Clusters to Clusters, and Network to Clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 327
Sylvain Brohée
19 Analyzing Biological Data Using R: Methods for Graphs and Networks. . . . . . . . . . 343
Nolwenn Le Meur and Robert Gentleman

PART III DYNAMICAL MODELLING

20 Detecting Structural Invariants in Biological Reaction Networks . . . . . . . . . . . . . . . . 377

Jörn Behre, Luı́s Filipe de Figueiredo, Stefan Schuster,
and Christoph Kaleta
21 Petri Nets in Snoopy: A Unifying Framework for the Graphical Display,
Computational Modelling, and Simulation of Bacterial Regulatory Networks . . . . . 409
Wolfgang Marwan, Christian Rohr, and Monika Heiner
22 Genetic Network Analyzer: A Tool for the Qualitative Modeling
and Simulation of Bacterial Regulatory Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
Grégory Batt, Bruno Besson, Pierre-Emmanuel Ciron, Hidde de Jong,
Estelle Dumas, Johannes Geiselmann, Regis Monte, Pedro T. Monteiro,
Michel Page, François Rechenmann, and Delphine Ropers
23 Logical Modelling of Gene Regulatory Networks with GINsim. . . . . . . . . . . . . . . . . 463
Claudine Chaouiya, Aurélien Naldi, and Denis Thieffry
24 Modelling the Evolution of Mutualistic Symbioses . . . . . . . . . . . . . . . . . . . . . . . . . . . 481
Maren L. Friesen and Emily I. Jones
25 Modelling the Onset of Virulence in Pathogenic Bacteria . . . . . . . . . . . . . . . . . . . . . . 501
Wilfred D. Kepseu, Frédérique Van Gijsegem,
and Jacques-Alexandre Sepulchre
26 Spatial and Stochastic Cellular Modeling with the Smoldyn Simulator . . . . . . . . . . . 519
Steven S. Andrews

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 543
Contributors

CEI ABREU-GOODGER European Molecular Biology Laboratory, European

Bioinformatics Institute, Hinxton, Cambridge, UK
STEVEN S. ANDREWS Fred Hutchinson Cancer Research Center, Seattle, WA, USA
MASANORI ARITA Department of Computational Biology, Graduate School
of Frontier Sciences, The University of Tokyo, Kashiwanoha, Kashiwa, Japan
GRÉGORY BATT INRIA Paris – Rocquencourt, Domaine de Voluceau, Le Chesnay,
France
ANAÏS BAUDOT Spanish National Cancer Research Centre, Madrid, Spain
JÖRN BEHRE Theoretical Systems Biology, Institute of Food Research (IFR), Norwich
Research Park, Colney Lane, Norwich, United Kingdom
BRUNO BESSON Genostar, Montbonnot, France
EMMANUELLE BOUVERET LISM – CNRS UPR9027, Marseille, France
SYLVAIN BROHÉE ESAT-SCD-SISTA (Bioinformatics group), Katholieke Universiteit
Leuven, Belgium
CHRISTINE BRUN TAGC, INSERM U928, Marseille, France
MICK CHANDLER Laboratoire de Microbiologie et Génétique Moléculaire (UMR
5100), Toulouse, France
CLAUDINE CHAOUIYA Instituto Gulbenkian de Ciência, Oeiras, Portugal
PIERRE-EMMANUEL CIRON Genostar, Montbonnot, France
JULIO COLLADO-VIDES Programa de Genómica Computacional, Centro de Ciencias
Genómicas, Universidad Nacional Autónoma de México, Cuernavaca, Morelos,
Mexico
LUÍS FILIPE DE FIGUEIREDO Chemoinformatics and Metabolism Team, European
Bioinformatics Institute (EBI), Wellcome Trust Genome Campus, Hinxton,
Cambridgeshire, United Kingdom
HIDDE DE JONG INRIA Rhône-Alpes, Montbonnot, St. Ismier, France
ESTELLE DUMAS INRIA Rhône-Alpes, Montbonnot, St. Ismier, France
KRISTOF ENGELEN Centre of Microbial and Plant Genetics,
Katholieke Universiteit Leuven, Heverlee, Belgium
KAROLINE FAUST Research group of Bioinformatics and (eco-)systems biology,
Department of Structural Biology; Microbiology Unit (MICR), Department of
Applied Biological Sciences (DBIT), Vrije Universiteit Brussel, Bruxelles, Belgium
ANA CAROLINA FIERRO Centre of Microbial and Plant Genetics, Katholieke
Universiteit Leuven, Heverlee, Belgium
MAREN L. FRIESEN Department of Molecular and Computational Biology,
University of Southern California, Los Angeles, CA, USA
QIANG FU Centre of Microbial and Plant Genetics, Katholieke Universiteit Leuven,
Heverlee, Belgium
JAIR SANTIAGO GARCÍA-SOTELO Programa de Genómica Computacional,
Centro de Ciencias Genómicas, Universidad Nacional Autónoma de México,
Cuernavaca, Morelos, Mexico

ix
x Contributors

JOHANNES GEISELMANN Laboratoire Adaptation et Pathogénie des Microorganismes

(CNRS UMR 5163), Bâtiment Jean Roget, Faculté de Médecine-Pharmacie,
Université Joseph Fourier, La Tronche, France
ROBERT GENTLEMAN Program in Computational Biology, Fred Hutchinson Cancer
Research Center, M2-B876, Seattle, WA, USA
JOSEPH GERACI Ontario Cancer Institute/UHN, Toronto, ON, Canada
MONIKA HEINER Department of Computer Science, Brandenburg
University of Technology, Cottbus, Germany
PIERS J. INGRAM Department of Mathematics, Imperial College, South Kensington
Campus, London, UK
EMILY I. JONES School of Biological Sciences, Washington State University,
Pullman, WA, USA
IGOR JURISICA Ontario Cancer Institute/UHN, Toronto, ON, Canada
CHRISTOPH KALETA Department of Bioinformatics, Friedrich Schiller University Jena,
Jena, Germany
PETER D. KARP SRI International, Menlo Park, CA, USA
WILLIAM P. KELLY Theoretical Systems Biology group, Imperial College, London, UK
WILFRED D. KEPSEU Institut Non Linéaire de Nice, CNRS UMR 6618, Université de
Nice Sophia Antipolis, Valbonne, France
ANDREW D. KING Department of Industrial Engineering and Operations Research,
Columbia University, New York, NY, USA
MARIO LATENDRESSE SRI International, Menlo Park, CA, USA
NOLWENN LE MEUR IRISA, Equipe Symbiose, Université de Rennes I, Rennes, France
KAREN LEMMENS Centre of Microbial and Plant Genetics, Katholieke Universiteit
Leuven, Heverlee, Belgium
GIPSI LIMA-MENDEZ Laboratoire de Bioinformatique des Génomes et des Réseaux
(BiGRe), Université Libre de Bruxelles, Bruxelles, Belgium
GEOFFREY LIU Ontario Cancer Institute/UHN, Toronto, ON, Canada
ALEJANDRA LÓPEZ-FUENTES Programa de Genómica Computacional, Centro de
Ciencias Genómicas, Universidad Nacional Autónoma de México, Cuernavaca,
Morelos, Mexico
KATHLEEN MARCHAL Centre of Microbial and Plant Genetics, Katholieke Universiteit
Leuven, Heverlee, Belgium
IRMA MARTÍNEZ-FLORES Programa de Genómica Computacional, Centro de Ciencias
Genómicas, Universidad Nacional Autónoma de México, Cuernavaca, Morelos,
Mexico
WOLFGANG MARWAN Otto von Guericke University & Magdeburg Centre for Systems
Biology, c/o Max Planck Institute for Dynamics of Complex Technical Systems,
Magdeburg, Germany
PIETER MEYSMAN Centre of Microbial and Plant Genetics, Katholieke Universiteit
Leuven, Heverlee, Belgium
REGIS MONTE INRIA Rhône-Alpes, Montbonnot, St. Ismier, France
PEDRO T. MONTEIRO INRIA Rhône-Alpes, Montbonnot, St. Ismier, France
LUIS MUÑIZ-RASCADO Programa de Genómica Computacional, Centro de Ciencias
Genómicas, Universidad Nacional Autónoma de México, Cuernavaca, Morelos,
Mexico
Contributors xi

AURÉLIEN NALDI TAGC, INSERM U928, Marseille, France

MICHEL PAGE INRIA Rhône-Alpes, Montbonnot, St. Ismier, France
SUZANNE PALEY SRI International, Menlo Park, CA, USA
MATTEO PELLEGRINI Department of Molecular, Cell and Developmental Biology,
University of California, Los Angeles, CA, USA
LILIANA PORRÓN-SOTELO Programa de Genómica Computacional,
Centro de Ciencias Genómicas, Universidad Nacional Autónoma de México,
Cuernavaca, Morelos, Mexico
NATAŠA PRŽULJ Department of Computing, Imperial College London, London, UK
FRANÇOIS RECHENMANN INRIA Rhône-Alpes, Montbonnot, St. Ismier, France
CHRISTIAN ROHR Otto von Guericke University & Magdeburg Centre for Systems
Biology, c/o Max Planck Institute for Dynamics of Complex Technical Systems,
Magdeburg, Germany
DELPHINE ROPERS INRIA Rhône-Alpes, Montbonnot, St. Ismier, France
HELADIA SALGADO Programa de Genómica Computacional, Centro de Ciencias
Genómicas, Universidad Nacional Autónoma de México, Cuernavaca,
Morelos, Mexico
AMINAEL SANCHEZ-RODRIGUEZ Centre of Microbial and Plant Genetics, Katholieke
Universiteit Leuven, Heverlee, Belgium
STEFAN SCHUSTER Department of Bioinformatics, Friedrich Schiller University Jena,
Jena, Germany
HENNING SCHWÖBBERMEYER SunGene GmbH, Gatersleben, Germany
JACQUES-ALEXANDRE SEPULCHRE Institut Non Linéaire de Nice, CNRS UMR 6618,
Université de Nice Sophia Antipolis, Valbonne, France
HILDA SOLANO Programa de Genómica Computacional, Centro de Ciencias
Genómicas, Universidad Nacional Autónoma de México, Cuernavaca, Morelos,
Mexico
OUISSEM SOUIAI TAGC, INSERM U928, Marseille, France
MICHAEL P.H. STUMPF Theoretical Systems Biology group, Imperial College,
London, UK
HONG SUN Centre of Microbial and Plant Genetics, Katholieke Universiteit Leuven,
Heverlee, Belgium
DENIS THIEFFRY Institut de Biologie de l’Ecole Normale Supérieure (IBENS), UMR
ENS, CNRS 8197, INSERM 1024, Paris, France
INGE M. THIJS Centre of Microbial and Plant Genetics, Katholieke Universiteit
Leuven, Heverlee, Belgium
ARIANE TOUSSAINT Laboratoire de Bioinformatique des Génomes et des Réseaux
(BiGRe), Université Libre de Bruxelles, Bruxelles, Belgium
STIJN VAN DONGEN European Molecular Biology Laboratory, European
Bioinformatics Institute, Hinxton, Cambridge, UK
FRÉDÉRIQUE VAN GIJSEGEM Laboratoire Interactions Plantes Pathogènes, UMR 217
INRA/AgroParisTech/UPMC Univ Paris 06, Paris, France
JACQUES VAN HELDEN Laboratoire de Bioinformatique des Génomes et des Réseaux
(BiGRe), Université Libre de Bruxelles, Bruxelles, Belgium
SHANNON J. WILLIAMSON J. Craig Venter Institute, San Diego, CA, USA
RÖBBE W€ uNSCHIERS University of Applied Sciences, Mittweida, Germany
SHIBU YOOSEPH J. Craig Venter Institute, San Diego, CA, USA
.
Chapter 1

Bacterial Molecular Networks: Bridging the Gap Between

Functional Genomics and Dynamical Modelling
Jacques van Helden, Ariane Toussaint, and Denis Thieffry

Abstract
This introductory review synthesizes the contents of the volume Bacterial Molecular Networks of the
series Methods in Molecular Biology. This volume gathers 9 reviews and 16 method chapters describing
computational protocols for the analysis of metabolic pathways, protein interaction networks, and
regulatory networks. Each protocol is documented by concrete case studies dedicated to model bacteria
or interacting populations. Altogether, the chapters provide a representative overview of state-of-the-
art methods for data integration and retrieval, network visualization, graph analysis, and dynamical
modelling.

Key words: Protein interactions, Metabolic pathways, Regulatory networks, Graph motifs,
Clustering, Dynamical modelling, Bacteria, Metagenomics, Computational biology, Bioinformatics,
Systems biology

1. Bacterial Models
for Molecular
Interaction
Networks Bacteria are everywhere, from the most hospitable to the most
hostile environment. They are an important component of the
microfauna that supports all biogeochemical (or nutrient) cycles.
Besides their well-known role as infectious agents, bacteria are
essential for the nutrition of animals (gut microbiome) and
plants (nitrogen fixing and other rhizosphere bacteria) and more
generally, contribute to the well being of water, soil, and air
ecosystems.
For over a century, bacteria have also been fruitful model
systems for deciphering fundamental biological mechanisms.
In particular, Escherichia coli, Bacillus subtilis, and a few others
have been extensively used to define the basic principles of gene
expression and its regulation (1–6).

Jacques van Helden et al. (eds.), Bacterial Molecular Networks: Methods and Protocols, Methods in Molecular Biology,
vol. 804, DOI 10.1007/978-1-61779-361-5_1, # Springer Science+Business Media, LLC 2012

1
2 J. van Helden et al.

Nowadays, with over a thousand bacterial genomes sequenced,

even greater opportunities have opened for experimental or
computational global analysis of metabolism, physiology, and evo-
lution. In particular, the access to comprehensive sets of molecular
components (genes, proteins, regulatory signals) is at the basis of
the development of novel integrative approaches, aiming at under-
standing the function of specific sets of these components (operons,
regulons, metabolic pathways, protein complexes, etc.) in the con-
text of the whole organism. Biological processes rely on the com-
bined activities of molecules interconnected to form a precisely
wired network. The integration of all the cellular processes results
in a complex network, comprising several thousands of molecules
and interactions. A schematic picture of such networks can nowa-
days be obtained from the huge data sets gathered from high-
throughput technologies (“omics”). These biological networks
are usually represented as graphs in the mathematical sense of
the term: a set of nodes (representing biomolecules) and edges
(describing their interactions). The topological structure of these
graphs shapes their dynamical properties, as demonstrated by
the pioneering work of René Thomas on the role of positive and
negative feedback circuits (7, 8).
The availability of genome-scale molecular networks insti-
gated the swift development of sophisticated software tools to
store, query, and visualize the datasets, analyze network struc-
tures, identify components, analyze dynamical properties, gener-
ate relevant, and testable functional predictions. Today, biologists
are confronted to several dozens of software tools that can be
combined to address complementary questions about networks
of interest. However, using these tools requires to understand
not only the biological concepts underlying the datasets, but
also the related algorithmic and statistical aspects. It is easy to
get lost when confronted to the multiple possibilities for analyzing
a given data set: which algorithm should be chosen to answer a
particular question? How should the parameters be tuned? How
should the results be interpreted? In order to put computer tools
in the hands of life sciences researchers, there is a crucial need for
well-documented protocols, illustrated by relevant study cases,
providing general guidelines as well as more detailed comments
about the impact of parametric choices and the potential traps.
This volume targets students and researchers working in the
field of experimental or computational biology. Firstly, it offers
an overview of the state-of-the-art in network biology and its
applications to decipher various types of molecular networks in
bacteria. Furthermore, it is a practical guide for computational
methods and software tools that can be used to build, retrieve,
visualize, analyze, and model biological networks, with concrete
microbial applications (Table 1).
Table 1
Software tools covered by the method chapters

Software tool Website Interface Chapter(s)

BioCyc http://biocyc.org/ Web browser Latendresse et al.
COLOMBOS http://ibiza.biw.kuleuven.be/colombos/ Web browser Fu et al.
DISTILLER http://ibiza.biw.kuleuven.be/DISTILLER/ Web browser Fu et al.
GINsim http://gin.univ-mrs.fr/GINsim/ Java application Chaouiya et al.
GNA http://www-helix.inrialpes.fr/gna Java application Batt et al.
MAVisto http://mavisto.ipk-gatersleben.de Java webstart application Schwobbermeyer and
1

Wunschiers
MCL http://micans.org/mcl/ Unix shell prompt van Dongen and
Abreu-Goodger
NeAT http://rsat.bigre.ulb.ac.be/neat/ Web browser Brohée
Lima-Mendez
Faust and van Helden
netZ http://www3.imperial.ac.uk/theoreticalsystemsbiology/ R package Kelly et al.
data-software
PRODISTIN http://crfb.univ-mrs.fr/webdistin/ Web browser Baudot et al.
R graph methods http://www.r-project.org/ R packages Le Meur and Gentleman
http://www.bioconductor.org/
http://www.graphviz.org/
http://www.boost.org/
RegulonDB http://regulondb.ccg.unam.mx Web browser or MySQL or SOAP Salgado et al.
RNSC http://www.cs.utoronto.ca/~juris/data/rnsc/ Unix shell prompt King et al.
Bacterial Molecular Networks: Bridging the Gap. . .

SmolDyn http://www.smoldyn.org Unix shell prompt Andrews

Snoopy http://www-dssz.informatik.tu-cottbus.de/index.html?/ Stand alone application for Mac OSX, Marwan et al.
software/snoopy.html Windows and Linux
3
4 J. van Helden et al.

2. Experimental
and In Silico
Approaches
to Unravel The first section of the book provides a general overview of
Interactions, Infer currently available experimental and in silico approaches used
Pathways, and to characterize molecular interactions, which constitute the
components of biomolecular pathways and networks.
Build Networks
This section starts with a review by Bouveret and Brun (Chap. 2),
(Part I)
covering the techniques available to microbiologists for studying
protein–protein interactions in bacteria, and the relevance of these
interactions for high-throughput studies, including the deciphering
of various bacterial interactomes.
Beyond description at the level of the organism, recent devel-
opments of high-throughput technologies allow for the direct
sequencing of DNA from bacterial communities, providing access
to the diversity and versatility of microbial populations in various
environments and in function of various physicochemical para-
meters. Williamson and Yooseph (Chap. 3) elaborate upon the
different steps required for such metagenomic analyses, including
DNA extraction, sequencing, and computational analyses.
Microorganisms not only flirt and compete in their natural
niches, but they also constantly exchange biomolecules, including
DNA. Toussaint and Chandler (Chap. 4) summarize the reticulated
relationships between prokaryotic species resulting from the
exchanges of mobile genetic elements, and propose routes to
improve their annotation and to decipher their evolutionary rela-
tionships. Directly related to this review, the method chapter by
Lima-Mendez (Chap. 5) explains how to build a reticulate represen-
tation of the relationships between microbial genomes, where
genetic material is transmitted by both vertical transmission and
horizontal exchanges.
The next chapters deal with in silico resources (software tools
and databases) to build, store, and analyze molecular networks.
As discussed by Arita (Chap. 6), in the case of metabolic networks,
the mapping of molecular interactions onto a graph usually implies
some loss of information, which can be limited by selecting specific
topological transformations that depend on the biological ques-
tions addressed. This is further detailed in the following chapter,
where Faust and van Helden (Chap. 7) present a pragmatic utili-
zation of graph-based representations of metabolic reactions for
inferring metabolic pathways from sets of functionally related
enzyme-coding genes. The method is illustrated with a study case
directly relevant to bacteria, the inferring of metabolic pathways
from operons. The same approach can also be applied to sets of
functionally related enzyme-coding genes obtained from other
sources such as co-expression clusters or phylogenetic profiles.
The protocol by Fu et al. (Chap. 8) introduces COLUMBUS
1 Bacterial Molecular Networks: Bridging the Gap. . . 5

and DISTILLER, two algorithms dedicated to the mining of

public resources for transcriptome data and to infer transcriptional
networks by a combined analysis of transcriptional profiles and
cis-regulatory motifs. Pellegrini et al. (Chap. 9) summarize the
concepts and methods at the basis of phylogenetic profiling. The
approach relies on the analysis of presence/absence of orthologs
across a set of phylogenetically related species, in order to predict
sets of genes that are likely to be involved in related functions.
The section ends with a presentation of novel interfaces
to the most popular databases on bacterial gene regulation
(RegulonDB) and metabolism (BioCyc). The protocol by Salgado
et al. (Chap. 10), describes three ways to retrieve transcriptional
networks from RegulonDB, adapted to various types of utiliza-
tion: simple download, programmatic access via Web services, and
direct access to the SQL database management system. The result-
ing networks can then be used as input to perform different types
of computational analyses, e.g., detect over-represented network
motifs, identify regulatory modules, or characterize degree distri-
butions. The protocol by Latendresse et al. (Chap. 11) introduces
a set of flexible tools specifically designed to visualize bacterial
metabolic and regulatory networks annotated in the BioCyc
database. Although E. coli serves as the main reference, the tools
described should apply to many other bacterial species, which will
be covered by the MetaCyc database once sufficient annotation
becomes available about their regulatory networks.

