![UNAFold 5
(or DG37
) and enthalpy, ∆H, are easily computed. From this, the
entropy, ∆S, is easily computed as:
∆ ×
∆ − ∆
S =
H G
T
1000 ,
where T is the hybridization temperature (K) and the factor of 1000
expresses ∆S in e.u. (entropy units; 1 e.u. = 1cal/mol/K). The
terms “free energy,” “enthalpy,” and “entropy” are really changes in
free energy, enthalpy, and entropy with respect to the “random coil
state.” In reality, “random coil” is a large ensemble of states. The free
energy and enthalpy of this ensemble of states may be (and is) set to
zero. Only entropy cannot be arbitrarily assigned. Thus a ∆S term
is Rln(N/Nref
), where R is the universal gas constant, 1.9872 e.u.;
N is the number of states consistent with hybridization; and Nref
is the
number of states, always larger, in the unconstrained “random coil.”
The simplest model for hybridization assumes that there are
two states: hybridized, where the structure is unique and known,
and “random coil.” It is then a simple matter to deduce a melt-
ing temperature, Tm
, at which half of the dimers have dissociated.
The formula is given by:
T
f
m =
H
S + R
1000
ln(Ct
×
∆
∆ )
, [1]
where ∆H, ∆S (expressed in e.u.) and R have already been defined, Ct
is the total strand concentration, and f = 4 when the two strands are
different and f = 1 when self-hybridization takes place. This formula
implicitly assumes that both strands are present in equal concentra-
tions. Competition with folding and with other possible hybridiza-
tions, such as homodimer formation, is not considered at all.
UNAFold offers a number of increasingly sophisticated ways
to compute melting temperatures and also computes entire melt-
ing profiles: UV absorbance at 260nm, heat capacity (Cp
), and
mole fractions of various single and double-stranded molecu-
lar species as a function of temperature. Even when Equation
[1] is used, the hybridization is computed by minimizing free
energy. The various levels of complexity and choice of methods
are described in the following.
It is important to issue a warning to those who wish to apply
these methods to microarrays. Hybridization on microarrays is
complicated by the fact that one of each hybridizing pair is immo-
bilized. It is difficult to compute the effective “solution concentra-
tion” for such molecules, and diffusion is probably an important
factor in slowing the time necessary to reach equilibrium. How-
ever, even for hybridization in solution, kinetic simulations that
treat different probe-target pairs independently can lead to totally
incorrect results for the time required to reach equilibrium as well
as the equilibrium concentrations themselves (21).](https://image.slidesharecdn.com/56076-250414223214-1a415799/85/Bioinformatics-Structure-Function-and-Applications-1st-Edition-Nicholas-R-Markham-22-320.jpg)
Bioinformatics Structure Function and Applications 1st Edition Nicholas R. Markham
- 1. Bioinformatics Structure Function and Applications 1st Edition Nicholas R. Markham pdf download https://ebookfinal.com/download/bioinformatics-structure- function-and-applications-1st-edition-nicholas-r-markham-2/ Explore and download more ebooks or textbooks at ebookfinal.com
- 2. Here are some recommended products for you. Click the link to download, or explore more at ebookfinal Bioinformatics Structure Function and Applications 1st Edition Nicholas R. Markham https://ebookfinal.com/download/bioinformatics-structure-function-and- applications-1st-edition-nicholas-r-markham/ The insects structure and function Fifth Edition R F Chapman https://ebookfinal.com/download/the-insects-structure-and-function- fifth-edition-r-f-chapman/ Starch in food structure function and applications 1st Edition Robert A. Disilvestro https://ebookfinal.com/download/starch-in-food-structure-function-and- applications-1st-edition-robert-a-disilvestro/ Keratin Structure Properties and Applications Structure Properties and Applications 1st Edition Renke Dullaart https://ebookfinal.com/download/keratin-structure-properties-and- applications-structure-properties-and-applications-1st-edition-renke- dullaart/
- 3. Protein Bioinformatics From Sequence to Function 1st Edition M. Michael Gromiha https://ebookfinal.com/download/protein-bioinformatics-from-sequence- to-function-1st-edition-m-michael-gromiha/ Structure and Function From clause to discourse and beyond Christopher Butler https://ebookfinal.com/download/structure-and-function-from-clause-to- discourse-and-beyond-christopher-butler/ Joint Structure And Function A Comprehensive Analysis 4th Edition Pamela Levangie https://ebookfinal.com/download/joint-structure-and-function-a- comprehensive-analysis-4th-edition-pamela-levangie/ Bioinformatics algorithms techniques and applications 1st Edition Ion Ma■Ndoiu https://ebookfinal.com/download/bioinformatics-algorithms-techniques- and-applications-1st-edition-ion-ma%cc%86ndoiu/ Neurotransmitter Transporters Structure Function and Regulation 2nd Edition Maarten E. A. Reith https://ebookfinal.com/download/neurotransmitter-transporters- structure-function-and-regulation-2nd-edition-maarten-e-a-reith/