3. Topological
Analysis of
Bacterial Networks
(Part II) After having collected the components of a given molecular
network, biologists soon realize that it is impossible to intuitively
grasp its global properties. Irrespective of the method used to
obtain the network (case-by-case experiments, high-throughput
technologies, computer-based inference), large-scale biomolecular
networks are complex, intricate, and often noisy. This complexity
stimulated the development of a wide variety of theoretical
approaches to analyze network topology (see, e.g., ref. 9, 10) and
extract relevant modules. This section presents a set of reviews and
method chapters describing a variety of topological approaches:
analysis of degree distributions, detection of network motifs,
network-based clustering. The chapter by Geraci et al. (Chap. 12)
introduces the key concepts and describes standard and novel
approaches to extract significant network characteristics (e.g.,
node degree properties, over-represented motifs) in bacterial
molecular networks. The following method chapters document
specific tools to perform different types of topological analyses.
Kelly et al. (Chap. 13) present an R library enabling to fit alternative
6 J. van Helden et al.

models onto the degree distribution of a given network and to

select the most likely model. This type of analysis challenges
simplistic models that do not adequately fit the data (see also
ref. 11). Schwoebbermeyer and Wuenschiers (Chap. 14) describe
how to use the software MAVisto to mine networks for recurrent
motifs by combining a versatile motif search algorithm with interac-
tive exploration methods and specialized visualization techniques.
The three following chapters present graph-based clustering
algorithms that can be used to extract functional modules from
biological networks. van Dongen et al. (Chap. 15) propose several
protocols and study cases where the algorithm MCL is applied to
partition interaction networks derived from protein sequence
similarities or correlations between gene expression profiles.
King et al. (Chap. 16) apply the algorithm RNSC to identify
densely connected subgraphs that might reveal protein complexes
in protein–protein interaction networks. Baudot et al. (Chap. 17)
illustrate the use of the PRODISTIN tool, which combines hier-
archical clustering and analysis of annotated protein functions in
order to extract consistent clusters of interacting proteins asso-
ciated to particular biological processes (Gene Ontology annota-
tions), thereby enabling novel functional predictions.
The two last chapters of this section present generic software
packages enabling various types of network analysis. The protocol
by Brohée (Chap. 18) explains the use of the Network Analysis
Tools (NeAT), a Web-based software suite combining a variety of
tools to analyze, compare, and classify biological networks (co-
expression, transcriptional networks, GO annotations, etc.), with
a special emphasis on the definition of robust metrics and control
procedures (generation of random networks and network altera-
tions). The chapter by Le Meur and Gentlemen (Chap. 19)
demonstrates the power of R statistical programming language
and packages to analyze statistical and topological properties of
graph-based representations of biological networks.

4. Dynamical
Modelling of
Bacterial
Networks A proper understanding of biological processes requires an
(Part III) appreciation of their temporal and spatial behaviour. Although
mathematical biologists have been working for decades on the
development of dynamical models that recapitulate in silico some
essential aspects of the behaviour of living organisms (e.g., cell
cycle), the correlation of such models with functional genomic
data sets is seldom made. Here again, bacteria offer unmatched
opportunities because of their limited complexity compared to
multicellular organisms, their amenability to experiments, and the
existence of rich data sets (in particular for E. coli). The last part of
1 Bacterial Molecular Networks: Bridging the Gap. . . 7

the book presents a series of qualitative and quantitative approaches

to model the temporal behaviour of biological networks in bacteria.
Focusing on metabolism, Behre et al. (Chap. 20) review estab-
lished methods to detect and analyze structural invariants, point-
ing to conserved moieties or elementary fluxes at the basis of
asymptotical stationary behaviour. Marwan et al. (Chap. 21)
present the Petri net formalism, a framework enabling the devel-
opment of discrete, continuous, or stochastic models, or even
combination thereof. Petri nets are particularly suited to the mod-
elling of metabolism and have been recently applied to signal
transduction pathways, and, to a lesser extent, to transcriptional
regulatory networks. This is illustrated by their case study: the
modelling of phosphate regulation in enteric bacteria. Batt et al.
(Chap. 22) focus on the regulatory network controlling E. coli
response to carbon starvation, and describe the progressive build-
ing of a predictive qualitative dynamical model for this response
and its parameterization. The protocol proposed by Chaouiya
et al. (Chap. 23) uses a logical formalism, that allows for the
simulation of bacteriophage l development, while hinting at
the specific roles of feedback circuits found at the core of the
phage regulatory network.
Most detailed mechanistic dynamical models are limited to the
level of single organisms. In contrast, mathematical ecology and
evolutionary studies emphasize the relationships between variants
within a population, or between different species. Bacteria are
involved in a wide spectrum of interspecies relationships with a
wide variety of organisms, from commensalism to mutualism and
from symbiosis to parasitism. Analysis and modelling of these rela-
tionships is an important emerging field of research illustrated
here by two chapters. Friesen and Jones (Chap. 24) review eco-
evolutionary studies of adaptive dynamics, inclusive fitness, and
population genetic models, providing insight into the strengths
and weaknesses of each approach and into how current evolutionary
methods can contribute to the deciphering of the mechanistic basis
of host–symbiont interactions. Focusing on bacterial virulence,
Kepseu et al. (Chap. 25) review the main mechanisms used by
bacteria to trigger the production of virulence factors. The dynami-
cal modelling of these mechanisms leads to the identification of two
qualitatively distinct infectious transitions, while an irreversible
switch behaviour is at the basis of an efficient onset of virulence.
The studies mentioned above refer to coarse-grain dynamical
models of bacterial molecular networks, where variables represent
the mean behaviour of population of biomolecules, cells, or
species. Justified for large numbers of components, this approxi-
mation does not reflect the variety of behaviour that can occur
at the microscopic level, in particular when the number of inter-
acting components is small. This consideration has led to the
development of stochastic methods for the modelling of the
8 J. van Helden et al.

temporal behaviour of individual components (e.g., molecules

or cells) in spatially distributed interacting networks. Andrews
(Chap. 26) presents Smoldyn, a computer program for modelling
cellular systems with spatial and stochastic detail. Thanks to a
novel rule-based modelling, chemical species and reactions are
automatically generated as they arise in simulations, enabling
computationally efficient simulations of bacterial systems.

5. Outlook:
Bridging the Gap
Between
Bioinformatics During the recent years, sustained by the progress in high-
and Dynamical throughput “omics” approaches, our perception of the bacterial
Modelling world, its dynamics and evolution have undergone a revolution.
Various microbiomes and viromes, sampled from niches as
diverse as mammalian gut, oceans or polluted environments, can
now be analyzed to answer questions about diseases and environ-
mental equilibrium, which could not even have been formulated a
decade ago. Several recently published studies (12–14) illustrate
how several high-throughput methodologies can altogether pro-
vide a comprehensive picture of transcriptional activity, protein
expression and interactions, along with metabolic activity in
bacteria such as Mycoplasma pneumoniae. The multiplication of
such investigations raises a pressing, increasing need to combine
methods and expertise in mathematics, informatics, and biology
to extract significant insights from the flood of molecular data.
By combining the efficient computational tools such as those
presented in this and other volumes of the MiMB series (15–18),
researchers will be able to organize, analyze, and visualize large-
scale molecular datasets and, most importantly, to assess the sta-
tistical relevance of the features of interest. Such tools are essential
to infer the molecular behaviour of bacteria in specific conditions
(i.e., response to available nutriments, hosts for symbiosis, or
pathogenesis, etc.). By and large, high-throughput technologies
are providing us with a snapshot of the network state, but a real
understanding of the biological systems’ behaviour would require
a proper formalization of the underlying causal relationships.
This problem is at the core of the domain of mathematical
biology since half a century. In the recent years, sophisticated
methods and software tools have been developed to model
the spatial–temporal behaviour of bacterial networks at different
scales and with different levels of details. Arguably, qualitative
approaches are better adapted to cope with the current lack of
quantitative data. However, with the raise of large-scale quantita-
tive measurements, we can expect more and more opportunities
for the application of detailed kinetic or stochastic approaches
to the predictive dynamical modelling of bacterial molecular
1 Bacterial Molecular Networks: Bridging the Gap. . . 9

networks. Irrespective of the mathematical formalism, it remains

challenging to explicitly articulate functional genomics data
analysis and dynamical modelling, even for a single cell. Another
recurrent difficulty lies in relating components, interactions, and
behaviours, from molecules to microbial populations or ecosys-
tems and that occur at different temporal and spatial scales.
Although microbes have recurrently proved to be much more
complex than expected, bacteria still offer excellent opportunities
to develop novel concepts, methods, and tools to bridge the gaps
between computational biology and dynamical modelling. This is
a necessary step to reach the level of understanding commended
by “systems biology” and, even more, for the development
of synthetic biology, which beyond the characterization and
predictive modelling of existing organisms aims at engineering
new living systems. Among these, bacteria may be engineered
with properties to cope with economic, environmental, or medical
challenges (for a recent review on synthetic bacteriology, see
ref. 19). Here again, efficient design requires a proper blend of
computational and experimental expertise (see e.g., ref. 20).
A first important step in the synthesis of a new bacterial
cell was recently reported (21). A Mycoplasma chromosome
was synthesized, assembled, and introduced into cells of a differ-
ent Mycoplasma species, generating cells capable of continuous
genome replication and cell division with properties characteristic
of the sole synthetic genome. This amazing technical accomplish-
ment does not imply that we can master bacterial reproduction
and evolution. Indeed, a systematic exploration of the conse-
quences of rewiring regulatory interactions among global regula-
tors in the best known bacteria, E. coli, led to puzzling and
unpredictable results in terms of gene expression profiles and
global fitness (cf. 22; Isalan, personal communication). Explicit
dynamical modelling of the underlying networks may hopefully
allow for the understanding of such non intuitive behaviour.
On a longer range, further methodological developments
along the lines outlined in this volume will no doubt contribute
to go from the understanding and possible manipulation of
individual bacterial cells and cellular networks to the management
of bacterial populations, of their interactions, as well of their
response to surrounding viral and other predators populations.

Acknowledgments

The collaboration between the TAGC (DT) and BiGRe (JvH

and AT) laboratories is supported by the Belgian Program on
Interuniversity Attraction Poles, initiated by the Belgian Federal
10 J. van Helden et al.

Science Policy Office, project P6/25 (BioMaGNet). The BiGRe

laboratory is further supported by the MICROME Collaborative
Project funded by the European Commission within its FP7
Programme, under the thematic area “BIO-INFORMATICS –
Microbial genomics and bio-informatics” (contract number
222886–2).

References
1. Jacob F, Monod J. (1961) Genetic regulatory 13. Kuhner S, van Noort V, Betts MJ, Leo-Macias
mechanisms in the synthesis of proteins. J Mol A, Batisse C, Rode M, Yamada T, Maier T,
Biol, 3:318–356. Bader S, Beltran-Alvarez P, Castano-Diez D,
2. Monod J, Jacob F. (1961) Teleonomic Chen WH, Devos D, Guell M, Norambuena
mechanisms in cellular metabolism, growth, T, Racke I, Rybin V, Schmidt A, Yus E, Aeber-
and differentiation. Cold Spring Harb Symp sold R, Herrmann R, Bottcher B, Frangakis
Quant Biol, 26:389–401. AS, Russell RB, Serrano L, Bork P, Gavin
3. Brock TD. (1990) The Emergence of Bacterial AC. (2009) Proteome organization in a
Genetics. Cold Spring Harbor Laboratory genome-reduced bacterium. Science,
Press, Cold Spring Harbor, NY. 326:1235–1240.
4. Beckwith JR, Silhavy TJ. (1992) The Power of 14. Yus E, Maier T, Michalodimitrakis K, van
Bacterial Genetics: A Literature-Based Course. Noort V, Yamada T, Chen WH, Wodke JA,
Cold Spring Harbor Laboratory Press, Plain- Guell M, Martinez S, Bourgeois R, Kuhner S,
view, NY. Raineri E, Letunic I, Kalinina OV, Rode M,
Herrmann R, Gutierrez-Gallego R, Russell
5. Thieffry D. (1996) Escherichia coli as a model
RB, Gavin AC, Bork P, Serrano L. (2009)
system with which to study cell differentiation.
Impact of genome reduction on bacterial
Hist Philos Life Sci, 18:163–193.
metabolism and its regulation. Science,
6. Morange M. (1998) A History of Molecular 326:1263–1268.
Biology. Harvard University Press, Cambridge,
MA. 15. Fenyö D. (2010) Computational Biology,
Methods in Molecular Biology. Methods in
7. Thomas R, D’Ari R. (1990) Biological Feed- Molecular Biology, Vol. 673. Springer.
back. CRC Press, Boca Raton, FL.
16. Ladunga I. (2010) Computational Biology of
8. Thieffry D. (2007) Dynamical roles of Transcription Factor Binding. Methods in
biological regulatory circuits. Brief Bioinform, Molecular Biology, Vol. 674. Springer.
8:220–225.
17. Wu CH, Chen C. (2011) Bioinformatics for
9. Alon U. (2007) An Introduction to Systems Comparative Proteomics. Methods in Molecular
Biology: Design Principles of Biological Circuits. Biology, Vol. 694. Springer.
Chapman & Hall/CRC, Boca Raton, FL.
18. Hamacher M, Eisenacher M, Stephan C.
10. Martinez-Antonio A, Janga SC, Thieffry D.
(2011) Data Mining in Proteomics. Methods
(2008) Functional organisation of Escherichia
in Molecular Biology, Vol. 696. Springer.
coli transcriptional regulatory network. J Mol
Biol, 381:238–247. 19. Michalodimitrakis K, Isalan M. (2009) Engi-
neering prokaryotic gene circuits. FEMS
11. Lima-Mendez G, van Helden J. (2009) The
Microbiol Rev, 33:27–37.
powerful law of the power law and other
myths in network biology. Mol Biosyst, 20. Stricker J, Cookson S, Bennett MR, Mather
5:1482–1493. WH, Tsimring LS, Hasty J. (2008) A fast,
robust and tunable synthetic gene oscillator.
12. Guell M, van Noort V, Yus E, Chen WH,
Nature, 456:516–519.
Leigh-Bell J, Michalodimitrakis K, Yamada T,
Arumugam M, Doerks T, Kuhner S, Rode M, 21. Gibson DG, Glass JI, Lartigue C, Noskov VN,
Suyama M, Schmidt S, Gavin AC, Bork P, Chuang RY, Algire MA, Benders GA, Monta-
Serrano L. (2009) Transcriptome complexity gue MG, Ma L, Moodie MM, Merryman C,
in a genome-reduced bacterium. Science, Vashee S, Krishnakumar R, Assad-Garcia N,
326:1268–1271. Andrews-Pfannkoch C, Denisova EA,
1 Bacterial Molecular Networks: Bridging the Gap. . . 11

Young L, Qi ZQ, Segall-Shapiro TH, Calvey 22. Isalan M, Lemerle C, Michalodimitrakis K,

CH, Parmar PP, Hutchison CA, 3rd, Smith Horn C, Beltrao P, Raineri E, Garriga-Canut
HO, Venter JC. (2010) Creation of a bacterial M, Serrano L. (2008) Evolvability and hierar-
cell controlled by a chemically synthesized chy in rewired bacterial gene networks.
genome. Science, 329:52–56. Nature, 452:840–845.
Part I

Experimental and In Silico Approaches to

Unravel Network Components
Chapter 2

Bacterial Interactomes: From Interactions to Networks

Emmanuelle Bouveret and Christine Brun

Abstract
In order to ensure their function(s) in the cell, proteins are organized in machineries, underlaid by a
complex network of interactions. Identifying protein interactions is thus crucial to our understanding of
cell functioning. Technical advances in molecular biology and genomic technology now allow for the
systematic study of the interactions occurring in a given organism. This review first presents the techniques
readily available to microbiologists for studying protein–protein interactions in bacteria, as well as their
usability for high-throughput studies. Two types of techniques need to be considered: (1) the isolation of
protein complexes from the organism of interest by affinity purification, and subsequent identification of
the complex partners by mass spectrometry and (2) two-hybrid techniques, in general based on the
production of two recombinant proteins whose interaction has to be tested in a reporter cell. Next, we
summarize the bacterial interactomes already published. Finally, the strengths and pitfalls of the techni-
ques are discussed, together with the potential prospect of interactome studies in bacteria.

Key words: Protein–protein interaction, Tandem affinity purification, Bacterial two-hybrid, Yeast
two-hybrid, Protein networks, Interactome

1. Introduction

1.1. From Up to recently, classical biochemical and genetic methods have

Interactions... been used to identify interactions between genes and between
proteins at a slow pace. Usually, these experiments follow a protocol
established for the elucidation of the functions of one or a few
genes/proteins of interest or for a better understanding of a partic-
ular biological process. Beyond the success of these approaches, the
genomics era has introduced major modifications in the way biol-
ogists envisage to get answers to their questioning and as a result,
plan experiments at a completely different scale. Genome-wide and
more generally global approaches open the possibility to tackle cell
functioning as a whole, namely as a system.
Understanding the biological processes at a system-level
implies the analysis of interaction networks. The deciphering of
these networks first necessitates identifying the relations and

Jacques van Helden et al. (eds.), Bacterial Molecular Networks: Methods and Protocols, Methods in Molecular Biology,
vol. 804, DOI 10.1007/978-1-61779-361-5_2, # Springer Science+Business Media, LLC 2012

15
16 E. Bouveret and C. Brun

interactions that exist between the parts of the system, i.e., the
genes and the proteins. To this aim, large-scale and high-
throughput methods have been developed to discover interactions
of different nature. Whereas DNA microarray technologies point
at related co-expressed genes, ChIP-on-chip and ChIP-seq meth-
odologies reveal protein–DNA interactions occurring in vivo, and
synthetic genetic array analyses identify genetically interacting
genes such as synthetic lethals.
Regarding protein–protein interactions (PPI), two techno-
logies, originally devised to identify specific interactions, protein
complex affinity purification and two-hybrid methods, have been
optimized to allow for large-scale analyses. The first detects direct
and indirect protein–protein interactions occurring within protein
complexes by identifying the protein partners of a given protein
using affinity purification (1). The second detects direct binary
interactions and is generally based on molecular or genetic comple-
mentarity principles. For instance, a binary interaction existing
between a bait and a prey allows for the reconstitution of a functional
transcription factor, which in turn, activates the expression of a
reporter gene (2–5).

1.2. ...To Once identified by these large-scale methodologies, interactions

Protein–Protein can be assembled into protein–protein interaction maps. These
Interaction Networks... “interactomes” (6) represent a set of possible biophysical interac-
tions between the tested proteins of the studied organism. Their
exploration leads either to functional predictions for uncharacterized
proteins based on the composition of their network neighbourhood,
or provides information about the organization of the biological
processes into the cell, their coordination, their actors, etc.