- 5. Bioinformatics Structure Function and Applications 1st Edition Nicholas R. Markham Digital Instant Download Author(s): Nicholas R. Markham, Michael Zuker (auth.), Jonathan M. Keith PhD (eds.) ISBN(s): 9781603274289, 1603274286 Edition: 1 File Details: PDF, 9.37 MB Year: 2008 Language: english
- 8. 460. Essential Concepts in Toxicogenomics, edited by Donna L. Mendrick and William B. Mattes, 2008 459. Prion Protein Protocols, edited by Andrew F. Hill, 2008 458. Artificial Neural Networks: Methods and Applica- tions, edited by David S. Livingstone, 2008 457. Membrane Trafficking, edited by Ales Vancura, 2008 456. Adipose Tissue Protocols, Second Edition, edited by Kaiping Yang, 2008 455. Osteoporosis, edited by Jennifer J. Westendorf, 2008 454. SARS- and Other Coronaviruses: Laboratory Proto- cols, edited by Dave Cavanagh, 2008 453. Bioinformatics, Volume II: Structure, Function and Applications, edited by Jonathan M. Keith, 2008 452. Bioinformatics, Volume I: Data, Sequence Analysis and Evolution, edited by Jonathan M. Keith, 2008 451. Plant Virology Protocols: From Viral Sequence to Pro- tein Function, edited by Gary Foster, Elisabeth Johansen, Yiguo Hong, and Peter Nagy, 2008 450. Germline Stem Cells, edited by Steven X. Hou and Shree Ram Singh, 2008 449. Mesenchymal Stem Cells: Methods and Protocols, edited by Darwin J. Prockop, Douglas G. Phinney, and Bruce A. Brunnell, 2008 448. Pharmacogenomics in Drug Discovery and Develop- ment, edited by Qing Yan, 2008 447. Alcohol: Methods and Protocols, edited by Laura E. Nagy, 2008 446. Post-translational Modification of Proteins: Tools for Functional Proteomics, Second Edition, edited by Christoph Kannicht, 2008 445. Autophagosome and Phagosome, edited by Vojo Deretic, 2008 444. Prenatal Diagnosis, edited by Sinhue Hahn and Laird G. Jackson, 2008 443. Molecular Modeling of Proteins, edited by Andreas Kukol, 2008. 442. RNAi: Design and Application, edited by Sailen Barik, 2008 441. Tissue Proteomics: Pathways, Biomarkers, and Drug Discovery, edited by Brian Liu, 2008 440. Exocytosis and Endocytosis, edited by AndreiI.Ivanov, 2008 439. Genomics Protocols, Second Edition, edited by Mike Starkey and Ramnanth Elaswarapu, 2008 438. Neural Stem Cells: Methods and Protocols, Second Edition, edited by Leslie P. Weiner, 2008 437. Drug Delivery Systems, edited by Kewal K. Jain, 2008 436. Avian Influenza Virus, edited by Erica Spackman, 2008 435. Chromosomal Mutagenesis, edited by Greg Davis and Kevin J. Kayser, 2008 434. Gene Therapy Protocols: Volume 2: Design and Char- acterization of Gene Transfer Vectors, edited by Joseph M. LeDoux, 2008 433. Gene Therapy Protocols: Volume 1: Production and In Vivo Applications of Gene Transfer Vectors, edited by Joseph M. LeDoux, 2008 432. Organelle Proteomics, edited by Delphine Pflieger and Jean Rossier, 2008 431. Bacterial Pathogenesis: Methods and Protocols, edited by Frank DeLeo and Michael Otto, 2008 430. Hematopoietic Stem Cell Protocols, edited by Kevin D. Bunting, 2008 429. Molecular Beacons: Signalling Nucleic Acid Probes, Methods and Protocols, edited by Andreas Marx and Oliver Seitz, 2008 428. Clinical Proteomics: Methods and Protocols, edited by Antonia Vlahou, 2008 427. Plant Embryogenesis, edited by Maria Fernanda Suarez and Peter Bozhkov, 2008 426. Structural Proteomics: High-Throughput Methods, edited by Bostjan Kobe, Mitchell Guss, and Huber Thomas, 2008 425. 2D PAGE: Sample Preparation and Fractionation, Volume 2, edited by Anton Posch, 2008 424. 2D PAGE: Sample Preparation and Fractionation, Volume 1, edited by Anton Posch, 2008 423. Electroporation Protocols: Preclinical and Clinical Gene Medicine, edited by Shulin Li, 2008 422. Phylogenomics, edited by William J. Murphy, 2008 421. Affinity Chromatography: Methods and Protocols, Sec- ond Edition, edited by Michael Zachariou, 2008 420. Drosophila: Methods and Protocols, edited by Christian Dahmann, 2008 419. Post-Transcriptional Gene Regulation, edited by Jeffrey Wilusz, 2008 418. Avidin–Biotin Interactions: Methods and Applications, edited by Robert J. McMahon, 2008 417. Tissue Engineering, Second Edition, edited by Hannsjörg Hauser and Martin Fussenegger, 2007 416. Gene Essentiality: Protocols and Bioinformatics, edited by Svetlana Gerdes and Andrei L. Osterman, 2008 415. Innate Immunity, edited by Jonathan Ewbank and Eric Vivier, 2007 414. Apoptosis in Cancer: Methods and Protocols, edited by Gil Mor and Ayesha