1.3. ...In Bacteria Whole genome sequencing programs revealed that almost a third
of the protein-coding genes in every genome is functionally
uncharacterized and occasionally lacks homologs in distant species
(7). Even for the deeply studied Escherichia coli model bacterium,
Gene Ontology annotations (8) for E. coli genes (August 2009
release) show that ~30% of the genes cannot be assigned any
functional annotations (Fig. 1). Forty-five percent of E. coli
genes have had either their molecular or cellular function experi-
mentally elucidated, yet both have been proven for only 15%. The
extent of our knowledge of gene functions in the model bacterium
is thus amazingly limited despite more than half a century of
investigations. The situation is obviously even worth for less stud-
ied bacterial species. Consequently, it is rather unexpected that
despite the relatively small size of bacterial genomes, only few
high-throughput protein–protein interaction studies have been
performed so far in these organisms. This is quite detrimental to
the field of systems biology since, one could reasonably predict
that if the functioning of any living organism comes to be fully
2 Bacterial Interactomes: From Interactions to Networks 17

Molecular Function Biological Process

25%
27%

39% 51%

22%
36%

Not Annotated Experimental Evidences Electronic Annotations

Fig. 1. Extent of the gene ontology annotations for Escherichia coli genes.

understood in a near future a bacterium such as E. coli or bacteria

with smaller genomes appear as the likeliest candidates.
This review focuses first onto the principles of the methods
allowing the identification of protein–protein interactions which
have already been used or can be used to produce the bacterial
interactomes. Second, available resources for bacterial interaction
data are presented. Finally, the strengths and weaknesses of the
experimental as well as the conceptual approaches are discussed.

2. Methods

2.1. Tandem Affinity Protein partners in a given complex can be identified through
Purification affinity purification. The principle of the methodology consists
in (1) fusing a tag to a given protein called the bait, in the
organism of interest, (2) producing the protein in physiological
conditions that warrant the formation of the natural complexes,
and (3) purifying the complexes by affinity. Purified proteins can
then be identified by mass spectrometry analyses. If in theory any
couple of tag and cognate affinity resin may be chosen, the tandem
affinity purification (TAP) has proven overtly successful.
TAP was originally developed in Saccharomyces cerevisiae
in Bertrand Séraphin’s laboratory at EMBL (1). The method is
based on a TAP tag composed of two IgG binding domains
originating from Staphylococcus aureus protein A and the
Calmodulin Binding Peptide (CBP) such that the tagged pro-
teins can be purified on IgG and Calmodulin-coated beads,
respectively (Fig. 2a). The two parts of the tag are separated by a
sequence of seven residues (ENLYFQG), which is specifically
18 E. Bouveret and C. Brun

a
CBP ENLYFQG protein A C-terminal TAP tag (21 kDa)
protein X
TEV cleavage site

CBP ENLYFQG 3xFlag C-terminal SPA tag (8 kDa)

TEV cleavage site

SBP ENLYFQG protein G C-terminal TAP(SG) tag (20 kDa)

TEV cleavage

CBP LEVLFQGP GST C-terminal gTAP tag (30 kDa)

3C cleavage site
b
chromosome
genetic enginee-

Protein extract
preparation

IgG beads

ENLYFQG protein A
Protein A binding CBP

on IgG beads X

Washes

IgG beads

CBP ENLYFQG protein A

TEV cleavage
X

CBP binding on Camodulin beads

Calmodulin beads CBP ENLYFQ
X

washes
EGTA

Elution with EGTA

CBP ENLYFQ
X

Purification
analysis by mass spectrometry

Fig. 2. Principle of TAP-like purifications. (a) The organization of different TAP-like tags
described in the text is shown. CBP calmodulin binding peptide, SBP streptavidin
binding peptide. TAP, SPA, TAP(SG), and gTAP tags have been described in refs. 1,
12, 22, 23. (b) Procedure of the TAP method.
2 Bacterial Interactomes: From Interactions to Networks 19

recognized and cleaved by the Tobacco Etch Virus (TEV) prote-

ase between the last Q and G residues. In total, the TAP tag is
about 20 kDa long and can be placed either at the N terminus or
at the C terminus of the bait protein. Purification is performed in
two successive steps (Fig. 2b; see ref. 9). First, the protein extract
(prepared from a culture of a strain producing the tagged protein)
is incubated on IgG-coated beads. After a washing step, the pur-
ified material is eluted from the column using the TEV protease.
The eluted material is then incubated on Calmodulin-coated
beads, the column is washed and the purified material can be eluted
by addition of EGTA that chelates the Ca2+ ions necessary to the
CBP/Calmodulin interaction. The composition of the isolated
protein complex is then analyzed by mass spectrometry, with or
without prior separation by polyacrylamide gel electrophoresis.
The two successive purification steps rely on different tags and
affinity systems and the elution procedures at each step do not
modify the buffer characteristics. Hence, this method permits
the recovery of a highly pure preparation, while preserving inter-
actions. Yet, the technique not only relies on the purification
procedure, but also on the nature of the initial biological material
from which the protein extract is prepared. In order to get the best
results, the tagged protein must be produced by the cells under
physiological conditions. Thus, ideally, the recombinant protein
should replace the endogenous wild-type protein and should not
be overproduced. When the studied organism is amenable to
genetic manipulation, this is the solution of choice. In yeast, the
process consists in introducing a cassette coding for the TAP tag
and a selection marker at the extremity of the genes of interest
by homologous recombination. However, in most organisms,
intermediate solutions exist. The tagged proteins can for instance
be produced in the corresponding mutants, in an expression
system that does not overproduce the tagged protein, or preserves
the natural promoter of the gene. Equivalents of the yeast TAP-
tagging cassettes have been constructed for E. coli (10–12). The
TAP method has been applied successfully to low-scale as well as
large-scale purification of protein complexes in bacteria as diverse
as E. coli (13–15), B. subtilis (16), Synechocystis (17), Helicobacter
pylori (18), or Mycoplasma pneumoniae (19).

2.2. Design of New Since the first publication of the TAP method, several new tags
Tags for TAP-like have been described aiming at improving the binding, the stabil-
Methods ity, and the functionality of the tagged protein. These methods
retain the principle of two successive and independent affinity
steps to ensure the physiological relevance of the identified protein
interactions.
A wide range of tags is available (20), which could theoreti-
cally be used for TAP-like protein complex purification in diverse
combinations. We do not describe all of them here, but shall
20 E. Bouveret and C. Brun

highlight those that have been or can be used in bacteria for

systematic protein–protein interaction studies.
For purification in E. coli, the sequential peptide affinity (SPA)
tag contains three repeats of the Flag epitope instead of protein A
used in the original TAP tag (Fig. 2a). The first step of purification is
thus performed on an anti-Flag column (12). The SPA tag is much
smaller than the TAP tag (5 kDa instead of 20 kDa), and has been
shown to increase the stability and the functionality of E. coli tagged
proteins (12). Very useful and detailed protocols describe the TAP
and SPA methods in E. coli (21). SPA tag vectors have also been
developed for chromosomal tagging in Gram-positive bacteria (22).
Furthermore, a new tag (gTAP; Fig. 2a) has been engineered, in
which the TEV cleavage site has been replaced by a protease 3C site
and the protein A part by the GST tag (22), one advantage being
that unlike the TEV protease, protease 3C is active at 4 C.
For purification in mammalian cells, the SG and GS tags have
been developed (23). In this case, protein G replaces the protein A
part of the TAP tag, and a Strep tag replaces the CBP tag (Fig. 2a).
IgG and streptavidin sepharose columns are respectively used for
the purification steps and complexes are then eluted with biotin.
In addition, the codon usage and the exact sequences of the SG
and GS tags for animal cells and other variants in plants (24) have
been optimized. These improved tags should work in bacteria
where they remain to be tested.
Other combinations of tags have been described and used for
protein complex purification in Salmonella typhimurium (25, 26),
which combine those described above and others such as the
6-histidine tag or the BIO tag (in vivo biotinylation signal peptide)
(25) in order to replace one or both purification steps.
All the variants described above involve two steps of purifica-
tion as in the initial TAP method and use elution processes that
preserve the interactions. Depending on the organism of interest,
the researcher will choose the most adapted constructions.

2.3. Two-Hybrid The techniques described above are based on protein co-purification
Approaches from natural protein extracts prepared from the organism of
interest. In contrast, alternative techniques grouped under the gen-
eral term of “two-hybrid” have been developed in order to test
whether two given proteins do or not interact. These techniques
only provide information about the existence of binary biophysical
interactions regardless of the context in which they naturally occur.
In general, these methods rely on the in vivo detection of an interac-
tion between two proteins, though often in a heterologous expres-
sion system. The two proteins to be tested are each fused to separated
domains of a same functional protein and produced together in a
reporter cell (Fig. 3). Interaction between the two recombinant
proteins is detected in vivo in the reporter cell by enabling the
reconstitution of either a missing biological activity, or a novel
2 Bacterial Interactomes: From Interactions to Networks 21

YN YC

Fig. 3. Different types of two-hybrid techniques. UAS upstream activating sequence, DNA BD DNA-binding domain, AD
activation domain, RNAP RNA polymerase. GFP, YFP, mRFP: green, yellow, or Red fluorescent protein. YN, YC N-terminal
or C-terminal domains of YFP. Rluc Luciferase of Renilla reniformis.
22 E. Bouveret and C. Brun

activity. Depending on the chosen two-hybrid technique, the

restored biological activity consists in an active transcriptional activa-
tion factor, an enzymatic activity, or a fluorescent protein.

2.3.1. Classical Yeast Two- The best-known and most widely used two-hybrid technique
Hybrid: Transcriptional is the yeast two-hybrid system based on the modularity of
Activation Factor transcription activators, first described in refs. 3, 4 (Fig. 3a).
Reconstitution By bringing together a DNA-binding domain and a transcription
activator domain, the interaction of two proteins can be detected
through the activation of a reporter gene with a promoter
sequence that includes a DNA-binding site for the transcription
factor. The two proteins are fused one to a DNA-binding domain
(usually from the bacterial LexA repressor or the yeast Gal4
transcription factor), the other to an activation domain (usually
from E. coli B42 or yeast Gal4) on two distinct expression vectors.
These are then introduced together in an engineered yeast strain
bearing the appropriate reporter gene. In order to be detected,
the interaction has to occur in the yeast nucleus. Schematically,
two strategies can be adopted. First, a matrix of known proteins
for which ORFs are cloned in the two expression vectors can be
investigated systematically. Alternatively, a given “bait” cloned in
the DNA-binding fusion vector can be screened against a genomic
library constructed in the activator domain fusion vector, in order
to identify without a priori new protein partners called the preys.
A huge literature about the yeast-two hybrids methodology
is available. For technical reference, the reader can begin with
the review of Brent and Finley (27). Although this review precedes
the genome-wide applications of the two-hybrid method, it
nevertheless clearly exposes the requirements for validating new
interactions, discusses strengths and pitfalls, and the problem of
false positives and false negatives, all recommendations that are
still fully valid. The issue 328 of Methods of Enzymology (2000)
contains the detailed protocols used by various groups. More
recent reviews of protocols for screening can be found in
specialized journal issues or books such as ref. 28 for array-based
screens and ref. 29 for genomic library screening.
The yeast-two hybrid technique has been applied with great
success to the identification of interactions between bacterial
proteins such as the six proteins of the E. coli Tol/Pal system in
which interaction domains were mapped (30). It also contributed
to identify new partners of the E. coli small GTPase Era by genomic
library screening (31) and overall was used for the majority of
high-throughput screens in bacteria (see below) (32–36).
A two-hybrid system based on the reconstitution of a tran-
scription activation factor has also been designed to detect PPI
in vivo in E. coli cells. It uses the lcI DNA-binding domain and
the N-terminal domain of the a subunit of RNA polymerase
(RNAP), which is thus directly recruited to the promoter by the
2 Bacterial Interactomes: From Interactions to Networks 23

interaction (37, 38). This system appears more physiological for

studies of bacterial protein interactions and is amenable to geno-
mic library screening.

2.3.2. Enzymatic-Protein The second type of two-hybrid approaches consists in the recon-
Reconstitution (Protein stitution of an enzymatic activity or of a protein detectable by its
Fragment fluorescence emission. This approach called Protein Fragment
Complementation Assay) Complementation Assay (PCA) has been developed using several
types of proteins such as adenylate cyclase, DHFR, or GFP
(Fig. 3b) (39–41). A variety of systems is available, which all use
two distinct expression vectors. Although these methods have not
contributed to bacterial network deciphering so far, they should
be amenable to high-throughput studies either by cloning the
ORFeome of a bacterium, or by screening genomic libraries
against a subset of baits.

Bacterial Two-Hybrid The BACTH (bacterial adenylate cyclase two hybrid) technique
has been increasingly used for studies on bacterial proteins in E. coli
(42). The method is based on the reconstitution of adenylate
cyclase activity in an E. coli cya mutant, by bringing together the
T18 and T25 domains of adenylate cyclase from Bordetella pertussis
(Fig. 3b). The restored cAMP production is then indirectly
detected by its positive effect on the transcription of the lactose
and maltose operons, which in turn, can be visualized on reporter
colour plates (XGal or Mac Conkey), by b-galactosidase assay, or
by selection on lactose or maltose minimal media. Because this
technique relies on the reconstitution of an enzymatic activity and
a diffusible second messenger, it can be used to detect interactions
between membrane protein (13, 39). This method has been
increasingly used for systematically testing protein–protein inter-
actions in a subset of proteins of interest from diverse bacteria
(39, 42, 43) and its recent adaptation to genomic library screens
(44) opens the possibility of high-throughput analysis.

Other PCA Systems Used In theory, any enzyme that can be split in two non-functional
fragments that can re-associate through interaction with a third-
part could be used for PCA, as long as the reconstituted activity
can be easily detected or selected in vivo Murine Dihydrofolate
Reductase (mDHFR) has, for instance, been used for genomic
library screening in E. coli (45). The power of the PCA method
based on mDHFR has been highlighted by its transposition to
yeast, allowing the construction of an in vivo map of the yeast
protein interactome (46). Another very successful approach is the
Bimolecular Fluorescence Complementation (BiFC) technique,
which takes advantage of the refolding of a functional GFP variant
upon interaction between two hybrid proteins (Fig. 3b) (47).
This technique was used in E. coli (48), Agrobacterium tumefa-
ciens (49), or Bacillus (50). Although none of the bacterial
24 E. Bouveret and C. Brun

interactomes published so far have used PCA methods, all these

methods fulfil the requirements for being used in large-scale
studies: they allow for library screening and for an easy detection
or selection of the interactions in vivo.

2.3.3. Transfer A third type of method to detect binary protein–protein interac-

of Fluorescence tions in vivo is also based on the co-production in the cell of two
distinct hybrid proteins. In this case, the interaction brings in close
proximity two distinct fluorescent or luminescent proteins that
can transfer their energy (Fig. 3c). These FRET and BRET (Fluo-
rescence and Bioluminescence Resonance Energy Transfer,
respectively) techniques have been used in bacteria (51, 52).
However, interpretation of the data obtained with these techni-
ques is often complex, and hence whether they may be suitable for
high-throughput studies in bacteria remains an open question.
Nevertheless, FRET and BRET allow for real time analyses (52),
and complex technological developments such as combining
GFP-based PCA with FRET or BRET in a single assay allow for
the in vivo detection of multiple interactions, at a time in a given
complex (53). This makes fluorescence-based techniques very
powerful and promising.

3. Resources

3.1. High-throughput Although bacteria, essentially due to the relative small size and low
Bacterial Interactomes complexity of their genomes, appear as the organisms of choice for
systems biology projects, only few and still partial interactomes
have been deciphered (Table 1). Two among the techniques
described above have been used: affinity purification and the
classical yeast two-hybrid methodology.

3.1.1. Affinity Using the technique developed by Court’s laboratory (11)

Purification Studies and the SPA approach (12), Emili’s and Greenblat’s groups con-
structed a collection of E. coli tagged strains to produce an E. coli
interactome. Strains expressing the TAP or the SPA tags at the
C-terminus of ~2,000 proteins were constructed (15, 55) in
the recombination inducible DY330 strain. Essential and soluble
proteins (15), then orphan proteins (55) were targeted, adding
up to ~1/3 of the E. coli proteome. The 857 tagged E. coli
strains described in the first study (15) are now distributed by
Open Biosystems (now part of Thermo Fisher Scientific). The
global results have been organized as web-accessible resources
(http://www.compsysbio.org/bacteriome/ (56) and http://
ecoli.med.utoronto.ca (55)). Due to the large number of identified
interactions and the high coverage of the E. coli proteome, the last
study allowed for the prediction of functions for orphan proteins,
2 Bacterial Interactomes: From Interactions to Networks 25

Table 1
Bacterial large-scale interaction screens

Total Number of Number of Number of

Bacterial ORF tested identified identified
species Methods number baits preys interactions References
Helicobacter Y2H 1,590 261 480 (?) 1,280 Rain et al.
pylori 26695 (32)
Escherichia coli TAP- 4,236 648 1,369 5,254 Butland et al.
K12 DY330 TAG (15)
Escherichia coli His- 4,236 2,667 2,012 11,511 Arifuzzaman
K12 W3110 TAG et al. (57)
Escherichia coli TAP- 4,236 1,241 1,316 5,997 Hu et al.
K12 W3110 TAG (55)
Mycoplasma TAP- 689 212 411 1,058 K€
uhner et al.
pneumoniae TAG (19)
M129
Campylobacter Y2H 1,654 637 1,248 11,687 Parrish et al.
jejuni (33)
NCTC11168
Synechocystis Y2H 3,569 1,037 1,352 3,236 Sato et al.
PCC6803 (34)
Mesorhizobium Y2H 7,277 1,542 1,084 3,121 Shimoda
loti et al. (35)
Treponema Y2H 1,039 606 357 3,640 Titz et al.
pallidium (36)
Steptococcus PING 2,126 28 40 157 Gerber et al.
pneumoniae (70)
Steptococcus Y2H 2,126 – – 77 MPIDB (61)
pneumoniae

thanks to their clustering in protein complexes or functional

modules (55).
In a second high-throughput PPI study of E. coli using affinity
purification (57), all E. coli ORFs were tested using a different,
less physiological approach. Four thousand three hundred and
thirty-nine 6-histine N-terminal tagged proteins were produced
on plasmids under the control of an IPTG-inducible T5-lac
promoter. It is to note that a single round of purification was
used in this study (cf. Subheading 2.1).
Taken as a model for the proteome organization of a genome-
reduced bacterium, the interactome of M. pneumoniae was
recently deciphered by the TAP approach (19). In this case,
all TAP-fusions were produced from a genomic integration
26 E. Bouveret and C. Brun

Table 2
Main databases for bacterial interactions.

Main Interaction Number of bacterial

databases Species types interactions reported URL
IntAct All Binary 38,116 http://www.ebi.ac.uk/intact/
MINT All Binary 7,456 http://mint.bio.uniroma2.it/
MPIDB Bacteria Binary 23,580 http://www.jcvi.org/mpidb/
Bacteriome Escherichia Direct, indirect 11,209 http://www.compsysbio.org/
coli and inferred bacteriome/

under the control of a unique promoter (456 strains compared to

the 689 protein coding genes of the bacterium). Ultimately, the
interactome covers approximately 85% of the predicted soluble
proteome of M. pneumoniae.

3.1.2. Two-Hybrid Screens Five high-throughput bacterial protein–protein interactions

screens using the yeast two-hybrid have been published so far.
Here again, two approaches have been followed. First, nearly all
the ORFs of the organism were cloned in the yeast two-hybrid
vectors and systematically tested in all possible combinations. This
array-based approach was performed in Treponema pallidum (36)
and Campylobacter jejuni (33), which both have relatively small
genomes (1,039 and 1,654 predicted ORFs, respectively).
Second, in H. pylori (32), Synechocystis sp. (34), and Mesorhi-
zobium loti (35), a subset of “baits” were cloned and screened
against a genomic library. In each study, baits were chosen accord-
ing to specific research interests or traits of the studied bacterium.
For example, genes involved in symbiosis or conserved in the host
Arabidopsis thaliana and genes of unknown function were studied
in M. loti (35) and Synechocystis sp., respectively (34).

3.2. Databases Interactions identified with high-throughput methods constitute

large datasets, which are compiled within interaction databases.
Besides general databases such as IntAct (58) or MINT (59)
that contain protein–protein interactions detected in all types of
species including bacteria (and for which a detailed protocol has
been described in ref. 60), two databases are especially dedicated
to bacterial protein–protein interactions (Table 2). Whereas
Bacteriome.org ((56), http://www.compsysbio.org/bacteriome/)
only collects results of systematic proteomic studies on E. coli cited
above, MPIDB ((61), http://www.jcvi.org/mpidb/about.php)
contains 20,000 binary interactions from almost 200 bacterial
species and strains. Interestingly, this latter database provides an
2 Bacterial Interactomes: From Interactions to Networks 27

additional dataset of ~800 interactions identified by classical

low-throughput experiments and extracted from publication
abstracts. This database complies with the standards for inter-
action descriptions (62) and participates to international curation
efforts (IMEx, http://imex.sourceforge.net/).
Although inferred protein–protein interactions are beyond
the scope of this review, orthology information may come as a
rescue in face of the paucity of the data. The STRING database
((63), http://string-db.org/), which integrates interaction data
from low and large-scale experiments with other types of data
(co-expression, genomic context, etc.) allows for inferring inter-
actions by transfer between organisms, where applicable.

3.3. Strengths Eukaryotic and bacterial interactomes are far from compre-
and Pitfalls hensive, and their quality and current coverage are difficult to
estimate. Many of the available data suffer from (1) the presence
of false-positive interactions, i.e., that are detected experimentally
but do not exist in the organism in physiological conditions and
(2) the absence of a large number of true interactions, the false
negatives, which are not identified during the screens. Sources of
these pitfalls and ways to overcome them are developed below.

3.3.1. Bias One caveat of the TAP-like methods is that the procedure is long
and uses several washing steps such that only strong interactions
are retained (64). The TAP method is thus best suited for
the identification of stable and abundant complexes. In vivo
cross-linking prior the TAP-like purifications has been proposed
as a way to overcome this limitation. This additional step does not
appear to prevent subsequent protein identification by mass spec-
trometry. This approach has been successful for protein complex
purification from H. pylori cells with the TAP tag (18) or with the
His-Bio tag on yeast cells (25). TAP-like methods are also biased
toward soluble and abundant proteins. Small proteins are missed
due in part to the purification procedure, but also to the reliability
of mass spectrometry to identify small peptides. On the contrary,
the two-hybrid methods are not sensitive to the natural abundance
of proteins because these are not expressed in their natural context,
and the identification of the partners by sequencing is unequivocal.

3.3.2. Membrane Proteins The second pitfall of the TAP-like methods is inherent to affinity
purification procedures that fail to easily deal with membrane
proteins. Some specific protocols for the solubilization of mem-
brane proteins before TAP have been described (65) but these may
interfere with the interactions. A combination of cross-linking and
membrane solubilization might be the solution to this problem
of membrane complex purification. In the case of two-hybrid like
techniques, the problem can be circumvented by testing only
the soluble domains of membrane proteins. Furthermore, the
28 E. Bouveret and C. Brun

BACTH and PCA techniques have been used for the detection of
full length integral inner membrane proteins (13, 39).

3.3.3. Recombinant Protein A common problem in all the methods reported here is that they
Expression rely on the production of recombinant fusion proteins (the bait
and prey in two-hybrid like methods, only the bait in affinity
purifications). These recombinant proteins may either not retain
their function, or not be correctly produced or not localize
correctly in the cell or a combination of these. This could lead to
missing or nonphysiological interactions. In most two-hybrid
methods, recombinant proteins are over-produced. Although
this may be an advantage when allowing for the detection of
interactions involving proteins of low abundance or proteins
only produced in specific conditions, it may represent a disadvan-
tage since over-expression can provoke nonphysiological interac-
tions. On the contrary, TAP-like methods are based on the
physiological expression of all the partners of the putative protein
complexes and are therefore more reliable.

3.3.4. Contaminants Both TAP-like and two-hybrid-like methods produce false

positives, but not for the same reasons. In TAP-like methods,
false positives correspond to co-purifying contaminants, mostly
abundant enzymes and other proteins, such as ribosomal proteins,
but also chaperones that bind to the recombinant bait or its TAP
tag region. It is thus important to maintain a list of the usual
TAP contaminants found in a given organism following a given
procedure. Such a list was published for the TAP study of E. coli
(55). It contains pyruvate dehydrogenase subunits, chaperones,
and ribosomal proteins. The two-hybrid methods are parasitized
by (1) the so-called sticky proteins, which are not correctly folded
or are produced in a wrong cell compartment and (2) the numer-
ous auto-activating proteins that promote transcription, without
any interaction partner. The long experience with yeast two-
hybrid in eukaryotes leads to setting up procedures aiming at
cleaning up the raw data. Furthermore, any new interaction
(whether identified by TAP-like or by two-hybrid-like methods)
needs to be validated. The doctrine first described for yeast two-
hybrid data remains relevant (27): validation of the interaction by
another method, selection of mutations that break and restore the
interaction, specificity of the interaction, etc. Noticeably, a tool kit
composed of several high-throughput protein interaction
assays has been recently benchmarked against true and random
interactions in order to calculate a probability of existence for each
identified interaction (66). All the methods included in the tool
kit being based on the two-hybrid principle (classical yeast two-
hybrid, PCA, etc.), one can ideally imagine using it for retesting
bacterial interactions.
2 Bacterial Interactomes: From Interactions to Networks 29

3.3.5. Sampling Sensitivity Several large-scale screens may be required for achieving an
acceptable coverage of a given interactome (67). The sampling
effect is particularly visible when comparing several interaction
datasets obtained from the same species. Different affinity purifi-
cation methods used to generate three sets of data on E. coli
(cf. Subheading 3.1) identified only 62/17,049 interactions in
common (55). Without a priori about data quality, this result
shows that the interaction space of E. coli is still far from being
covered.

3.3.6. Confidence Score All potential problems discussed so far including false-positive and
false-negative interactions, stimulated the development of confi-
dence scores for interactions. Besides the experimental confidence
score for binary interactions derived from the use of the tool kit
evoked previously (66), confidence scores are usually based on
other experimental data obtained during the identification process
such as the frequency of detection of a given interaction as well
as independent data such as co-expression or/and subcellular
colocalization of the two interacting partners (32, 33, 68, 69).
The issue of the reliability of the interactions produced by the
high-throughput methods is crucial. Although one user may feel
more comfortable with interactions that are coherent with prior
knowledge on the protein partners, one should remember that a
lack of fit may suggest false negatives among the secondary data
(taken as prior knowledge) but also that over-validated interac-
tions only point toward already beaten tracks and impair the
discovery process.

3.3.7. Feasibility of High- Finally, the methods may be compared and evaluated regarding
throughput Studies their practicability and strengths as high-throughput methods.
Two-hybrid methods are clearly favoured in this case, due to
their simplicity compared to TAP-like methods that rely on
tedious steps of bacterial culture, biochemistry, and mass spec-
trometry.

4. Outlook

The fact that different methodologies have been used to generate

different datasets of interactions that have little in common (55)
calls for a comparison of the techniques’ sensitivity and coverage.
This will enable the estimation of the completeness and quality of
the identified interactions, as it was recently done for eukaryotes
(66). These high-throughput studies provide a wealth of informa-
tion to the microbiologist community, especially for the decipher-
ing of gene function at the genomic scale. However, beyond the
description of every protein–protein interactions that can occur in
30 E. Bouveret and C. Brun

a cell, the gain of information about dynamics of protein network

is a next step toward a global understanding of cell systems. To
understand how the interactome responds and adapts to modifi-
cation of the cell environment, interactomes obtained in different
physiological conditions will have to be compared. This can be
achieved experimentally in vivo using the TAP-like methods
and bioinformatically via the integration of pertinent genomic,
transcriptomic, and proteomic data with interactomes deciphered
by two-hybrid methods. Hopefully, such comparisons will provide
better understanding of the dynamics of the networks, resulting
from genetic regulation and changing interactions.

Acknowledgments

Work by E.B. is funded by the CNRS, a young research grant from

ANR, and FRM. Work by C.B. is funded by the CNRS, the
Inserm, the Université de la Méditerranée and grants from ANR.

References
1. Rigaut G, Shevchenko A, Rutz B, Wilm M, functional genomics. Nat Biotechnol,
Mann M, Seraphin B. (1999) A generic pro- 18:609–613.
tein purification method for protein complex 8. Ashburner M, Ball CA, Blake JA, Botstein D,
characterization and proteome exploration. Butler H, Cherry JM, Davis AP, Dolinski K,
Nat Biotechnol, 17:1030–1032. Dwight SS, Eppig JT, Harris MA, Hill DP,
2. Legrain P, Selig L. (2000) Genome-wide Issel-Tarver L, Kasarskis A, Lewis S, Matese
protein interaction maps using two-hybrid JC, Richardson JE, Ringwald M, Rubin GM,
systems. FEBS Lett, 480:32–36. Sherlock G. (2000) Gene ontology: tool for
3. Chien CT, Bartel PL, Sternglanz R, Fields S. the unification of biology. The Gene Ontology
(1991) The two-hybrid system: a method to Consortium. Nat Genet, 25:25–29.
identify and clone genes for proteins that 9. Puig O, Caspary F, Rigaut G, Rutz B, Bou-
interact with a protein of interest. Proc Natl veret E, Bragado-Nilsson E, Wilm M, Sera-
Acad Sci USA, 88: 9578–9582. phin B. (2001) The tandem affinity
4. Fields S, Song O. (1989) A novel genetic sys- purification (TAP) method: a general proce-
tem to detect protein–protein interactions. dure of protein complex purification. Meth-
Nature, 340:245–246. ods, 24:218–229.
5. Walhout AJ, Vidal M. (2001) High-through- 10. Datsenko KA, Wanner BL. (2000) One-step
put yeast two-hybrid assays for large-scale pro- inactivation of chromosomal genes in Escher-
tein interaction mapping. Methods, ichia coli K-12 using PCR products. Proc Natl
24:297–306. Acad Sci U S A, 97:6640–6645.
6. Sanchez C, Lachaize C, Janody F, Bellon B, 11. Yu D, Ellis HM, Lee EC, Jenkins NA, Cope-
Roder L, Euzenat J, Rechenmann F, Jacq B. land NG, Court DL. (2000) An efficient
(1999) Grasping at molecular interactions and recombination system for chromosome engi-
genetic networks in Drosophila melanogaster neering in Escherichia coli. Proc Natl Acad Sci
using FlyNets, an Internet database. Nucleic U S A, 97:5978–5983.
Acids Res, 27:89–94. 12. Zeghouf M, Li J, Butland G, Borkowska A,
7. Galperin MY, Koonin EV. (2000) Who’s your Canadien V, Richards D, Beattie B, Emili A,
neighbor? New computational approaches for Greenblatt JF. (2004) Sequential Peptide
2 Bacterial Interactomes: From Interactions to Networks 31

Affinity (SPA) system for the identification of 23. Burckstummer T, Bennett KL, Preradovic A,
mammalian and bacterial protein complexes. Schutze G, Hantschel O, Superti-Furga G,
J Proteome Res, 3:463–468. Bauch A. (2006) An efficient tandem affinity
13. Gully D, Bouveret E. (2006) A protein purification procedure for interaction proteo-
network for phospholipid synthesis uncovered mics in mammalian cells. Nat Methods,
by a variant of the tandem affinity purification 3:1013–1019.
method in Escherichia coli. Proteomics, 24. Rohila JS, Chen M, Cerny R, Fromm ME.
6:282–293. (2004) Improved tandem affinity purification
14. Gully D, Moinier D, Loiseau L, Bouveret E. tag and methods for isolation of protein
(2003) New partners of acyl carrier protein heterocomplexes from plants. Plant
detected in Escherichia coli by tandem affinity J, 38:172–181.
purification. FEBS Lett, 548:90–96. 25. Tagwerker C, Zhang H, Wang X, Larsen LS,
15. Butland G, Peregrin-Alvarez JM, Li J, Yang W, Lathrop RH, Hatfield GW, Auer B, Huang L,
Yang X, Canadien V, Starostine A, Richards D, Kaiser P. (2006) HB tag modules for PCR-
Beattie B, Krogan N, Davey M, Parkinson J, based gene tagging and tandem affinity purifi-
Greenblatt J, Emili A. (2005) Interaction net- cation in Saccharomyces cerevisiae. Yeast,
work containing conserved and essential pro- 23:623–632.
tein complexes in Escherichia coli. Nature, 26. Chowdhury SM, Shi L, Yoon H, Ansong C,
433:531–537. Rommereim LM, Norbeck AD, Auberry KJ,
16. Pompeo F, Luciano J, Galinier A. (2007) Moore RJ, Adkins JN, Heffron F, Smith RD.
Interaction of GapA with HPr and its homo- (2009) A method for investigating protein–
logue, Crh: novel levels of regulation of a key protein interactions related to salmonella
step of glycolysis in Bacillus subtilis? J Bacter- typhimurium pathogenesis. J Proteome Res,
iol, 189:1154–1157. 8:1504–1514.
17. Brown AP, Affleck V, Fawcett T, Slabas AR. 27. Brent R, Finley RL, Jr. (1997) Understanding
(2006) Tandem affinity purification tagging of gene and allele function with two-hybrid
fatty acid biosynthetic enzymes in Synechocystis methods. Annu Rev Genet, 31:663–704.
sp. PCC6803 and Arabidopsis thaliana. J Exp 28. Rajagopala SV, Uetz P. (2009) Analysis of
Bot, 57:1563–1571. protein–protein interactions using array-
18. Stingl K, Schauer K, Ecobichon C, Labigne A, based yeast two-hybrid screens. Methods Mol
Lenormand P, Rousselle JC, Namane A, de Biol, 548:223–245.
Reuse H. (2008) In vivo interactome of Heli- 29. Serebriiskii IG, Fang R, Latypova E, Hopkins
cobacter pylori urease revealed by tandem affin- R, Vinson C, Joung JK, Golemis EA. (2005) A
ity purification. Mol Cell Proteomics, combined yeast/bacteria two-hybrid system:
7:2429–2441. development and evaluation. Mol Cell Prote-
19. Kuhner S, van Noort V, Betts MJ, Leo-Macias omics, 4:819–826.
A, Batisse C, Rode M, Yamada T, Maier T, 30. Walburger A, Lazdunski C, Corda Y. (2002)
Bader S, Beltran-Alvarez P, Castano-Diez D, The Tol/Pal system function requires an inter-
Chen WH, Devos D, Guell M, Norambuena T, action between the C-terminal domain of
Racke I, Rybin V, Schmidt A, Yus E, Aebersold TolA and the N-terminal domain of TolB.
R, Herrmann R, Bottcher B, Frangakis AS, Mol Microbiol, 44:695–708.
Russell RB, Serrano L, Bork P, Gavin AC. 31. Zhang J, Inouye M. (2002) MazG, a nucleo-
(2009) Proteome organization in a genome- side triphosphate pyrophosphohydrolase,
reduced bacterium. Science, 326:1235–1240. interacts with Era, an essential GTPase in
20. Terpe K. (2003) Overview of tag protein Escherichia coli. J Bacteriol, 184:5323–5329.
fusions: from molecular and biochemical fun- 32. Rain JC, Selig L, De Reuse H, Battaglia V,
damentals to commercial systems. Appl Reverdy C, Simon S, Lenzen G, Petel F, Woj-
Microbiol Biotechnol, 60:523–533. cik J, Schachter V, Chemama Y, Labigne A,
21. Babu M, Butland G, Pogoutse O, Li J, Green- Legrain P. (2001) The protein–protein inter-
blatt JF, Emili A. (2009) Sequential peptide action map of Helicobacter pylori. Nature,
affinity purification system for the systematic 409:211–215.
isolation and identification of protein com- 33. Parrish JR, Yu J, Liu G, Hines JA, Chan JE,
plexes from Escherichia coli. Methods Mol Mangiola BA, Zhang H, Pacifio S, Fotouhi F,
Biol 564:373–400. DiRita VJ, Ideker T, Andrews P, Finley RL, Jr.
22. Yang X, Doherty GP, Lewis PJ. (2008) (2007) A proteome-wide protein interaction
Tandem affinity purification vectors for use in map for Campylobacter jejuni. Genome Biol,
gram positive bacteria. Plasmid, 59:54–62. 8:R130.
32 E. Bouveret and C. Brun

34. Sato S, Shimoda Y, Muraki A, Kohara M, 46. Tarassov K, Messier V, Landry CR, Radinovic
Nakamura Y, Tabata S. (2007) A large-scale S, Serna Molina MM, Shames I, Malitskaya Y,
protein–protein interaction analysis in Synecho- Vogel J, Bussey H, Michnick SW. (2008) An in
cystis sp. PCC6803. DNA Res, 14:207–216. vivo map of the yeast protein interactome.
35. Shimoda Y, Shinpo S, Kohara M, Nakamura Y, Science, 320:1465–1470.
Tabata S, Sato S. (2008) A large scale analysis 47. Kerppola TK. (2008) Bimolecular fluores-
of protein–protein interactions in the nitro- cence complementation (BiFC) analysis as a
gen-fixing bacterium Mesorhizobium loti. probe of protein interactions in living cells.
DNA Res, 15:13–23. Annu Rev Biophys, 37:465–487.
36. Titz B, Rajagopala SV, Goll J, Hauser R, 48. Magliery TJ, Wilson CG, Pan W, Mishler D,
McKevitt MT, Palzkill T, Uetz P. (2008) Ghosh I, Hamilton AD, Regan L. (2005)
The binary protein interactome of Treponema Detecting protein–protein interactions with a
pallidum – the syphilis spirochete. PLoS green fluorescent protein fragment reassembly
ONE, 3:e2292. trap: scope and mechanism. J Am Chem Soc,
37. Dove SL, Hochschild A. (1998) Use of artifi- 127:146–157.
cial activators to define a role for protein–pro- 49. Cascales E, Atmakuri K, Liu Z, Binns AN,
tein and protein–DNA contacts in Christie PJ. (2005) Agrobacterium tumefa-
transcriptional activation. Cold Spring Harb ciens oncogenic suppressors inhibit T-DNA
Symp Quant Biol, 63:173–180. and VirE2 protein substrate binding to the
38. Dove SL, Joung JK, Hochschild A. (1997) VirD4 coupling protein. Mol Microbiol,
Activation of prokaryotic transcription 58:565–579.
through arbitrary protein–protein contacts. 50. Soufo HJ, Graumann PL. (2003) Actin-like
Nature, 386:627–630. proteins MreB and Mbl from Bacillus subtilis
39. Karimova G, Dautin N, Ladant D. (2005) are required for bipolar positioning of replica-
Interaction network among Escherichia coli tion origins. Curr Biol, 13:1916–1920.
membrane proteins involved in cell division 51. Sourjik V, Berg HC. (2002) Binding of the
as revealed by bacterial two-hybrid analysis. J Escherichia coli response regulator CheY to its
Bacteriol, 187:2233–2243. target measured in vivo by fluorescence reso-
40. Nyfeler B, Michnick SW, Hauri HP. (2005) nance energy transfer. Proc Natl Acad Sci U S
Capturing protein interactions in the secretory A, 99:12669–12674.
pathway of living cells. Proc Natl Acad Sci U S 52. Xu Y, Piston DW, Johnson CH. (1999) A
A, 102:6350–6355. bioluminescence resonance energy transfer
41. Remy I, Campbell-Valois FX, Michnick SW. (BRET) system: application to interacting cir-
(2007) Detection of protein–protein interac- cadian clock proteins. Proc Natl Acad Sci U S
tions using a simple survival protein-fragment A, 96:151–156.
complementation assay based on the enzyme 53. Sourjik V, Vaknin A, Shimizu TS, Berg HC.
dihydrofolate reductase. Nat Protoc, (2007) In vivo measurement by FRET of path-
2:2120–2125. way activity in bacterial chemotaxis. Methods
42. Karimova G, Pidoux J, Ullmann A, Ladant D. Enzymol, 423:365–391.
(1998) A bacterial two-hybrid system based on 54. Rebois RV, Robitaille M, Petrin D, Zylbergold
a reconstituted signal transduction pathway. P, Trieu P, Hebert TE. (2008) Combining
Proc Natl Acad Sci U S A, 95:5752–5756. protein complementation assays with reso-
43. Marbouty M, Saguez C, Cassier-Chauvat C, nance energy transfer to detect multipartner
Chauvat F. (2009) ZipN, an FtsA-like orches- protein complexes in living cells. Methods,
trator of divisome assembly in the model cya- 45:214–218.
nobacterium Synechocystis PCC6803. Mol 55. Hu P, Janga SC, Babu M, Diaz-Mejia JJ, But-
Microbiol, 74(2):409–420. land G, Yang W, Pogoutse O, Guo X, Phanse
44. Karimova G, Robichon C, Ladant D. (2009) S, Wong P, Chandran S, Christopoulos C,
Characterization of YmgF, a 72-residue inner Nazarians-Armavil A, Nasseri NK, Musso G,
membrane protein that associates with the Ali M, Nazemof N, Eroukova V, Golshani A,
Escherichia coli cell division machinery. J Bac- Paccanaro A, Greenblatt JF, Moreno-Hagel-
teriol. 191:333–346. sieb G, Emili A. (2009) Global functional atlas
45. Michnick SW, Remy I, Campbell-Valois FX, of Escherichia coli encompassing previously
Vallee-Belisle A, Pelletier JN. (2000) Detec- uncharacterized proteins. PLoS Biol, 7:e96.
tion of protein–protein interactions by protein 56. Su C, Peregrin-Alvarez JM, Butland G, Phanse
fragment complementation strategies. Meth- S, Fong V, Emili A, Parkinson J. (2008) Bac-
ods Enzymol, 328:208–230. teriome.org – an integrated protein
2 Bacterial Interactomes: From Interactions to Networks 33