Alvero, 2008 413. Protein Structure Prediction, Second Edition, edited by Mohammed Zaki and Chris Bystroff, 2008 412. Neutrophil Methods and Protocols, edited by Mark T. Quinn, Frank R. DeLeo, and Gary M. Bokoch, 2007 411. Reporter Genes: A Practical Guide, edited by Don Anson, 2007 410. Environmental Genomics, edited by Cristofre C. Martin, 2007 409. Immunoinformatics: Predicting Immunogenicity In Silico, edited by Darren R. Flower, 2007 408. Gene Function Analysis, edited by Michael Ochs, 2007 407. Stem Cell Assays, edited by Vemuri C. Mohan, 2007 406. Plant Bioinformatics: Methods and Protocols, edited by David Edwards, 2007 405. Telomerase Inhibition: Strategies and Protocols, edited by Lucy Andrews and Trygve O. Tollefsbol, 2007 METHODS IN MOLECULAR BIOLOGY™ John M. Walker, SERIES EDITOR
- 9. ME T H O D S I N MO L E C U L A R BI O L O G Y ™ Bioinformatics Volume II Structure, Function and Applications Edited by Jonathan M. Keith, PhD School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland,Australia
- 10. ISBN: 978-1-60327-428-9 e-ISBN: 978-1-60327-429-6 ISSN 1064-3745 e-ISSN: 1940-6029 DOI: 10.1007/978-1-60327-429-6 Library of Congress Control Number: 20082922946 © 2008 Humana Press, a part of Springer Science+Business Media, LLC 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, 999 Riverview Drive, Suite 208, Totowa, NJ 07512 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. While the advice and information in this book are believed to be true and accurate at the date of going to press, nei- ther the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Cover illustration: Fig. 1A, Chapter 23, “Visualization,” by Falk Schreiber (main image); and Fig. 4, Chapter 5, “The Classification of Protein Domains,” by Russell L. Marsden and Christine A. Orengo (surrounding images) Printed on acid-free paper 9 8 7 6 5 4 3 2 1 springer.com Editor Jonathan M. Keith School of Mathematical Sciences Queensland University of Technology Brisbane, Queensland, Australia j.keith@qut.edu.au Series Editor John Walker Hatfield, Hertfordshire AL10 9NP UK
- 11. Preface Bioinformatics is the management and analysis of data for the life sciences. As such, it is inherently interdisciplinary, drawing on techniques from Computer Science, Statis- tics, and Mathematics and bringing them to bear on problems in Biology. Moreover, its subject matter is as broad as Biology itself. Users and developers of bioinformatics methods come from all of these fields. Molecular biologists are some of the major users of Bioinformatics, but its techniques are applicable across a range of life sciences. Other users include geneticists, microbiologists, biochemists, plant and agricultural scientists, medical researchers, and evolution researchers. The ongoing exponential expansion of data for the life sciences is both the major challenge and the raison d’être for twenty-first century Bioinformatics. To give one example among many, the completion and success of the human genome sequencing project, far from being the end of the sequencing era, motivated a proliferation of new sequencing projects. And it is not only the quantity of data that is expanding; new types of biological data continue to be introduced as a result of technological development and a growing understanding of biological systems. Bioinformatics describes a selection of methods from across this vast and expanding discipline. The methods are some of the most useful and widely applicable in the field. Most users and developers of bioinformatics methods will find something of value to their own specialties here, and will benefit from the knowledge and experience of its 86 con- tributing authors. Developers will find them useful as components of larger methods, and as sources of inspiration for new methods. Volume II, Section IV in particular is aimed at developers; it describes some of the “meta-methods”—widely applicable mathematical and computational methods that inform and lie behind other more special- ized methods—that have been successfully used by bioinformaticians. For users of bioin- formatics, this book provides methods that can be applied as is, or with minor variations to many specific problems. The Notes section in each chapter provides valuable insights into important variations and when to use them. It also discusses problems that can arise and how to fix them. This work is also intended to serve as an entry point for those who are just beginning to discover and use methods in bioinformatics. As such, this book is also intended for students and early career researchers. As with other volumes in the Methods in Molecular Biology™ series, the intention of this book is to provide the kind of detailed description and implementation advice that is