interaction database for E. coli. Nucleic Acids teins and their functional interactions in 630
Res, 36:D632–D636. organisms. Nucleic Acids Res, 37:
57. Arifuzzaman M, Maeda M, Itoh A, Nishikata D412–D416.
K, Takita C, Saito R, Ara T, Nakahigashi K, 64. Jensen LJ, Bork P. (2008) Biochemistry. Not
Huang HC, Hirai A, Tsuzuki K, Nakamura S, comparable, but complementary. Science,
Altaf-Ul-Amin M, Oshima T, Baba T, 322:56–57.
Yamamoto N, Kawamura T, Ioka-Nakamichi 65. Gavin AC, Aloy P, Grandi P, Krause R,
T, Kitagawa M, Tomita M, Kanaya S, Wada C, Boesche M, Marzioch M, Rau C, Jensen LJ,
Mori H. (2006) Large-scale identification of Bastuck S, Dumpelfeld B, Edelmann A, Heur-
protein–protein interaction of Escherichia coli tier MA, Hoffman V, Hoefert C, Klein K,
K-12. Genome Res, 16:686–691. Hudak M, Michon AM, Schelder M, Schirle
58. Kerrien S, Alam-Faruque Y, Aranda B, Ban- M, Remor M, Rudi T, Hooper S, Bauer A,
carz I, Bridge A, Derow C, Dimmer E, Feuer- Bouwmeester T, Casari G, Drewes G, Neu-
mann M, Friedrichsen A, Huntley R, Kohler bauer G, Rick JM, Kuster B, Bork P, Russell
C, Khadake J, Leroy C, Liban A, Lieftink C, RB, Superti-Furga G. (2006) Proteome sur-
Montecchi-Palazzi L, Orchard S, Risse J, vey reveals modularity of the yeast cell machin-
Robbe K, Roechert B, Thorneycroft D, ery. Nature, 440:631–636.
Zhang Y, Apweiler R, Hermjakob H. (2007) 66. Braun P, Tasan M, Dreze M, Barrios-Rodiles
IntAct – open source resource for molecular M, Lemmens I, Yu H, Sahalie JM, Murray RR,
interaction data. Nucleic Acids Res, 35: Roncari L, de Smet AS, Venkatesan K, Rual JF,
D561–D565. Vandenhaute J, Cusick ME, Pawson T, Hill
59. Zanzoni A, Montecchi-Palazzi L, Quondam DE, Tavernier J, Wrana JL, Roth FP, Vidal
M, Ausiello G, Helmer-Citterich M, Cesareni M. (2009) An experimentally derived confi-
G. (2002) MINT: a Molecular INTeraction dence score for binary protein–protein inter-
database. FEBS Lett, 513:135–140. actions. Nat Methods, 6:91–97.
60. Chatr-Aryamontri A, Zanzoni A, Ceol A, 67. Venkatesan K, Rual JF, Vazquez A, Stelzl U,
Cesareni G. (2008) Searching the protein Lemmens I, Hirozane-Kishikawa T, Hao T,
interaction space through the MINT database. Zenkner M, Xin X, Goh KI, Yildirim MA,
Methods Mol Biol, 484:305–317. Simonis N, Heinzmann K, Gebreab F, Sahalie
61. Goll J, Rajagopala SV, Shiau SC, Wu H, Lamb JM, Cevik S, Simon C, de Smet AS, Dann E,
BT, Uetz P. (2008) MPIDB: the microbial Smolyar A, Vinayagam A, Yu H, Szeto D,
protein interaction database. Bioinformatics, Borick H, Dricot A, Klitgord N, Murray RR,
24:1743–1744. Lin C, Lalowski M, Timm J, Rau K, Boone C,
62. Hermjakob H, Montecchi-Palazzi L, Bader G, Braun P, Cusick ME, Roth FP, Hill DE, Taver-
Wojcik J, Salwinski L, Ceol A, Moore S, nier J, Wanker EE, Barabasi AL, Vidal M.
Orchard S, Sarkans U, von Mering C, Roe- (2009) An empirical framework for binary
chert B, Poux S, Jung E, Mersch H, Kersey P, interactome mapping. Nat Methods, 6:83–90.
Lappe M, Li Y, Zeng R, Rana D, Nikolski M, 68. Zhang B, Park BH, Karpinets T, Samatova NF.
Husi H, Brun C, Shanker K, Grant SG, Sander (2008) From pull-down data to protein inter-
C, Bork P, Zhu W, Pandey A, Brazma A, Jacq action networks and complexes with biological
B, Vidal M, Sherman D, Legrain P, Cesareni relevance. Bioinformatics, 24:979–986.
G, Xenarios I, Eisenberg D, Steipe B, Hogue 69. Han JD, Bertin N, Hao T, Goldberg DS,
C, Apweiler R. (2004) The HUPO PSI’s Berriz GF, Zhang LV, Dupuy D, Walhout
molecular interaction format – a community AJ, Cusick ME, Roth FP, Vidal M. (2004)
standard for the representation of protein Evidence for dynamically organized modular-
interaction data. Nat Biotechnol, ity in the yeast protein–protein interaction
22:177–183. network. Nature, 430:88–93.
63. Jensen LJ, Kuhn M, Stark M, Chaffron S, 70. Gerber D, Maerkl SJ, Quake SR. (2009) An in
Creevey C, Muller J, Doerks T, Julien P, vitro microfluidic approach to generating pro-
Roth A, Simonovic M, Bork P, von Mering tein-interaction networks. Nat Methods,
C. (2009) STRING 8 – a global view on pro- 6:71–74.
Chapter 3

From Bacterial to Microbial Ecosystems (Metagenomics)

Shannon J. Williamson and Shibu Yooseph

Abstract
Metagenomics is revolutionizing the field of microbial ecology through techniques that eliminate the
prerequisite of culturing. Metagenomic studies of microbial populations in different environments
reveal the incredible diversity and adaptive capabilities of these organisms. With the advent of cheaper,
high-throughput sequencing technologies, these studies are also producing vast amounts of sequence
data. Here, we discuss the different components of a metagenomic study including sample collection,
DNA extraction, sequencing, and informatics. We highlight their issues and challenges, and review the
solutions that are currently in use. We conclude with examples of metagenomic studies conducted on
environments of varying complexities.

Key words: Metagenomics, Microbial ecology, Sample complexity, Shotgun sequencing,

Pyrosequencing, Metagenomic assembly, Binning, Gene finding, Metadata

1. Introduction

Metagenomics, or community genomics, is a rapidly emerging

discipline that seeks to understand microbial ecosystems by study-
ing the genomic content of the constituent microbes in their
natural community setting (1). This is extremely useful and
important since microbes that can be cultivated in the laboratory
are thought to account for less than 1% of the microbes that exist
in many environments (2). In a typical metagenomic study,
nucleic acids are extracted from a sample that is collected from
an environment of interest and directly sequenced using a random
shotgun approach, thus eliminating the need for in vitro culture.
We differentiate this from the paradigm (not discussed in this
chapter) that uses targeted survey sequencing of marker genes,
like the 16S rRNA gene, to interrogate a microbial community.
Metagenomic shotgun sequence data are typically used to study
two aspects of a microbial community – its taxonomic content
(who’s there?) and functional content (what are they doing?).

Jacques van Helden et al. (eds.), Bacterial Molecular Networks: Methods and Protocols, Methods in Molecular Biology,
vol. 804, DOI 10.1007/978-1-61779-361-5_3, # Springer Science+Business Media, LLC 2012

35
36 S.J. Williamson and S. Yooseph

Data from metagenomic studies have provided clues to the roles

that constituent microbes play in their communities, and have also
pointed to the incredible diversity of these organisms both at the
genome and protein levels.
While early metagenomic studies used the Sanger dideoxy
sequencing method (3) to generate sequence data, recent ones
have switched to the more cost effective next-generation sequenc-
ing technologies like the array-based pyrosequencing technology
from 454 Inc. (4) and the primer extension with fluorescent
reversible terminators method from Illumina Inc. (5). These
newer sequencing technologies also produce vastly more data per
run (compared to Sanger), and this has resulted in metagenomic
projects becoming dominant contributors to sequence databases.
In this chapter, we review various experimental and informat-
ics aspects of metagenomics. We describe sample collection, DNA
extraction, library construction, and sequencing. Among the
informatics aspects that are addressed are sequence assembly,
gene identification and annotation, taxonomic binning, sample
comparison, and linking sequence-level observations to the
sample metadata that is collected. The complexity of the various
analyses is a function of several variables including the diversity of
the microbial community, the depth of sequencing, and the choice
of sequencing platform. We conclude with examples of metage-
nomic studies conducted in environments of varying complexity.

2. Experimental
Approaches
Metagenomic approaches have been applied to a diverse range
2.1. Sample Collection of samples originating from a variety of environments; from
and Processing marine waters to the human gut (6–15). The unifying constant
in metagenomic studies is DNA, therefore, the type of sample that
is suitable for metagenomic analysis is only limited by the quantity
and quality of nucleic acid that can be extracted from it (16).
Technological advances in the amplification of very low concen-
trations of DNA (e.g., linker-adapter amplification and whole
genome amplification via the phi29 DNA polymerase) (17, 18)
have further expanded the spectrum of samples that are amenable
to metagenomic investigation. However, various environments
present their own unique challenges regarding the collection of
representative microbial communities that must be addressed
before DNA extraction can be considered.
The collection of microbes from certain environments, such
as aquatic ecosystems, is rather straightforward, as these tend
to be fairly homogenous compared to other ecosystems such
as soils. Aquatic ecosystems, particularly in the marine environ-
ment, typically contain one million bacterial cells per milliliter
Other documents randomly have
different content
to bring the little ones often to the Church, and let them see them
there.”
I like to think that a great master, painting in those years, had a
thought of dear, little, chubby, dark-eyed boys and girls being taken,
one of the days soon to follow, into the church, when the
masterpiece should be hanging in its place, and making friends with
his “Bambino.”
Not by “sight” alone shall you teach your little ones to know God.
“While they lie against your breast, and you feed them with your
own substance, feed their souls with the sweetness of the love of
Jesus, that wells first from your own heart; let the first words they
utter be: ‘Jesus, Ave Maria, Deo Gratias, Pater noster qui es in coelo.’
“Have a little altar for the children,” the author counsels, “and
teach them the different colours and the different vestments for the
several festivals. Let them ring the ‘Hours’ with their own small bell.
Nor were it ill done to let them preach to you; to the which
preaching do you listen with all due attention and reverence. So
shall they learn more easily, and more exactly, their Christian
Doctrine, and you will have an opportunity of judging what progress
they are really making in it.”
Dominici loves to see the young things gambol about. “As for
games,” he says, “let them run and jump, and gambol and play”—
but never so as to scare away the little playmate, Jesukin. He throws
in a word of warning about the choice of their companions among
the neighbour’s children.
Dominici is not in favour of sending children to the schools, as
things then were. (It must be remembered that he was engaged in a
work of reform among those responsible for those schools.) He
advises the mother to teach the children as much as she can at
home herself. Here the question of what they should study meets
him. Is it safe to let their young and innocent minds come into
contact with pagan morality? He strongly regrets the old days and
the old ways. Our ancestors were wiser. First they taught the Psalter,
and Sacred Doctrine. Then, if the child was to go further: the
Morality of Cato, the Fictions of Aesop, the profitable wisdom of
Prospero (i.e., certain “sentences” taken from the works of Saint
Augustine); the Fidei Confessio of Bœthius; the philosophy of the
“Eva Columello,”[10] the “Tres leo Naturas”[11]; and to help them to
memorize the Sacred Story, the poem “Aethiopium Terras.”
He was writing this book for a woman who had known much
trouble, who had seen her husband’s family driven into exile by the
Medici. It behoves her then to rear up her children in the possession
of that liberty of which no Cosimo can dispossess them—the liberty
which is in the heart of every true man who has emancipated his
Will from the thraldom of his passions. Again and again, he returns
to this point: train their will. Teach them to know Good from Evil,
and to choose good. No man can be free who is not free from these
three things: free from sin, free from vengeance, free from debt. Nor
can a man be free whose soul is in bondage to the appetites of his
body. Rear your children hardily; so shall they have no fear of future
evil fortune. Again he goes into detail: “Teach them to eat bitter
things, lest too great daintiness be their undoing.” And again: “if
they are sick, do not show them too much compassion, for so shall
you take from them the opportunity of practising patience.”
Be careful with whom your children associate. “None of the things
entrusted to you are so precious as your children. Their souls are
worth more in God’s eyes than Heaven and Earth, and the whole of
the irrational creation; and you do Him a greater service in bringing
up your children well than if you possessed the whole world, and
gave all away to the poor.”
When treating of the relation of the child to his parents, Dominici
lays great stress on the observance of those outward forms, which
express the reverence due from him to them. In the presence of
parents, children shall not sit down unless desired to do so; they
must stand in a respectful attitude, humbly bow the head when any
command is addressed to them, and uncover when they meet their
father or mother. “Twice a day at least shall they kneel and beg your
blessing. The child must say: ‘Benedicite,’ and you shall answer:
‘May God bless thee with an everlasting blessing,’ or ‘may the
blessing of God be always with thee.’ And let the child, kneeling to
ask a blessing from you, remind you to ask a blessing from your
Father who is in Heaven—not twice a day only, but as often as you
change your occupation.”
Very noble are the precepts he gives Madonna Bartolommea, to
whom Florence had shown herself, indeed, but an unkind
stepmother. “You owe your child to your country. Therefore, teach
him the duties of a good citizen, and morning and evening let him
pray for the Patria.” Those are precepts we in Ireland might follow
with profit.
I have lingered a little on the educational theories of the great
Dominican, because many of them were put in practice in the school
which Vittorino da Feltre founded at Mantua, under the protection of
the Gonzagas that it is almost unthinkable that Vittorino and
Madonna Paola were not directly influenced by it. We shall now see
these theories in practice.

Part II.—La Casa Giocosa.

On the banks of the Mincio, the Gonzagas, very splendid even in
their Condottiere days, had built a stately Pleasure House, which (for
reasons on which it pleases certain historians yet to dispute) they
called the “Casa Giocosa”—the “Joyous House,” if you care to
translate it thus.
When Vittorino da Feltre came to Mantua, it would seem that the
two elder of the Gonzaga boys were already being educated in this
house. The Court School, where the sons of the reigning house and
of the gentlemen in their service were educated, was an old
European institution. The Palace Schools of Charlemagne are notable
examples.
No trace of the “Casa Giocosa” is now to be found. Even its
situation is a matter of uncertainty; but it has been described for us
so often, and so vividly, that we need but to close our eyes, and
there it stands again on broad meadow lands, that sweep down to
the “slow-gliding Mincio,” in the midst of its fair lawns and terraced
gardens, with its avenues of acacias and plane trees, its hedges of
roses, its shady courts and fountains and statues, its marble “loggie,”
and frescoed, spacious chambers, full of air and sunshine—a
memory and an inspiration for all the schools that have followed it
since.
Now, if ever there was a man who was that rare thing, a
schoolmaster born, it was Vittorino da Feltre. He had discovered his
vocation in his student days at the University of Padua, where he
had kept body and soul together by teaching backward students,
and helping them to keep up with their classes. He had developed it
at Venice, where he had shared the management of a school with
the great Greek scholar, Guarino. But at Mantua, whither he came, a
ripe man of forty-five or so, with a large experience of boy-nature, a
thorough mastery of methods of teaching, a store of well-tested
theories, and a new field on which to exercise them, he was like a
prince coming into his own kingdom. There seems something
providential in the way things were arranged for Vittorino da Feltre.
One is reminded of the Providence which we shall see in a later
paper, surrounding the foundation of another great educational
institution—Madame de Maintenon’s Saint-Cyr.
One condition Vittorino made before he took over the “Casa
Giocosa.” He was to be supreme master in the establishment. Even
the parents were not to interfere, and there was to be no appeal
from his decision, whether as to the studies of the children, their
food and manner of life, or their companions. On the last point, he
was destined to meet with grave opposition. Some families claimed it
as one of their immemorial privileges to send their children to the
Court School, to be educated with the Princes. But, privilege or no
privilege, Vittorino would tolerate nobody at the school whose
example was likely to be harmful. Here you see him putting into
practice one of the most constantly reiterated of the precepts of the
Blessed Giovanni Dominici: “Be careful of your children’s
companions.”
Not alone of the companions of the Princes was he careful, but of
the servants. It would seem that when Vittorino took over the “Casa
Giocosa,” he found a whole troup of liveried menials ready to
minister to the slightest wish of the young Princes. His first care was
to send them all off, replacing them by a few trustworthy attendants,
not numerous enough to make discipline difficult. He put porters at
the gate, on whom he could rely, wishing to secure for his school the
atmosphere of quiet, and work, and prayer, and order, and
wholesome austerity in which the young souls confided to him might
grow to their full perfection. The vicinity of a splendid court made his
precautions all the more necessary.
With the troops of servants disappeared the soft carpets, and
luxurious couches, the gold and silver plate, and, above all, the rich
foods which were playing such havoc with poor Lodovico’s figure—
not to speak of anything else. Everything was to be plain and
wholesome and abundant, but there was to be no luxury. Had not
the Blessed Giovanni warned Madonna Bartolommea against rearing
her children too softly? “Rear them hardily” had been his advice.
“Teach them to eat bitter food and things unpleasant,” so shall they
be able to say “we care not” when life is hard with them in the years
to come.
The little picture-children for whom Dominici pleads, that Madonna
Bartolommea’s boys and girls may make friendship with them early,
were not forgotten by Vittorino. Be sure, when, at his request,
Madonna Paola sent her painter to cover the walls of the study-halls
and galleries with “frescoes of children playing,” the tiny Jesukin,
and dear Saint John, who was as an elder brother to Him, were to
be seen there, playing together—in the carpenter’s shop, mayhap,
when Elizabeth had taken her boy to Nazareth to visit his small
cousin, or by the covered well under the palm-tree in Zachary’s
garden, when the sweet spring days had called Mary and her
bambino to the hill-country.
Poor Lodovico must have felt the change rather hard at first. He
had been accustomed to get up whatever time he liked, do as little
as he pleased, and have his interest aroused by nothing except
questions of eating and drinking. One really thinks there must have
been something diseased in poor Lodovico’s extraordinary appetite.
Our ancestors would have seen in him a fellow victim of Cathal
MacFinguine, King of Munster, in whom there entered the Demon of
Gluttony through eating the apples, whereon the Scholar of Fergal,
son of Maelduin, had put his heathen charms, “so that the demon
abode in the throat of Cathal, to the ruin of the men of Munster
during three half-years; and it is likely he would have ruined Ireland
during another half-year.”
At all events, Vittorino took Lodovico in hand at once. He was only
allowed to eat at mealtimes; but at first his meals were set at short
intervals, and there was no stint on the quantity of food; which,
however, was as plain as possible, so that the appetite should not be
over-stimulated.
Gradually, the meals became fewer, and Vittorino, discovering that
Lodovico’s voracity was as much the result of an empty mind and
starved interests as anything else, had the inspiration to accompany
them with little entertainments. He had singers, and musicians, and
storytellers placed in the dining-hall; and, lo! Lodovico, listening to
them, forgot his plate, for the nonce, and did not notice the loss of
it, when an attendant, on a sign from Vittorino, bore it quietly away.
With Carlo, the second son, the master pursued quite a different
plan. Carlo was growing too fast, and spending his energies too
rapidly in the ardour which he put into his games. Vittorino saw to it
that the boy should have as much to eat as he wished at meal times.
Between meals he had permission to eat, too—but only bread.
His attention to the food of his pupils and to their bodily welfare
give the key to Vittorino’s whole educational system. The “Mens sana
in corpore sano,” is as an educational ideal, not the exclusive
possession of the Greeks. I have tried to show how vitally it
influenced education in the Middle Ages; but, undoubtedly, Vittorino
found new ardour for his pursuit of it in the image of the “Academy,”
which his Greek studies had conjured up for him, and kept
constantly before his mind.
But the little boys and girls, for whom Vittorino was going to
assume responsibility, had something more than a body and a mind
to be developed. They had each an immortal soul. The recognition
of that fact must, logically, alter any system of education, not
founded on it, into something very different. And so the Christian
school of Mantua, forming colonists for this world, and citizens for
heaven, was something essentially different from the Platonic
Academy, however much it may have borrowed from it.
The strengthening and developing of body, and mind, and soul—
that is, in a few words, what the whole system of Vittorino aimed at.
It is in the harmonious ordering of the different studies and
exercises, which he chose for the perfecting of the three parts of
men, that the chief excellence of the “Casa Giocosa” consists.
“You are rearing your children for God,” the Blessed Giovanni
reminded Madonna Bartolommea. Vittorino never forgot this for a
moment, nor did he ever allow his pupils to forget that they had
been “created and placed” in this world by God, “to know Him, love
Him, and serve Him, and by that means to gain Everlasting Life.”
The common day began with hearing Mass in the school chapel.
After Mass, the Office of Our Lady was recited, and a short
instruction in some point of Christian Doctrine was given to the
school. But this teaching was not theoretic only. He taught them not
only what Christians must believe, but showed them how Christians
must act. And so we find his pupils, in the after years, distinguished,
among all the men and women of Italy, by their practical Christianity.
Lodovico, once the self-indulgent, grew into the Marchese Lodovico,
chaste and sober, and wise and kind—the best of the Gonzaga
Princes. Little Frederick Montefeltro, sent as a hostage to the
Mantuan Court, rejoiced, in the years when men spoke of him as the
“Good Duke of Urbino,” that the chances of war had brought him
such good fortune as to make him the pupil of such a master. Ever
before his eyes he would have that master’s image, and as much as
might be, he would carry out his system of life in the order of the
Ducal Palace. So Vittorino’s portrait adorned one of the walls of the
famous palace, with these words written under it: “In honour of his
saintly master, Vittorino da Feltre, who by word and example,
instructed him in all human excellence, Frederico, has set this here.”
And better still, “the Court of Urbino was framed on the precepts
which the Duke had learnt in the Casa Giocosa, and became in its
turn a school where Italian princes sent their sons to be trained in
knightly exercises and elegant manners.” And so we trace, in a direct
line of descent from the “Casa Giocosa,” “the best book that was
ever written upon good breeding,” as Doctor Johnson testified to
Boswell: “‘Il Cortegiano’—the best book, I tell you, and you should
read it.” An advice which one could not do better than repeat.
In an age when men lied and deceived shamelessly, Vittorino’s
pupils were known for their absolute sincerity. This love of truth and
hatred of falsehood was not won without careful efforts on the part
of the master. He would not punish for a fault that was bravely
confessed, and so took away one of the occasions of lying from timid
children. A funny little story is told of Alessandro Gonzaga. The little
fellow was ill, and had been ordered not to drink any water. But he
was horribly thirsty, and disobeyed the commands. There was no
great danger of being “found out,” but the boy was uneasy, until he
had confessed. “Do you know what I did?” he said to Vittorino. “I
took a big, big drink of water. Wasn’t I very good?” “Well,” said
Vittorino, seeing that it could not be helped now, “at least, you were
very good to tell it.”
He never allowed his pupils to utter a profane word. When Carlo
was quite grown up, he swore a soldier’s oath in his master’s
presence. And lo! the little man was upon him in an instant, boxing
his ears, as if he were still a schoolboy. To the honour of Carlo, it
must be said that he bore the indignity meekly, feeling that he had
deserved it.
Like Dominici, Vittorino loved to see the children run about, and
laugh, and leap, and play. He found two of them, during recreation
hour one day, confabbing in a corner about their lessons. Do you
think he was pleased? Not a bit of it. Out he routed them, and made
them take part in the other children’s games. For, long before the
English Duke, he had found out for himself that many a battle yet to
be fought was being won already on those meadows by the Mincio,
where his pupils were playing merrily.
One of the outstanding features in Vittorino’s system was the
importance he attached to games—and all sorts of physical
exercises. He held as a fundamental principle that “the human spirit
cannot exercise its faculties fully, if the physical organs which it must
use are defective.” He insisted on outdoor exercise, whatever the
weather. He had his pupils taught riding, and swimming, and
wrestling, and fencing. He organised hunting and fishing expeditions
for them; and, remembering that many of his pupils were to be
soldiers, he liked to teach them the art of warfare, by occupying
mimic trenches, and pitching mimic camps, and taking mimic towns
—according to the most approved methods.
These rougher plays were not for the girls, though they, in
general, shared the lessons of the boys. For them there were
dancing lessons, where every movement of their body was trained to
an exquisite grace. They had riding lessons, too, and hunted and
played “palla,” or tennis. No game was tolerated for them which
would tend to make them ungraceful—as so many of the games our
girls play to-day really seem to do.
Plenty of fresh air and exercise, plenty of good, simple food, to
which they brought the sauce of a healthy hunger, sound and
dreamless sleep, soon made the youngsters of the “Casa Giocosa” a
healthy, happy band, whom it was a delight to see at their lessons.
He had the supreme gift of the good teacher, our Vittorino, that of
knowing how to interest his pupils in their work. The maxim of
Quintilian “do not allow the boy to conceive an aversion for the
studies which he cannot yet love,” was adopted by him, and, until
the young minds were ripe enough to love learning for its own sake,
it was the master’s care to surround it with attractions. So we find
him reviving the Quintilianian device of teaching the youngsters to
read by means of painted cards and wooden blocks. As we have
seen already, our Irish ancestors had a similar plan. In Whitley
Stokes’ “Lives of the Irish Saints from the Book of Lismore,” there is
a charming story of Saint Columbkille, as a child, learning his letters
from a cake, on which they had been stamped.
In Vittorino’s school there was no place for the rod, which we have
seen play such an important part in the mediæval school-system. He
liked to appeal to the children through their sense of honour and
dignity. The greatest punishment for such children was to make
them feel ashamed.
As far as school-work went, however, there was little need for
punishment. The enthusiasm for letters which had seized on all Italy
had taken possession of the little people of the “Casa Giocosa” in an
extraordinary degree; and in some cases, especially that of
Gianlucido, the master’s care was rather to restrain than to urge
them on. That great scholar, Ambrogio Traversari, General of the
Camaldolese Order, writing to his friend, the celebrated Niccolò
Nicoli, mentioned the boy’s marvellous achievements in Latin and
Greek. On the occasion of a later visit, which he described for the
great Cosimo de’ Medici, he listened to a Latin poem of about two
hundred verses, wherein Gianlucido celebrated the coming of the
Emperor Sigismund to Mantua.
That same letter to Cosimo makes mention of Cecilia, whom,
perhaps, my readers think I have left too long undistinguished,
among the band of merry children, playing by the Mincio. “There is
also a daughter of the Marquis[12] at the school,” writes Ambrogio,
“who, though only ten years’ old, writes Greek with such elegance
that, I am ashamed to acknowledge, scarcely any of my own pupils
can approach it.” This tribute, indeed, hardly does full credit to
Cecilia’s astonishing attainments. At eight years old, we are told, she
read the works of Saint John Chrysostom, and wrote elegant Latin
verses. She had begun the study of Greek at the mature age of six.
Nor did she excel less in the feminine arts, on which her master’s
educational system laid such stress. Those little, high-born, Italian
girls learned to dance almost as soon as they learned to walk, and a
suggestion of the music which accompanied their early dancing
lessons lingered in every graceful movement. Music, too, was as
general as speech, and the child learned it as naturally. But, general
as it was, it was never cheapened by being wedded to unworthy
words. When a little girl learned to sing, there was food for her
intellect in the lesson, too; for, in those days, men set sonnets of
Petrarch and passages from Virgil to music, and the lute made a
charming accompaniment.
The “speaking” voice was even more carefully trained than the
“singing” voice. To quote from the delightful “Life of Vittorino da
Feltre” in the Saint Nicholas Series: “the greatest trouble was taken
with the cultivation of the voice, the manner of breathing,
pronunciation, and all the other details which go to make up an easy
and elegant delivery.... Like the ancient Romans, the master
attached to this exercise a certain hygienic value.” It was a rare treat
to hear Cecilia, in that golden voice of hers, declaim some of her
own verses.
It is worth while to examine, in some detail, the system which led
to such astonishing results.
Those painted cards and blocks, of which I have spoken, had been
designed to teach the child to read Latin. The thing was not so
surprising in those days as it would be in ours. As a matter of fact, it
was as short a step towards the “unknown from the known” (the
safest of pedagogic principles) to teach a child, whose mother-
tongue was the speech of Lombardy, to read Latin, as to teach him
or her to read Italian. So the children learned to read Latin very
young indeed. Unless Cecilia was an exception, they learned to read
Greek very young, too. The practice was to translate Greek into
Latin.
Later on, the pupils took up the study of Grammar. The rules of
Latin Grammar were deduced from a careful study of the works of
Virgil and Cicero, while those of Greek were formulated while the
pupils studied Homer and Demosthenes. The barbarous system,
from which we are just emerging, which made the study of
grammatical rules precede all else, was the unfortunate discovery of
the century following Vittorino’s.
History was studied in the pages of Sallust, Valerius Maximus, and
Livy.
Vittorino’s practice was to make his pupils read aloud, insisting on
good pronunciation and artistic delivery. He made them learn off by
heart, too, the best passages of the poets, orators, and
philosophers. And so the children had faultless models to hand,
when it was time for them to address themselves to original
composition.
To balance any tendency this practice might have had to make his
pupils adopt other people’s thoughts ready-made, he put them
through a very thorough course of mental gymnastics. He aimed,
with these exercises, to win for his pupils rather strength and vigour
of mentality than subtilty. “I want to teach them to think, and not to
split hairs,” expresses a pedagogic maxim of his, of which all his
biographers have taken note. He made the youngsters propose
difficulties to him, or raised them himself for them, and helped them
to solve them. The Mathematical training given in the “Casa Giocosa”
was the best in all Italy. At none of the Universities were
Mathematics taught in a manner so profitable, or their educative
value so fully realised.
When the children were a little older they took part in certain
oratorical exercises, the idea of which the master had borrowed from
the “Schools of Rhetoric” of Antiquity. Many of these boys would, in
the years to come, be employed in Diplomatic Missions, and nothing
could more fittingly prepare them for such work than these
“Disputations.”
Such, in broadest outline, was the education which made of Cecilia
Gonzaga, at the age of sixteen, one of the most charming and
accomplished ladies in all Italy. Many a princely suitor came riding
over the long bridge of San Giorgio to lay his hand and heart at the
feet of the Marchese’s golden-haired girl, whose beauty and
attainments had set their poets singing. One, in particular, found
favour with her father—Oddantonio of Montefeltro, elder brother of
that Frederico who had been Cecilia’s fellow-pupil at the “Casa
Giocosa.” Oddantonio saw Cecilia, for the first time, on the occasion
of Margherita’s marriage with Leonello of Este, fell in love with her
immediately, and formally asked for her hand six weeks later. The
alliance proposed was one that offered immense political advantages
to the House of Gonzaga, and the Marchese eagerly accepted it,
though the reputation of his prospective son-in-law was none of the
best. In the off-hand way fathers had in those days, when it was a
question of arranging their daughters’ matrimonial affairs, the
Marquis sent for Cecilia one day, and told her to hold herself ready
to be married.
But another Lover than Oddantonio had won Cecilia for Himself.
The little Jesukin, with Whom she had played her childish games,
Who, grown to manhood, had changed the water of the old
philosophies into the wine of truth for her drinking, Who had
sanctified the dust of the earth’s materialism because His Feet had
touched it, and made the World a Sacred Place because He had died
in it, was the Beloved of her heart.
As Magdalene, on a day, had broken the “alabaster box of
precious ointment,” and anointed His Feet therewith, so, too, was
Cecilia ready to pour out at His Feet all the treasures of heart, and
mind, and soul, which Vittorino’s teaching had helped her to gather;
as Magdalene spent the beauty of her hair to wipe the Feet she had
anointed, so, too, was Cecilia ready to lay down her beauty for His
dear sake.
Not many yards from the Castello of Mantua was a Convent of the
Poor Clares, founded by Madonna Paola. Here was the abode which
her Beloved had chosen for her, and here He offered her the habit
which Francis had bestowed on Clare, the rope-girdle, the coarse
veil. Oddantonio’s jewels and gifts, the satins and laces, and cloth of
gold in which he would have decked his bride, were dross in the
eyes of her, whose chosen ornaments were the jewels of Madonna
Poverty.
When she announced her intention of becoming a Clarice to her
father, his rage knew no bounds. One blushes to tell it (for
Gianfrancesco, with all his faults, has a way of making us like him),
but it must be told: he actually used physical violence to the poor
girl to compel her to do what she was told. But steadfast as Clare
herself, Cecilia stood firm, finding a little comfort in the unfailing
sympathy of her mother and her teacher.
For two years the struggle lasted—Vittorino and Paola managing,
between them, to dissuade the Marquis from forcing on his
daughter’s marriage with a Prince, whose name was beginning to be
in all men’s mouths as that of a notorious libertine. It may well be
that Gianfrancesco was a little ashamed of himself when he was
forced to recognise the true character of his chosen son-in-law. But
ashamed or not, he was no more ready to see his brilliant girl bury
herself in a Poor Clare’s Cell. To the last, he refused his permission
for her entrance into religion; and Cecilia, fearing to bring disaster
on the Convent she had chosen, was forced to acquiesce.
But a day came when Gianfrancesco had power to make his will
felt no longer—a day when he lay very still and cold on a bier of
black velvet, and was borne to his tomb in San Francesco.
Curiously enough, the Church wherein the great Marchese lay
buried was part of the Convent Cecilia had been so anxious to make
her home. Nothing stood in the way of the accomplishment of her
heart’s desire now; and so it came to pass that the Vows he had so
long refused to allow his daughter to take were uttered over
Marchese Gianfrancesco’s tomb.
MARGARET MORE
A Little Schoolgirl of Tudor England