crucial for getting optimal results out of any given method, yet which often is not incorporated into journal publications. Thus, this series provides a forum for the com- munication of accumulated practical experience. The work is divided into two volumes, with data, sequence analysis, and evolution the subjects of the first volume, and structure, function, and application the subjects of the second. The second volume also presents a number of “meta-methods”: techniques that will be of particular interest to developers of bioinformatic methods and tools. Within Volume I, Section I deals with data and databases. It contains chapters on a selection of methods involving the generation and organization of data, including v
- 12. sequence data, RNA and protein structures, microarray expression data, and func- tional annotations. Section II presents a selection of methods in sequence analysis, beginning with multiple sequence alignment. Most of the chapters in this section deal with methods for discovering the functional components of genomes, whether genes, alternative splice sites, non-coding RNAs, or regulatory motifs. Section III presents several of the most useful and interesting methods in phylogenetics and evolution. The wide variety of topics treated in this section is indicative of the breadth of evolution research. It includes chapters on some of the most basic issues in phylogenet- ics: modelling of evolution and inferring trees. It also includes chapters on drawing infer- ences about various kinds of ancestral states, systems, and events, including gene order, recombination events and genome rearrangements, ancestral interaction networks, lateral gene transfers, and patterns of migration. It concludes with a chapter discussing some of the achievements and challenges of algorithm development in phylogenetics. In Volume II, Section I, some methods pertinent to the prediction of protein and RNA structures are presented. Methods for the analysis and classification of structures are also discussed. Methods for inferring the function of previously identified genomic elements (chiefly protein-coding genes) are presented in Volume II, Section II. This is another very diverse subject area, and the variety of methods presented reflects this. Some well-known techniques for identifying function, based on homology, “Rosetta stone” genes, gene neighbors, phylogenetic profiling, and phylogenetic shadowing are discussed, alongside methods for identifying regulatory sequences, patterns of expres- sion, and participation in complexes. The section concludes with a discussion of a technique for integrating multiple data types to increase the confidence with which functional predictions can be made. This section, taken as a whole, highlights the opportunities for development in the area of functional inference. Some medical applications, chiefly diagnostics and drug discovery, are described in Volume II, Section III. The importance of microarray expression data as a diagnostic tool is a theme of this section, as is the danger of over-interpreting such data. The case study presented in the final chapter highlights the need for computational diagnostics to be biologically informed. The final section presents just a few of the “meta-methods” that developers of bioinformatics methods have found useful. For the purpose of designing algorithms, it is as important for bioinformaticians to be aware of the concept of fixed parameter tractability as it is for them to understand NP-completeness, since these concepts often determine the types of algorithms appropriate to a particular problem. Clustering is a ubiquitous problem in Bioinformatics, as is the need to visualize data. The need to interact with massive data bases and multiple software entities makes the development of computational pipelines an important issue for many bioinformaticians. Finally, the chapter on text mining discusses techniques for addressing the special problems of interacting with and extracting information from the vast biological literature. Jonathan M. Keith vi Preface