Part I.
At the foot of the river-stairs nearest the Westminster Law Courts,
you might have seen (in the days when the sixteenth century was
yet in its teens, King Henry the Eighth, a slim young Prince—the very
flower of knighthood—and the Thames, a silver highway of
romance,) a private barge, with a couple of blue-coated serving
men, waiting for their master. And presently down the steps would
come a man with brown hair a little tumbled, and dress a little awry,
after a long, hard day’s work in the Courts. Something in the gait—a
little defect, one shoulder somewhat higher than the other—might
strike your attention, and you would turn to a water-bailiff near you
with a question: “Is this Master More?” Then—whether intentionally
or not, your whisper having carried further than the ear for which it
was ostensibly intended—you would see the uneven shoulders swing
suddenly round, and from half-way down the steps a clever face—
wonderfully attractive in its irregularity—with a humorous mouth,
and merry, grey eyes, would be lifted to you, and a laughing voice
would proclaim its owner, “Thomas More, indeed, and very much
more at your service.”
If upon being further pressed to know in what more he could
serve you, you were well enough advised to make the request to be
rowed down the river with him to Chelsea—there to make the
acquaintance of his daughter, Mistress Margaret, and the others of
his “Academia,” not to mention his second wife, Dame Alice (for
whose solid, if somewhat Philistine, qualities you have the highest
regard), and Master Gunnel, and John Harris, and Henry Pattieson—
with all of whom you already seem to yourself familiar, from
Erasmus, his letters—you would find yourself comfortably seated in
the stern of the barge (before you had time to enlarge on the
reasons, which had emboldened you to make your request), and
being borne on your pleasant way down the pleasant, shining
Thames.
Oh! a very pleasant way, in good sooth! The river covered with
barges that carry bright colours, and music and laughter, and its
banks covered with gardens that let the evening breeze rifle them
for sweetness; the wooded hills that fill in the distance, brave in
their new summer greenery, and the kindly sun, the giver of all
these good gifts, so loth to leave the sight of them that he sinks but
slowly, slowly to his bed in the West!
And yet methinks the most pleasant part of all would be yet to
come. It would be waiting for you at a certain steps, towards which
you might have seen your host, long since, strain his eyes. A group
of young things are standing at the top, waving their scarves. Two of
them, a little boy and a girl, so near a size that you take them for
twins, are in such haste to get to the barge that they are in danger
of tumbling right down the steps into the river. You can hear a girl’s
voice call at them anxiously, “Cecy! Jack!” and when the barge is
fastened to its moorings, and you are mounting the steps, leisurely
enough to give your host a good start of you, you look up and see
those two troublesome little monkeys held fast by the hands of a tall
girl of fifteen or so, and you know by the way her father turns to her,
first of all, that this is Margaret.
In the meantime your host is being pulled, very affectionately,
from one to the other. Margaret’s restraining hand is not strong
enough to keep Jack and Cecily in check any longer, and with those
two rifling his pockets for barley-sugar, and Bess and Daisy hanging
out of his arms, one on each side, and Margaret Giggs a little in the
background, and young Will Roper, and Jack Dancey, and Rupert
Alington dancing around, one understands why he cares not to be
over-careful of his clothes.
Going up the garden path to the great house you will meet a
stately lady stepping sedately down from it. If her welcome has a
touch of frigidity, lay it not to her charge, good lady. In truth, her
lord might have given her a little warning that a stranger was
coming to supper. Then had she time to get Gillian to add a dish of
black-caps and a lèche to the bill of fare, and herself to change into
her scarlet gown and coif. Whereas, now!
Indeed, the fare is plain enough, as you will presently discover
when you are seated at Master More’s right hand at the long table in
the great hall. But dainty though you be at your sizes, on ordinary
occasions, it will be odds if you have ever set down to a meal more
to your taste, or eaten anything with a greater appetite than the salt
meat, and coarse fish, and thick slices from the cob-loaf, flavoured,
as these are, with the “Attic Salt,” for which this house is famous.
After supper someone will suggest a stroll through the garden;
and you will accept the more readily since you hear Dame Alice say
that Gillian needs her superintendence in the kitchen. As you rise
from table your eye, through the long, wide lattice, catches a
glimpse of glowing flower-beds and blossoming hedges, and you
compliment your host on the beautiful home he has made for
himself. Is it fancy, or does a slight shadow really fall on his laughing
face, as if he felt in how short a time he must bid it all good-bye?
It would seem as if Margaret noticed the little cloud also, and her
homely, clever face, so like her father’s in colouring and feature and
expression, reflects it lovingly. But she knows how to conjure away
his sadness. “Shall we not go to the Academia first, and show you to
what good use we have put the day?” she asks him, laying her hand
on his arm and turning her dear face up to his.
“Well proposed, Meg,” he says, tucking her arm under his own,
and so leading the way up the broad oak stairs—you following
among the others.
“How charming!” is your first exclamation as you enter the
schoolroom. And, indeed, you are right. No more delightful room can
be imagined than this panelled-oak chamber, with deep, low, roomy
window-seats, and classic tapestry, flapping in the cool breeze from
the river. After you have spoken a word with Master Gunnel, the
tutor (whom you have noticed slip away early from the supper table,
and find again here with young John Clement, with a Greek text
between them) you will be conducted to the various desks, and
shown their contents by their several owners. On Bessy’s you will
find a “Livy” most probably, on Daisy Middleton’s a “Sallust,” and on
Margaret’s a “Saint Augustine,” with her father’s marks “where she is
to read and where desist.” Then Master Gunnel will conduct you to
his own high desk, and take therefrom some of their traductions, at
the purity and elegance of which, if you have any skill in Latin style,
you will be completely amazed.
Though you compliment Master Gunnel on the proficiency of his
pupils you know, and he knows, that the credit is all due to their
father. Even in his busiest years it has been his chief occupation. If
you had time to go over the letters which Margaret treasures so
dearly, and which you may see (tied up like a lover’s in blue ribbon)
in a safe corner of her desk, you would find, not once, but many
times repeated, words like these: “I beg you, Margaret, tell me
about the progress you are all making in your studies. For I assure
you that, rather than allow my children to be idle and slothful, I
would make a sacrifice of wealth, and bid adieu to other cares and
business, to attend to my children and my family, amongst whom
none is more dear to me than yourself, my beloved daughter.”
“Jack! Jack! What has become of Jack?” Margaret looks around
anxiously; but for once Jack is not in any particular mischief, and
comes up to his father with a look of self-satisfaction at the fact,
which is infinitely comical.
“Look,” says his father, “how the little monkey knows already that
he is going to be praised for the Latin letter he sent me to Court by
the hand of the Bristol merchant.”
He takes the little chap between his knees, and strokes his curls
while he talks to them all. Indeed, each of them had done very well,
and it was not only because he was their father and loved them
dearly that their letters had given him such pleasure. Their letters
were very good; the thought very well put; the Latin pure and
correct. But John’s pleased him best of all, because it was longer,
and showed that he had taken more trouble with it than the others
had done. It was funny too, and some of his own jokes had been
turned very wittily against himself; the which pleased him not a
little. But even in this matter John had remembered not to go too
far, and while he thrust and parried very prettily, he never forgot that
he was fencing with his father.
With that Cecy claps her hands in delight, for whatever of good or
ill befalls Jack is her hap, too.
“There is a mount for thee, too, Cecy,” her father promises her,
and takes her and Jack, one on each knee, and goes on with his
discourse.
When he is away from home he will expect a letter from each of
them every day. He will not take excuses such as Jack is wont to
make, that he has not time, or the carrier went off before he knew,
or, forsooth, he had nothing to write about. As for want of time, how
could it fail, since everyone who has anything to say in the division
of their day will let the letter to father take first place. And as for
keeping the carrier waiting, why not have the letters ready and
sealed, even before his coming? And as for having nothing to say—
did anyone ever hear of such an excuse from girls, who (he pulls
Cecy’s ears at this point) have always a world to say about nothing
at all. If there is nothing at all to write about, why! let them write
about “nothing at all.” But they know he likes to hear about their
studies and their games. But whatever they write, whether it be fun
or earnest, let them write it as carefully and with as much finish as
possible. It will be no harm to write out first the whole in English, for
then they will have much less trouble in turning it into Latin; not
having to look for the matter, their mind will be more free to attend
to the language. That, however, he will leave to their own choice;
but on another thing he will be strict. Whatever they have
composed, they must carefully examine before writing out clear; and
in this examination they must first scrutinise the whole sentence,
and then every part of it. Thus, if any solecisms have escaped them,
they will easily detect them. By this diligence they will soon be able
to turn out elegant productions.
“And have them shown to the Archbishop, or Dean Colet, or even
the King, as Erasmus did with a letter of Margaret’s,” says Cecily.
A little shade comes over the kind face above her curls. If there is
one thing he dreads for his girls, this wise father, it is vain-glory.
“Tilly-vally, Master More,” says Dame Alice, bustling in (just at the
right moment, to show what a sensible choice he has made of a
step-mother for his brilliant girls). “What comes over you to keep the
girls all idling here, while Gillian needs them in the dairy, being all of
a sweat, poor wench, a-trying to make the butter come? Off with
you all, girls, now, and take your turns at the churn until the cream
breaks, were it to keep you to morning.”
She leads the relief-party off to the dairy, and you find yourself
alone with your host and Master Gunnel.
“Shall we to the garden until the young ones come back to us?”
Master More inquires, and you need no second invitation.
What a beautiful garden it is! Even though so many of the flowers
have gone to sleep, you know you have never been in so beautiful a
garden in all your life. All sorts of sweet perfumes come to you as
you seat yourself between your host and Master Gunnel in the
pavilion that gives such a charming view of the river. You would like
to know some of the names of these so sweet-smelling flowers and
herbs that you might perfume your own garden with them.
Sayth Master Gunnel: “It is a pity that Mistress Margaret is not
here, for she knows the name of each of them, and their nature, and
their uses.”
Margaret’s father laughs. “If Margaret is not spoiled, methinks it is
not to her tutor she owes it—for he is always ready to blazon forth
her praises. I am glad to think, however, that she has good skill in
herbs. It is that the children may learn the uses of common things
that I keep in my garden and paddock many a plant which the
fastidious would cast forth. A woman should have good knowledge
of healing.”
“And of what else?” inquires Master Gunnel, innocently.
A merry laugh from your host. “Look what artifices he useth, this
good Gunnel, to get me to mount my favourite hobby. You must e’en
take the consequences, if it rides off with me.”
And with that he is off in good earnest, and you are minded to
lose no word of what he says about the way a girl should be
educated.
“In your country,” he says (turning to you), “which would have
been mine, too, had not one of my ancestors left Ireland for
England, I have heard it said that embroiderers ever kept before
them, stamped in a piece of leather, the pattern of that design which
they wished to imitate on church robes and vestments. Now, even
such a pattern, stamped on the imperishable leather of Holy Writ,
lies to our hands; and I know that good Master Gunnel here (of
whose work one may speak in a manner, not all too fanciful, as
resembling that of the embroiderer) puts in never a stitch without
looking carefully at the model. Is it not so, Master Gunnel?”
For answer Master Gunnel begins to quote the glorious words:
“Who shall find a valiant woman? far, and from the uttermost coast
is the price of her. The heart of her husband trusteth in her, and he
shall have no need of spoils.” And so on, till the picture is complete,
and the “Valiant Woman” stands out before you, strong, and wise,
and chaste, and kind, and sweet, in all her imperishable beauty.
The hour is exquisite. Sweeter and sweeter grows the garden, as
the dew distils new perfumes. The paling river is pricked here and
there by a rare star; but in the sky itself, from where you sit, you
can only see one, and that is Venus. In the faint light your host’s
face, raised to it, shows very soft and dreamy. Is he thinking of the
wife of his youth, the dead mother of his girls?
Presently he begins to talk again. “If I were a preacher, or a
moralist, or anything but a lawyer, trained only to look for the flaws
in all things, I could show you how in that one passage of Holy Writ
is contained a whole treatise on the Education of Women. But
Master Gunnel shall do it for me.”
“Right willingly,” declares Master Gunnel, “if you will but show me
how.”
“I would have you in the first place note that the ‘Mens Sana in
Corpore Sano’ hath never been more clearly indicated than in that
picture. For the healthy body, you shall see it mentioned not once
but many times, and you shall guess at it, too, by the laughter and
good humour which she carries down into her old age. ‘She hath
girded her loins with strength, and hath strengthened her arm’—as if
to show that this strength and suppleness of body, so admirable in a
woman, were to be cultivated by suitable exercises; as to which, to
speak sooth, none are so well adapted for the purpose as those she
finds ready to hand in her household tasks, sweeping, kneading
bread, churning, spinning.”
“At all of which she was proficient, this ‘Valiant Woman,’” puts in
Master Gunnel. “‘She hath looked well to the paths of her house,
and hath not eaten her bread idle’, and again: ‘She hath sought wool
and flax, and hath wrought by the counsel of her hands.’ ‘Her fingers
have taken hold of the spindle.’”
“As for her good humour,” continues your host, laughing a little, “I
would ask your opinion whether it is better shown in anything than
the excellent terms she always managed to maintain with her
husband.”
“Of a truth,” sayth Master Gunnel demurely, “the fact proveth that
she suffered not from megrims, to which effect I, for one (who
believe in the healthfulness of the morning hours), consider her
early rising much contributed.”
“Ah! Master Gunnel,” says More, standing up, “you will be able to
write that treatise without any help from me.”
Here you put in a word, and entreat Master More to develop the
matter further.
“And you will,” he promises you; “but let us climb to the roof of
the new building, where I have promised to have the young ones,
and question them on their knowledge of the stars.”
Under the great dome of the starry sky the conversation takes
another tone—deeper and more serious. He holds, you gather from
what you hear him say, that those who trained the mind and soul of
that woman were not afraid to feed them with the food of strong
thoughts. He discovers a strength and sureness about all her
dealings, a big and generous way of regarding things that show a
well-nourished, well-balanced mentality. That little touch about her
concern for the well-being of her household: that they be generously
fed, and warmly and comfortably clad, seems to him to indicate a
wider outlook than the prejudice which confines woman’s studies to
the petty things of life would tend to foster. “Be sure of this,” sayth
he, “she is not one of those who are penny-wise and pound-foolish,
saving a candle’s end and spoiling a velvet gown.”
“That she was well-read,” says Master Gunnel, “is not without
warranty.”
“Now, how may that be?” you inquire; and Master Gunnel
instances her clever and sensible conversation, which, he holds
reasonably enough, was not acquired without reading, and study,
and listening to the conversation of learned men. “I take it,” he says,
turning to More, “that we can interpret this, what is further said of
her: ‘She had opened her mouth to wisdom, and the law of
clemency is on her tongue.’”
But here there comes a sound of laughter thrilling through the
garden, and a scamper of light feet up the steps that lead to the flat
roof of the new building, and the whole Academia, with the
exception of Jack and Cecy (who have been attacked by “Johnny
Nods,” and carried off to their respective beds), are here to tell all
the frolic they had in the dairy, and how long it has taken for the
butter to come.

Part II.
When you found yourself last night in the oak bed-chamber, which
Dame Alice had assigned to you as the pleasantest in the house, you
felt strangely disinclined for slumber. So you set the wax candle
(which had been borne before you very ceremoniously, to light you
to your quarters) in a place secure from the night breeze, and,
unbolting the heavy wooden shutter as noiselessly as possible, you
opened your lattice, and stepped out into the balcony—out into the
night that was sweet with flowers and starlight.
Then, as you sought among the stars for those with which you
had made friendship when you formed one of the little group of star-
gazers on the roof of the new building, you seemed to hear again
the voice of your host. How droll he had been as he pulled Daisy’s
pink ear, and praised her for that she was able, on occasion, to tell
the sun from the moon! But, presently, the laughter and bantering
had died away, and you found yourself listening in a delicious
hushed expectancy for a whisper of the music of the spheres as your
host’s words made you think of them as moving harmoniously,
carrying each its appointed luminary like a blazing jewel set in a
crystal circlet! Ah! truly the “Almagest”[13] would make a man a poet
in spite of himself; and now you know how a certain look in
Margaret’s eyes came there. For who could gaze, night after night,
into the great spaces wherein the revolutions of the spheres make
melody, and around which the fixed stars are built, in their
firmament, into a mighty battlement, without carrying some of the
wonder and the glory of it all away in one’s own soul?
In the lighted hall, afterwards, cozy with candlelight and a great
log ablaze on the wide hearth, you came back very gently to earth.
Such a good earth and a kind earth; not so very far from heaven
either, since there was love, and light, and music to keep the
roadway free!
Here Dame Alice, taking her capable part in the concert of
instrumental music (which you learned is a nightly event in this
household), relaxed a little from her attitude of housewifely
overcarefulness, and showed you a pleasanter part of her nature.
And when you looked round the circle and saw the happy looks of
each little performer on harp, and lute, and monochord, and flute, it
is odds but your pity was stirred for certain little girls you left behind
you in the twentieth century, who spend such miserable, profitless,
lonely hours in “practising.” If the “practice hour” were such a jolly
re-union of the whole family as it is here, be sure our little maids
would get more good out of their music-lessons!
Gradually the instruments were laid carefully aside, and the maid-
servants, who had been busy with their spinning and sewing during
the concert, prepared the place for the night prayers. One thing
surprised you no little, and this was the accuracy with which
everyone in the household recited the alternate verses of the psalms
chosen: “Miserere,” “Ad Te Domine levavi,” and “Deus misereatur
nostri.”
Surely the blessing of God rested visibly on this home, where
everything was done to show Him perpetual honour! And so with a
sense of great spiritual peace in your heart you came away from
your star-watch on the balcony, and presently were lost in blissful
dreams in the huge four-poster bed, with its downy pillows and
sheets that smelled of lavender.
And now it is morning again, and the sun is streaming through the
chink in the wooden shutter, which you neglected to fasten properly
last night. Someone below in the garden is singing, and you speed
your toilet to the merry tune and time:—
“The Hunt is up, the Hunt is up,
And it is well nigh Daye,
Harry our King has gone hunting
To bring his Deere to baye.
The East is bright with Morning Lighte,
And Darkness it is fled,
And the merrie Horn wakes up the Morn
To leave his idle Bed.
Behold the Skies with golder Dies.
Tra la la la la la!”

“Up with it, young ones; up with the burden though it do come
before it’s time,” says the merry voice of your host. And certainly
they take him at his word, until the thrushes and blackbirds start
singing, too—in self-defence.
“Shall we visit the menagerie now?” queries Master More, when
there comes a moment’s silence in the wake of the “good-morrows”
which greeted your appearance. “Nay,” as you look round the groups
for an explanation, “these be not the only wild animals we keep in
the enclosure.”
And then you look again, and see that everybody has some
feeding stuff in his or her hand, and you find yourself presently
engaged on a round of visits to the quaintest and most varied
collection of pet-animals you ever dreamed of. There is one condition
laid down in this household for all who would own a pet in it, and
that is that the whole care of it devolves on the owner. Methinks in
this there is a fine training in thoughtfulness and in the sense of
responsibility as well as in Natural History.
There is a little time to spare yet, it seems, before the bell rings
for Mass, and you willingly accept the invitation to pass it in the
study in the new building. “And Meg shall come with us, too,” your
host promises, “but for the others, I would ask no four walls to try
and hold them while they be in such spirits.”
So off they go scampering round the garden, the wild young
things!
In the new building you find a long gallery lined with books, which
leads to a charming little room built all for study and retirement. On
the broad oak table lie leaves of the manuscript which has occupied
Master More during long hours while all the world beside slept. “Oh!
Father,” says Margaret reproachfully, “what a state your desk is in,”
and, thereupon, she sets about tidying it with deft hands, and an
understanding mind.
“Our Meg here,” says her father, laying a hand on her bonny
brown head, “is the only one of her sex one can trust among one’s
books and papers, with the hope of finding one’s way safely through
them after her. She is the tidy part of my own soul.”
“A part of his own soul!” Nobody shall know until the end has
come on earth how true are these words; what tender, holy secrets
are confided to this dear daughter alone in all the world; how much
apart he lives, even in the midst of that gay and happy family life, in
some respects, from all but her! But here as you note her flitting
among his books, finding out those which she feels he will need for
this work, looking up references and marking passages, you see how
closely she is identified with his intellectual interests. Here in this
little study she is as much at home as he. And in what lies beyond it,
in the little bare room where only a carved crucifix breaks the white
line of wall, and where her father seeks God and his own soul in
solitude, what is her place? Oh! truly a privileged one there, too—as
the world shall know at last when he shall have made the last
distribution of his gifts from the Tower—and to her falls his hair-shirt,
while Cecily has his “handkerchief,” and Elizabeth “a picture in
parchment with her name on the back thereof.”
The bell for Mass sounds from the Parish Church a little bit down
the river, and you follow your host and Margaret to the door to find
Dame Alice (more stately than ever in her blue cloak and white
head-dress), waiting to take Master More’s arm, and head the family
procession, by the path they have made for themselves by the end
of the meadows to the little church. What an appetite you carry
home with you, and how the sweetness of that morning hour in the
quaint old English church lingers with the band that seats itself for
breakfast in the long hall, afterwards, making the meal a veritable
“agape,” a feast of love! What merry jests and quips are bandied
round, and how heartily your host makes you feel yourself of the
company when you prove yourself not inapt to catch and throw back
the light and shining ball of words aimed at you by Henry Pattieson,
the official jester to the household. And so the morning hours pass.
And now it is time for Master More to make himself ready for the
day’s business in the Law Courts. It appears, however, that you need
not terminate your pleasant visit so quickly. It is proposed by the
master of the house himself, seconded most cordially by Dame Alice,
and passed with acclamation by the whole band of youngsters, that
you spend yet another day in this hospitable household, and
strengthen your acquaintance with its inmates. It is not to be
expected, as you perfectly agree when the fact is pointed out to you
by Dame Alice, that that good lady should spend much time with
you, having heavy household duties to attend to, but the girls will be
free presently, and as for the boys, having nothing more important
to do than lessons, they and Master Gunnel are ready to devote
themselves to you immediately.
But first Master More has to be seen off, and kerchiefs and sashes
have to be waved at him from the water-stairs, until his barge grows
smaller and smaller, and finally the speck it has become has been
caught into the distance. Then off go the girls, under the bustling
and energetic directions of Dame Alice, to help in the dairy or
kitchen, or attend to the wants of the poor, whose meals are as
regular a part of the household machinery as those of the family
themselves. In the meantime you go to the schoolroom with Master
Gunnel and the boys—young John Clement, and Jack Dancey, and
Will Roper, wards or protégés of Master More, and the son of the
house, young John More. A word or two puts you in possession of
the present position and future prospects of the lads. Young Clement
has a marked taste for medicine, and is already a distinguished
botanist. He has been taken into the household at the
recommendation of Dean Colet, at whose School of St. Paul’s he has
already distinguished himself, and while pursuing his own Greek and
Latin studies under Master Gunnel, preparatory to entering Oxford,
he acts as assistant tutor and directs the botanical researches of the
others. Will Roper is a ward of Master More, and Jack Dancey is the
son of a legal client, whom the good man has taken into his house
until his affairs can be settled. Otherwise it were ill for young
Dancey, of whose estates the lawyers alone draw the profits. To
balance matters, Dancey, very wisely, proposes to became a lawyer
himself. As for Jack More, you know a little of his abilities already—
but it needs no Master Gunnel to tell you presently when you shall
be alone with him—the boys being given a task to do, and sent into
the garden with it—that in the matter of brains, poor Jack can never
hope to compete with his sisters.
You venture to remark that it is a pity, but you do not find Master
Gunnel over ready to agree with you.
“For my part,” he says, “I hold with Master More that the harvest
will not be much affected whether it is a man or a woman who sows
the field. In truth, it is a matter on which he hath done me the
honour to put his views in writing, at some length—if you care to
see his letter, I have it at hand.”
Indeed, you care very much, and presently you are seated in a
comfortable window seat, with the treasured letter spread out on
your knees.
To your shame, be it spoken, you read the Latin with less ease
than Cecy or Jack would show; noting which, Master Gunnel
unostentatiously begins to read in English some of the more
important passages.
“Listen to this,” he counsels you, pointing to a marked passage,
and thereupon begins:—
“Nor do I think that the harvest will be much affected whether it is
a man or a woman who sows the field. They both have the same
human nature, which reason differentiates from that of beasts; both,
therefore, are equally suited for those studies by which reason is
perfectioned, and becomes fruitful like a ploughed land, on which
the seed of good lessons has been sown. If it be true that the soil of
woman’s brain be bad, and apter to bear bracken than corn, by
which saying many keep women from study, I think, on the contrary,
that a woman’s wit is, on that account, all the more diligently to be
cultivated, that nature’s defect may be redressed by industry. This
was the opinion of the ancients, of those who were most prudent as
well as most holy. Not to speak of the rest, St. Jerome and St.
Augustine not only exhorted excellent matrons and most noble
virgins to study, but also, in order to assist them, diligently explained
the abstruse meanings of Holy Scripture, and wrote for tender girls
letters replete with so much erudition, that now-a-days old men,
who call themselves professors of sacred science, can scarcely read
them correctly, much less understand them. Do you, my learned
Gunnel, have the kindness to see that my daughters thoroughly
learn these works of those holy men....”
“So that is the explanation of the Saint Augustine we found on
Margaret’s desk yesterevening?”
Master Gunnel nods confirmation, but he is much occupied in
finding the next suitable passage in the letter, and does not speak
immediately.
Then with his thumb on the paragraph selected, he looks up for a
moment out of kind, rather near-sighted eyes.
“Do you remember last night when he spoke of the ‘Valiant
Woman,’ and showed how all those who have girls to educate can
find in her an imperishable model? For his own daughters he hath
borne in mind, that, of all the virtues of the ‘Valiant Woman,’ it is her
fear of the Lord that alone giveth substance and value to the others.
‘Many daughters have gathered together riches: thou hast surpassed
them all. Favour is deceitful, and beauty is vain: the woman that
feareth the Lord, she shall be praised.’ Hark, how he drives home the
point. He hath been praising Elizabeth for her good conduct in her
mother’s absence.
“‘Let her understand that such conduct delights me more than all
possible letters I could receive from anyone. Though I prefer
learning joined with virtue, to all the treasures of kings, yet renown
for learning, when it is not united with a good life, is nothing else
than splendid and notorious infamy: this would be specially the case
in a woman. Since erudition in women is a new thing, and a
reproach to the sloth of men, many will gladly assail it, and impute
to literature what is really the fault of nature, thinking from the vices
of the learned to get their own ignorance esteemed as virtue. On the
other hand, if a woman (and this I desire and hope with you as their
teacher for all my daughters) to eminent virtue should add an
outwork of even moderate skill in literature, I think she will have
more real profit than if she had obtained the riches of Crœsus and
the beauty of Helen. I do not say this because of the glory which will
be hers, though glory follows virtue as a shadow follows a body, but
because the reward of wisdom is too solid to be lost like riches or to
decay like beauty, since it depends on the intimate conscience of
what is right, not on the talk of men, than which nothing is more
foolish or mischievous.
“‘It belongs to a good man, no doubt, to avoid infamy, but to lay
himself out for renown is the conduct of a man who is not only
proud, but ridiculous and miserable. A soul must be without peace
which is ever fluctuating between elation and disappointment from
the opinions of men. Among all the benefits that learning bestows
on men, there is none more excellent than this, that by the study of
books we are taught in that very study to seek not praise, but utility.
Such has been the teaching of the most learned men, especially of
philosophers, who are the guides of human life, although some may
have abused learning, like other good things, simply to court glory
and popular renown.’”
Master Gunnel interrupts himself a moment with a reminiscent
smile: “It may well have been that I was in danger of turning
Margaret’s attention to the wrong things, but, if this were so, I was
soon made to discover the mistake. Mark how gently I am brought
to task:
“‘I have dwelt so much on this matter, my dear Gunnel, because
of what you say in your letter, that Margaret’s lofty character should
not be abased. In this judgment I quite agree with you; but to me,
and, no doubt, to you also, that man would seem to abase a
generous character who should accustom it to admire what is vain
and low. He, on the contrary, raises the character who rises to virtue
and true goods, and who looks down with contempt from the
contemplation of what is sublime, on those shadows of good things
which almost all mortals, through ignorance of truth, greedily snatch
at as if they were true goods.’”
But here come the boys back with their finished tasks; and little
Cecy is at the door, with her stepmother’s compliments, and are you
fond of curds and cream? If so, you will come to the dairy and eat
them, with a dish of strawberries, gathered by Dame Alice herself
when the morning dew was yet on them, and carefully kept for you
until this moment on the coolest shelf of the cool dark pantry.
MARIE JEANNE D’AUMALE
A Little Schoolgirl of Saint-Cyr