- 13. Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Content of Volume I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii SECTION I: STRUCTURES 1. UNAFold: Software for Nucleic Acid Folding and Hybridization . . . . . . . . . . . . . . 3 Nicholas R. Markham and Michael Zuker 2. Protein Structure Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Bissan Al-Lazikani, Emma E. Hill, and Veronica Morea 3. An Introduction to Protein Contact Prediction . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 Nicholas Hamilton and Thomas Huber 4. Analysis of Mass Spectrometry Data in Proteomics. . . . . . . . . . . . . . . . . . . . . . . . . 105 Rune Matthiesen and Ole N. Jensen 5. The Classification of Protein Domains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 Russell L. Marsden and Christine A. Orengo SECTION II: INFERRING FUNCTION 6. Inferring Function from Homology. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149 Richard D. Emes 7. The Rosetta Stone Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169 Shailesh V. Date 8. Inferring Functional Relationships from Conservation of Gene Order . . . . . . . . . . 181 Gabriel Moreno-Hagelsieb 9. Phylogenetic Profiling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 Shailesh V. Date and José M. Peregrín-Alvarez 10. Phylogenetic Shadowing: Sequence Comparisons of Multiple Primate Species . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 Dario Boffelli 11. Prediction of Regulatory Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233 Albin Sandelin 12. Expression and Microarrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245 Joaquín Dopazo and Fátima Al-Shahrour 13. Identifying Components of Complexes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 Nicolas Goffard and Georg Weiller 14. Integrating Functional Genomics Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267 Insuk Lee and Edward M. Marcotte SECTION III: APPLICATIONS AND DISEASE 15. Computational Diagnostics with Gene Expression Profiles. . . . . . . . . . . . . . . . . . . 281 Claudio Lottaz, Dennis Kostka, Florian Markowetz, and Rainer Spang vii
- 14. 16. Analysis of Quantitative Trait Loci. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297 Mario Falchi 17. Molecular Similarity Concepts and Search Calculations . . . . . . . . . . . . . . . . . . . . . 327 Jens Auer and Jürgen Bajorath 18. Optimization of the MAD Algorithm for Virtual Screening . . . . . . . . . . . . . . . . . . 349 Hanna Eckert and Jürgen Bajorath 19. Combinatorial Optimization Models for Finding Genetic Signatures from Gene Expression Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363 Regina Berretta, Wagner Costa, and Pablo Moscato 20. Genetic Signatures for a Rodent Model of Parkinson’s Disease Using Combinatorial Optimization Methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379 Mou’ath Hourani, Regina Berretta, Alexandre Mendes, and Pablo Moscato SECTION IV: ANALYTICAL AND COMPUTATIONAL METHODS 21. Developing Fixed-Parameter Algorithms to Solve Combinatorially Explosive Biological Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395 Falk Hüffner, Rolf Niedermeier, and Sebastian Wernicke 22. Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423 Geoffrey J. McLachlan, Richard W. Bean, and Shu-Kay Ng 23. Visualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441 Falk Schreiber 24. Constructing Computational Pipelines. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451 Mark Halling-Brown and Adrian J. Shepherd 25. Text Mining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471 Andrew B. Clegg and Adrian J. Shepherd Index. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 493 viii Contents
- 15. Contributors BISSAN AL-LAZIKANI • Biofocus DPI, London, United Kingdom FÁTIMA AL-SHAHROUR • Department of Bioinformatics, Centro de Investigación Príncipe Felipe (CIPF), Valencia, Spain JENS AUER • Department of Life Science Informatics, Bonn-Aachen International Center for Information Technology (B-IT), Rheinische Friedrich-Wilhelms-University Bonn, Bonn, Germany JÜRGEN BAJORATH • Professor and Chair of Life Science Informatics, Department of Life Science Informatics, Bonn-Aachen International Center for Information Technology (B-IT), Rheinische Friedrich-Wilhelms-University Bonn, Bonn, Germany RICHARD W. BEAN • ARC Centre of Excellence in Bioinformatics, and Institute for Molecular Bioscience, The University of Queensland, Brisbane, Queensland, Australia REGINA BERRETTA • Centre of Bioinformatics, Biomarker Discovery and Information- Based Medicine, The University of Newcastle, Callaghan, New South Wales, Australia DARIO BOFFELLI • Children’s Hospital Oakland Research Institute, Oakland, CA ANDREW B. CLEGG • Institute of Structural Molecular Biology, School of Crystallography, Birkbeck College, University of London, London, United Kingdom WAGNER COSTA • School of Electrical Engineering and Computer Science, The University of Newcastle, Callaghan, New South Wales, Australia SHAILESH V. DATE • PENN Center for Bioinformatics, Department of Genetics, University of Pennsylvania School of Medicine, Philadelphia, PA JOAQUÍN DOPAZO • Department of Bioinformatics, Centro de Investigación Príncipe Felipe (CIPF), Valencia, Spain HANNA ECKERT • Department of Life Science Informatics, Bonn-Aachen International Center for Information Technology (B-IT), Rheinische Friedrich-Wilhelms-University Bonn, Bonn, Germany RICHARD D. EMES • Department of Biology, University College London, London, United Kingdom MARIO FALCHI • Twin Research and Genetic Epidemiology Unit, King’s College London School of Medicine, London, United Kingdom NICOLAS GOFFARD • Research School of Biological Sciences and ARC Centre of Excellence for Integrative Legume Research, The Australian National University, Canberra, Australian Capital Territory, Australia MARK HALLING-BROWN • Institute of Structural Molecular Biology, School of Crystallography, Birkbeck College, University of London, London, United Kingdom NICHOLAS HAMILTON • ARC Centre of Excellence in Bioinformatics, Institute for Molecular Bioscience and Advanced Computational Modelling Centre, The University of Queensland, Brisbane, Queensland, Australia EMMA E. HILL • The Journal of Cell Biology, Rockefeller University Press, New York, NY ix
- 16. MOU’ATH HOURANI • Newcastle Bioinformatics Initiative, School of Electrical Engineer- ing and Computer Science, The University of Newcastle, Callaghan, New South Wales, Australia THOMAS HUBER • School of Molecular and Microbial Sciences and Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, Brisbane, Queensland, Australia FALK HÜFFNER • Institut für Informatik, Friedrich-Schiller-Universität Jena, Jena, Germany OLE N. JENSEN • Department of Biochemistry and Molecular Biology, University of Southern Denmark, Odense, Denmark JONATHAN M. KEITH • School of Mathematical Sciences, Queensland University of Technology, Brisbane, Queensland, Australia DENNIS KOSTKA • Max Planck Institute for Molecular Genetics and Berlin Center for Genome-Based Bioinformatics, Berlin, Germany INSUK LEE • Center for Systems and Synthetic Biology, Institute for Molecular Biology, University of Texas at Austin, Austin, TX CLAUDIO LOTTAZ • Max Planck Institute for Molecular Genetics and Berlin Center for Genome-Based Bioinformatics, Berlin, Germany EDWARD M. MARCOTTE • Center for Systems and Synthetic Biology, and Department of Chemistry and Biochemistry, Institute for Molecular Biology, University of Texas at Austin, Austin, TX NICHOLAS R. MARKHAM • Xerox Litigation Services, Albany, NY FLORIAN MARKOWETZ • Max Planck Institute for Molecular Genetics and Berlin Center for Genome-Based Bioinformatics, Berlin, Germany RUSSELL L. MARSDEN • Biochemistry and Molecular Biology Department, University College London, London, United Kingdom RUNE MATTHIESEN • CIC bioGUNE, Bilbao, Spain GEOFFREY J. MCLACHLAN • ARC Centre of Excellence in Bioinformatics, Institute for Molecular Bioscience, and Department of Mathematics, The University of Queensland, Brisbane, Queensland, Australia ALEXANDRE MENDES • Centre of Bioinformatics, Biomarker Discovery and Information- Based Medicine, The University of Newcastle, Callaghan, New South Wales, Australia VERONICA MOREA • National Research Council (CNR), Institute of Molecular Biology and Pathology (IBPN), Rome, Italy GABRIEL MORENO-HAGELSIEB • Department of Biology, Wilfrid Laurier University, Waterloo, Ontario, Canada PABLO MOSCATO • ARC Centre of Excellence in Bioinformatics, and Centre of Bioin- formatics, Biomarker Discovery and Information-Based Medicine, The University of Newcastle, Callaghan, New South Wales, Australia SHU-KAY NG • Department of Mathematics, The University of Queensland, Brisbane, Queensland, Australia ROLF NIEDERMEIER • Institut für Informatik, Friedrich-Schiller-Universität Jena, Jena, Germany CHRISTINE A. ORENGO • Biochemistry and Molecular Biology Department, University College London, London, United Kingdom JOSÉ M. PEREGRÍN-ALVAREZ • Hospital for Sick Children, Toronto, Ontario, Canada x Contributors
- 17. ALBIN SANDELIN • The Bioinformatics Centre, Department of Molecular Biology and Biotech Research and Innovation Centre, University of Copenhagen, Copenhagen, Denmark FALK SCHREIBER • Leibniz Institute of Plant Genetics and Crop Plant Research (IPK) Gatersleben, Germany and Institute for Computer Science, Martin-Luther University Halle-Wittenberg, Germany ADRIAN J. SHEPHERD • Institute of Structural Molecular Biology, School of Crystallography, Birkbeck College, University of London, London, United Kingdom RAINER SPANG • Max Planck Institute for Molecular Genetics and Berlin Center for Genome-Based Bioinformatics, Berlin, Germany GEORG F. WEILLER • Research School of Biological Sciences and ARC Centre of Excellence for Integrative Legume Research, The Australian National University, Canberra, Australian Capital Territory, Australia SEBASTIAN WERNICKE • Institut für Informatik, Friedrich-Schiller-Universität Jena, Jena, Germany MICHAEL ZUKER • Mathematical Sciences and Biology Department, Rensselaer Polytechnic Institute, Troy, NY Contributors xi