Part I.
The little “new” girl had sobbed herself to sleep at last, and in all
the long, white dormitory there was no sound but that of the regular
breathing of healthy, sleeping children. Very gently, Madame de
Fontaine withdrew her hand from the lock of the little fingers which
had held it so long. Then, as she stooped to kiss the small face on
the tear-stained pillow, she heard a murmur of “Maman!” and saw
that the child was smiling in her sleep.
“She is dreaming of home,” said Madame de Fontaine to herself;
and, involuntarily, she turned to the unshuttered window, when she
was back in her cell at the end of the dormitory, and yielded her own
dreams to the spells the white moon was weaving for them.
Away across the park, long cords of light were stretched across
the dark mass of the Château, where a King and his courtiers held
revel. Now and then, the night wind whispering to the tall trees,
carried snatches of the music to which the dainty, jewelled feet of
the Court ladies moved rhythmically. But these things barely touched
the nun’s consciousness. Beyond the boundaries of the stately park,
far away from the echoes of courtly music, or the light of a King’s
presence, her dreams were following where those of little Marie
Jeanne d’Aumale had led—to an old “gentilhommière” in the heart of
the provinces, very shabby, and tumble-down, and dilapidated, but
where a little girl could be very happy, because she called it “home.”
It may well have been that more than one of the little sleepers in
the long row of little white beds was dreaming of just such an old
“noblesse”;[14] and that is why, as she looked into the moonlit park,
the nun could see it so plainly before her. Poor little girls! Two titles
had procured for them their right of entrance into Saint-Cyr: nobility
of birth, and poverty; and one was more clearly written across the
tumble-down walls, the grass-grown courtyard, the empty byres and
stables of their old provincial “gentilhommières,” than the other on
the Coat of Arms carved above the dilapidated doorway.
And was not one as honourable as the other? Nun as she was,
Madame de Fontaine was not yet dead to that noble pride, to which,
as Madame de Maintenon herself has finely said, “before having
died, one must have lived.” And, standing there at the window of
that establishment, whose foundation, four years ago, represented
an instalment of payment of the debt contracted by the Monarchy to
France, to the nobility of France, ruined in its service, she felt the
thrill of one whose order “hath chosen the better part.”
And all the time, from the lighted palace across the park, floated
the soft strains of dance-music! There, they who had made the other
choice, who had abandoned their homes, and their home duties,
who lived at Court, absentees from their estates, and deserters from
their “consigne,” were dancing their “branles,” and “courantes,” their
“menuets” and “passe-pied” in the light of the King’s presence. Let
them dance on! The true hope of France was in these little sleeping
girls, who, gathered together under the pious roof of Saint-Cyr, were
being trained for a womanhood, which should work out the
regeneration of a kingdom.
Never has a more splendid tribute been paid to women than in the
foundation of Saint-Cyr; and one runs the risk of failing to realize its
importance, both in the history of feminism, and in the history of
education, if one neglects to consider it, as much in the light of the
statesmanship of Louis XIV. as in that of the charity of Madame de
Maintenon. The primary idea was hers, no doubt—but it remained
for the King, not only to supply her with the means of putting her
project into execution, but to perceive the part it might play in the
economical reconstruction of his kingdom. Long wars had left the
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

Test Bank for Campbell Biology Concepts and Connections, 7th Edition: Reece instant download
100% (2)
Test Bank for Campbell Biology Concepts and Connections, 7th Edition: Reece instant download
49 pages
Water less Body Bath
No ratings yet
Water less Body Bath
10 pages
Bacilli and Agrobiotechnology - Phytostimulation
No ratings yet
Bacilli and Agrobiotechnology - Phytostimulation
334 pages
PDF Third Random
No ratings yet
PDF Third Random
27 pages
(Allameh, 2012) Isolation, Identification and Characterization of Leuconostoc Mesenteroides As A New Probiotic From Intestine of Snakehead Fish (Channa Striatus) .
100% (1)
(Allameh, 2012) Isolation, Identification and Characterization of Leuconostoc Mesenteroides As A New Probiotic From Intestine of Snakehead Fish (Channa Striatus) .
7 pages
Study of The Microbiological Quality and Physio-Chemical Properties of The Soft White Cheese Traditionally Produced in Wasit Province in Iraq
No ratings yet
Study of The Microbiological Quality and Physio-Chemical Properties of The Soft White Cheese Traditionally Produced in Wasit Province in Iraq
14 pages
Alkaline Diet Recipe Book
100% (3)
Alkaline Diet Recipe Book
132 pages
3 - Classification of Microorganisms
No ratings yet
3 - Classification of Microorganisms
27 pages
AQA - Biology - Cell Biology - GraspIT - GCSE - Re-Usable Worksheet
No ratings yet
AQA - Biology - Cell Biology - GraspIT - GCSE - Re-Usable Worksheet
6 pages
Aiats Medical Aipmt2015 Test1
50% (2)
Aiats Medical Aipmt2015 Test1
8 pages
Bacterial Cellulose Biotextiles For The Future of Sustainable Fashion A Review
No ratings yet
Bacterial Cellulose Biotextiles For The Future of Sustainable Fashion A Review
14 pages
U-1, Microbiology, Carewell Pharma
No ratings yet
U-1, Microbiology, Carewell Pharma
39 pages
Environmental Significance of Lichens and Biodeterioration: Gül Şah Çobanoğlu Özyiğitoğlu
No ratings yet
Environmental Significance of Lichens and Biodeterioration: Gül Şah Çobanoğlu Özyiğitoğlu
31 pages
Microbes
No ratings yet
Microbes
30 pages
Aquaculture: M.E. Azim, D.C. Little
No ratings yet
Aquaculture: M.E. Azim, D.C. Little
7 pages
Biohacking
No ratings yet
Biohacking
9 pages
Clostridium Perfingens
No ratings yet
Clostridium Perfingens
7 pages
Comprehention - Engy
No ratings yet
Comprehention - Engy
12 pages
Peleaz F 2006
No ratings yet
Peleaz F 2006
10 pages
SLE234 Lab Report 1 - 2018
No ratings yet
SLE234 Lab Report 1 - 2018
3 pages
DXM MicroScan WalkAway Brochure EN
No ratings yet
DXM MicroScan WalkAway Brochure EN
12 pages
Andrographis Paniculata A Review of Its Traditional Uses Phytochemistry and Pharmacology 2167 0412.1000169
No ratings yet
Andrographis Paniculata A Review of Its Traditional Uses Phytochemistry and Pharmacology 2167 0412.1000169
15 pages
Teks 1
No ratings yet
Teks 1
10 pages
Bio Investigatory Project
100% (1)
Bio Investigatory Project
18 pages
Ok lm1 Pitch Anything Blueprint Paid
89% (9)
Ok lm1 Pitch Anything Blueprint Paid
11 pages
Pandangan & Sardani - Activity-6-Isolation-Of-Microorganisms
No ratings yet
Pandangan & Sardani - Activity-6-Isolation-Of-Microorganisms
2 pages
Thermophilic Bacteria Presentation Ver3 Aug05
No ratings yet
Thermophilic Bacteria Presentation Ver3 Aug05
12 pages
Australian Biology Olympiad 2008
No ratings yet
Australian Biology Olympiad 2008
31 pages
J201 Quiz II
No ratings yet
J201 Quiz II
5 pages
Agriculture Junior Secondary School (Jss 2) First Term
100% (2)
Agriculture Junior Secondary School (Jss 2) First Term
4 pages
The Sympathizer: A Novel (Pulitzer Prize for Fiction)
From Everand
The Sympathizer: A Novel (Pulitzer Prize for Fiction)
Viet Thanh Nguyen
4.5/5 (141)
The Unwinding: An Inner History of the New America
From Everand
The Unwinding: An Inner History of the New America
George Packer
4/5 (45)
The Yellow House: A Memoir (2019 National Book Award Winner)
From Everand
The Yellow House: A Memoir (2019 National Book Award Winner)
Sarah M. Broom
4/5 (100)
A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story
From Everand
A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story
Dave Eggers
3.5/5 (233)
Manhattan Beach: A Novel
From Everand
Manhattan Beach: A Novel
Jennifer Egan
3.5/5 (919)
On Fire: The (Burning) Case for a Green New Deal
From Everand
On Fire: The (Burning) Case for a Green New Deal
Naomi Klein
4/5 (78)
Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America
From Everand
Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America
Gilbert King
4.5/5 (280)
The World Is Flat 3.0: A Brief History of the Twenty-first Century
From Everand
The World Is Flat 3.0: A Brief History of the Twenty-first Century
Thomas L. Friedman
3.5/5 (2289)
The Little Book of Hygge: Danish Secrets to Happy Living
From Everand
The Little Book of Hygge: Danish Secrets to Happy Living
Meik Wiking
3.5/5 (464)
A Man Called Ove: A Novel
From Everand
A Man Called Ove: A Novel
Fredrik Backman
4.5/5 (5181)
Fear: Trump in the White House
From Everand
Fear: Trump in the White House
Bob Woodward
3.5/5 (836)
Sing, Unburied, Sing: A Novel
From Everand
Sing, Unburied, Sing: A Novel
Jesmyn Ward
4/5 (1267)
Her Body and Other Parties: Stories
From Everand
Her Body and Other Parties: Stories
Carmen Maria Machado
4/5 (903)
The Constant Gardener: A Novel
From Everand
The Constant Gardener: A Novel
John le Carré
4/5 (278)
Team of Rivals: The Political Genius of Abraham Lincoln
From Everand
Team of Rivals: The Political Genius of Abraham Lincoln
Doris Kearns Goodwin
4.5/5 (244)
Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race
From Everand
Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race
Margot Lee Shetterly
4/5 (1022)
The Emperor of All Maladies: A Biography of Cancer
From Everand
The Emperor of All Maladies: A Biography of Cancer
Siddhartha Mukherjee
4.5/5 (298)
Wolf Hall: A Novel
From Everand
Wolf Hall: A Novel
Hilary Mantel
4/5 (4088)
Brooklyn: A Novel
From Everand
Brooklyn: A Novel
Colm Tóibín
3.5/5 (2133)
John Adams
From Everand
John Adams
David McCullough
4.5/5 (2546)
Yes Please
From Everand
Yes Please
Amy Poehler
4/5 (2016)
Grit: The Power of Passion and Perseverance
From Everand
Grit: The Power of Passion and Perseverance
Angela Duckworth
4/5 (650)
Rise of ISIS: A Threat We Can't Ignore
From Everand
Rise of ISIS: A Threat We Can't Ignore
Jay Sekulow
3.5/5 (144)
Angela's Ashes: A Memoir
From Everand
Angela's Ashes: A Memoir
Frank McCourt
4.5/5 (943)
Shoe Dog: A Memoir by the Creator of Nike
From Everand
Shoe Dog: A Memoir by the Creator of Nike
Phil Knight
4.5/5 (628)
Principles: Life and Work
From Everand
Principles: Life and Work
Ray Dalio
4/5 (643)
Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future
From Everand
Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future
Ashlee Vance
4.5/5 (582)
The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life
From Everand
The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life
Mark Manson
4/5 (6453)
Bad Feminist: Essays
From Everand
Bad Feminist: Essays
Roxane Gay
4/5 (1090)
The Light Between Oceans: A Novel
From Everand
The Light Between Oceans: A Novel
M.L. Stedman
4.5/5 (815)
The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers
From Everand
The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers
Ben Horowitz
4.5/5 (361)
The Art of Racing in the Rain: A Novel
From Everand
The Art of Racing in the Rain: A Novel
Garth Stein
4/5 (4372)
Steve Jobs
From Everand
Steve Jobs
Walter Isaacson
4.5/5 (1139)
A Tree Grows in Brooklyn
From Everand
A Tree Grows in Brooklyn
Betty Smith
4.5/5 (2033)
The Glass Castle: A Memoir
From Everand
The Glass Castle: A Memoir
Jeannette Walls
4.5/5 (1856)
The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are
From Everand
The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are
Brene Brown
4/5 (1175)
Never Split the Difference: Negotiating As If Your Life Depended On It
From Everand
Never Split the Difference: Negotiating As If Your Life Depended On It
Chris Voss
4.5/5 (1005)
The Woman in Cabin 10
From Everand
The Woman in Cabin 10
Ruth Ware
3.5/5 (2791)
The Perks of Being a Wallflower
From Everand
The Perks of Being a Wallflower
Stephen Chbosky
4.5/5 (4102)
Little Women
From Everand
Little Women
Louisa May Alcott
4.5/5 (2369)
The Outsider: A Novel
From Everand
The Outsider: A Novel
Stephen King
4/5 (2884)

Bacterial Molecular Networks Methods and Protocols 1st Edition Jacques Van Helden instant download

Uploaded by

Bacterial Molecular Networks Methods and Protocols 1st Edition Jacques Van Helden instant download

Uploaded by

Bacterial Molecular Networks Methods and

Protocols 1st Edition Jacques Van Helden pdf

Get Instant Ebook Downloads – Browse at https://ebookgate.com

Membrane Protein Structure Determination Methods and

Hemoglobin Disorders Molecular Methods and Protocols

Molecular Motors Methods and Protocols Methods in

Bacterial Secretion Systems Methods and Protocols 2nd ed.

Molecular Diagnosis of Cancer Methods and Protocols 2nd

Autoimmunity Methods and Protocols Methods in Molecular

Yeast Genetic Networks Methods and Protocols 1st Edition

Cystic Fibrosis Methods and Protocols Methods in Molecular

For further volumes:

Methods and Protocols

Jacques van Helden and Ariane Toussaint

ISSN 1064-3745 e-ISSN 1940-6029

Library of Congress Control Number: 2011938202

ª Springer Science+Business Media, LLC 2012

Printed on acid-free paper

Humana Press is part of Springer Science+Business Media (www.springer.com)

Network-based representations have become pervasive in most fields in biology. Focusing

Bruxelles, Belgium Jacques van Helden

1 Bacterial Molecular Networks: Bridging the Gap Between Functional

PART I EXPERIMENTAL AND IN SILICO APPROACHES TO UNRAVEL

2 Bacterial Interactomes: From Interactions to Networks. . . . . . . . . . . . . . . . . . . . . . . . 15

PART II TOPOLOGICAL ANALYSIS OF BACTERIAL NETWORKS

14 MAVisto: A Tool for Biological Network Motif Analysis . . . . . . . . . . . . . . . . . . . . . . . 263

PART III DYNAMICAL MODELLING

20 Detecting Structural Invariants in Biological Reaction Networks . . . . . . . . . . . . . . . . 377

CEI ABREU-GOODGER  European Molecular Biology Laboratory, European

JOHANNES GEISELMANN  Laboratoire Adaptation et Pathogénie des Microorganismes

AURÉLIEN NALDI  TAGC, INSERM U928, Marseille, France

Bacterial Molecular Networks: Bridging the Gap Between

Nowadays, with over a thousand bacterial genomes sequenced,

Software tool Website Interface Chapter(s)

SmolDyn http://www.smoldyn.org Unix shell prompt Andrews

and DISTILLER, two algorithms dedicated to the mining of

models onto the degree distribution of a given network and to

the book presents a series of qualitative and quantitative approaches

temporal behaviour of individual components (e.g., molecules

networks. Irrespective of the mathematical formalism, it remains

The collaboration between the TAGC (DT) and BiGRe (JvH

Science Policy Office, project P6/25 (BioMaGNet). The BiGRe

Young L, Qi ZQ, Segall-Shapiro TH, Calvey 22. Isalan M, Lemerle C, Michalodimitrakis K,

Experimental and In Silico Approaches to

Bacterial Interactomes: From Interactions to Networks

1.1. From Up to recently, classical biochemical and genetic methods have

1.2. ...To Once identified by these large-scale methodologies, interactions

Molecular Function Biological Process

Not Annotated Experimental Evidences Electronic Annotations

understood in a near future a bacterium such as E. coli or bacteria

CBP ENLYFQG 3xFlag C-terminal SPA tag (8 kDa)

TEV cleavage site

SBP ENLYFQG protein G C-terminal TAP(SG) tag (20 kDa)

CBP LEVLFQGP GST C-terminal gTAP tag (30 kDa)

CBP ENLYFQG protein A

CBP binding on Camodulin beads

Elution with EGTA

recognized and cleaved by the Tobacco Etch Virus (TEV) prote-

highlight those that have been or can be used in bacteria for

activity. Depending on the chosen two-hybrid technique, the

interaction (37, 38). This system appears more physiological for

interactomes published so far have used PCA methods, all these

2.3.3. Transfer A third type of method to detect binary protein–protein interac-

3.1.1. Affinity Using the technique developed by Court’s laboratory (11)

Total Number of Number of Number of

thanks to their clustering in protein complexes or functional

Main Interaction Number of bacterial

under the control of a unique promoter (456 strains compared to

3.1.2. Two-Hybrid Screens Five high-throughput bacterial protein–protein interactions

3.2. Databases Interactions identified with high-throughput methods constitute

additional dataset of ~800 interactions identified by classical

3.3.4. Contaminants Both TAP-like and two-hybrid-like methods produce false

The fact that different methodologies have been used to generate

a cell, the gain of information about dynamics of protein network

Work by E.B. is funded by the CNRS, a young research grant from

From Bacterial to Microbial Ecosystems (Metagenomics)

CEI ABREU-GOODGER European Molecular Biology Laboratory, European

JOHANNES GEISELMANN Laboratoire Adaptation et Pathogénie des Microorganismes

AURÉLIEN NALDI TAGC, INSERM U928, Marseille, France