- 18. Contents of Volume I SECTION I: DATA AND DATABASES 1. Managing Sequence Data Ilene Karsch Mizrachi 2. RNA Structure Determination by NMR Lincoln G. Scott and Mirko Hennig 3. Protein Structure Determination by X-Ray Crystallography Andrea Ilari and Carmelinda Savino 4. Pre-Processing of Microarray Data and Analysis of Differential Expression Steffen Durinck 5. Developing an Ontology Midori A. Harris 6. Genome Annotation Hideya Kawaji and Yoshihide Hayashizaki SECTION II: SEQUENCE ANALYSIS 7. Multiple Sequence Alignment Walter Pirovano and Jaap Heringa 8. Finding Genes in Genome Sequence Alice Carolyn McHardy 9. Bioinformatics Detection of Alternative Splicing Namshin Kim and Christopher Lee 10. Reconstruction of Full-Length Isoforms from Splice Graphs Yi Xing and Christopher Lee 11. Sequence Segmentation Jonathan M. Keith 12. Discovering Sequence Motifs Timothy L. Bailey SECTION III: PHYLOGENETICS AND EVOLUTION 13. Modeling Sequence Evolution Pietro Liò and Martin Bishop 14. Inferring Trees Simon Whelan 15. Detecting the Presence and Location of Selection in Proteins Tim Massingham 16. Phylogenetic Model Evaluation Lars Sommer Jermiin, Vivek Jayaswal, Faisal Ababneh, and John Robinson xiii
- 19. 17. Inferring Ancestral Gene Order Julian M. Catchen, John S. Conery, and John H. Postlethwait 18. Genome Rearrangement by the Double Cut and Join Operation Richard Friedberg, Aaron E. Darling, and Sophia Yancopoulos 19. Inferring Ancestral Protein Interaction Networks José M. Peregrín-Alvarez 20. Computational Tools for the Analysis of Rearrangements in Mammalian Genomes Guillaume Bourque and Glenn Tesler 21. Detecting Lateral Genetic Transfer: A Phylogenetic Approach Robert G. Beiko and Mark A. Ragan 22. Detecting Genetic Recombination Georg F. Weiller 23. Inferring Patterns of Migration Paul M.E. Bunje and Thierry Wirth 24. Fixed-Parameter Algorithms in Phylogenetics Jens Gramm, Arfst Nickelsen, and Till Tantau Index xiv Contents of Volume I
- 20. Chapter 1 UNAFold Software for Nucleic Acid Folding and Hybridization Nicholas R. Markham and Michael Zuker Abstract The UNAFold software package is an integrated collection of programs that simulate folding, hybridiza- tion, and melting pathways for one or two single-stranded nucleic acid sequences. The name is derived from “Unified Nucleic Acid Folding.” Folding (secondary structure) prediction for single-stranded RNA or DNA combines free energy minimization, partition function calculations and stochastic sampling. For melting simulations, the package computes entire melting profiles, not just melting temperatures. UV absorbance at 260nm, heat capacity change (Cp ), and mole fractions of different molecular species are computed as a function of temperature. The package installs and runs on all Unix and Linux platforms that we have looked at, including Mac OS X. Images of secondary structures, hybridizations, and dot plots may be computed using common formats. Similarly, a variety of melting profile plots is created when appropriate. These latter plots include experimental results if they are provided. The package is “command line” driven. Underlying compiled programs may be used individually, or in special combina- tions through the use of a variety of Perl scripts. Users are encouraged to create their own scripts to sup- plement what comes with the package. This evolving software is available for download at http:/ /www. bioinfo.rpi.edu/applications/hybrid/download.php. Key words: RNA folding, nucleic acid hybridization, nucleic acid melting profiles (or DNA melting profiles), free energy minimization, partition functions, melting temperature. The UNAFold software predicts nucleic acid foldings, hybridi- zations, and melting profiles using energy-based methods and a general computational technique known as dynamic program- ming (1). Early software for RNA folding predicted minimum 1. Introduction 1. Introduction Jonathan M. Keith (ed.), Bioinformatics, Volume II: Structure, Function and Applications, vol. 453 © 2008 Humana Press, a part of Springer Science+Business Media, Totowa, NJ Book doi: 10.1007/978-1-60327-429-6 Springerprotocols.com 3
- 21. 4 Markham and Zuker free energy foldings only (2–6). It became clear early on that such methods were unreliable in the sense that many different fold- ings, with free energies close to the computed minimum, could exist. Although constraints deduced from experiments or phylo- genetic analyses could be applied to reduce uncertainty, a method to compute a variety of close to optimal foldings was needed. The mfold software (7–10) computes a collection of optimal and suboptimal foldings as well as a triangular-shaped plot called an energy dot plot (EDP). The EDP contains a dot or other symbol in row i and column j (i < j) to indicate that the base pair between the ith and jth nucleotides can occur in a folding within some user prescribed free energy increment from the minimum. The Vienna RNA Package (Vienna RNA) (11–13) differs funda- mentally from mfold because the underlying algorithm computes partition functions, rather than minimum free energies. This leads naturally to the computation of base pair probabilities (14) and what they call “boxplots.” We call these triangular shaped plots probability dot plots (PDPs). In this case, all base pairs with probabilities above a certain threshold are plotted as boxes (or other symbols) whose area is proportional to the probability of that base pair. This software can compute all possible foldings close to optimal. The sfold package (15–17) also computes parti- tion functions, but it uses a simpler algorithm than the Vienna RNA package because base pair probabilities are not computed directly (exactly). Instead, it computes a “statistically valid sam- ple” (Gibbs sample) that permits the estimation of not only base pair probabilities, but probabilities of any desired motif(s). UNA- Fold encompasses all three methods by computing minimum and suboptimal foldings in the mfold style, and full partition func- tions that allow both exact base pair computations as well as sto- chastic sampling. In addition, stochastic sampling may be used to compute both ensemble enthalpy and ensemble heat capacity for single sequence folding. For practical applications, it is common to use crude methods, often ad hoc, to compute melting temperatures for dimers. It is usual to assume that hybridized strands are perfectly complementary. When mismatches are permitted, they are few and isolated. There is no question about base pairs and no computations are required to determine the hybridization. For example, the simple case: 5′-ACGGtCCAGCAA-3′ 3′-TGCCgGGTCGTT-5′ shows the dimerization of two almost complementary 12-mers. Note the single T⋅G wobble pair. A computer is not needed to compute the hybridization. A calculator would be helpful in add- ing up nearest neighbor free energies and enthalpies from pub- lished tables (18–20), and a computer would be needed to process thousands of these dimers. In any case, free energy at 37°C, ∆G
- 22. UNAFold 5 (or DG37 ) and enthalpy, ∆H, are easily computed. From this, the entropy, ∆S, is easily computed as: ∆ × ∆ − ∆ S = H G T 1000 , where T is the hybridization temperature (K) and the factor of 1000 expresses ∆S in e.u. (entropy units; 1 e.u. = 1cal/mol/K). The terms “free energy,” “enthalpy,” and “entropy” are really changes in free energy, enthalpy, and entropy with respect to the “random coil state.” In reality, “random coil” is a large ensemble of states. The free energy and enthalpy of this ensemble of states may be (and is) set to zero. Only entropy cannot be arbitrarily assigned. Thus a ∆S term is Rln(N/Nref ), where R is the universal gas constant, 1.9872 e.u.; N is the number of states consistent with hybridization; and Nref is the number of states, always larger, in the unconstrained “random coil.” The simplest model for hybridization assumes that there are two states: hybridized, where the structure is unique and known, and “random coil.” It is then a simple matter to deduce a melt- ing temperature, Tm , at which half of the dimers have dissociated. The formula is given by: T f m = H S + R 1000 ln(Ct × ∆ ∆ ) , [1] where ∆H, ∆S (expressed in e.u.) and R have already been defined, Ct is the total strand concentration, and f = 4 when the two strands are different and f = 1 when self-hybridization takes place. This formula implicitly assumes that both strands are present in equal concentra- tions. Competition with folding and with other possible hybridiza- tions, such as homodimer formation, is not considered at all. UNAFold offers a number of increasingly sophisticated ways to compute melting temperatures and also computes entire melt- ing profiles: UV absorbance at 260nm, heat capacity (Cp ), and mole fractions of various single and double-stranded molecu- lar species as a function of temperature. Even when Equation [1] is used, the hybridization is computed by minimizing free energy. The various levels of complexity and choice of methods are described in the following. It is important to issue a warning to those who wish to apply these methods to microarrays. Hybridization on microarrays is complicated by the fact that one of each hybridizing pair is immo- bilized. It is difficult to compute the effective “solution concentra- tion” for such molecules, and diffusion is probably an important factor in slowing the time necessary to reach equilibrium. How- ever, even for hybridization in solution, kinetic simulations that treat different probe-target pairs independently can lead to totally incorrect results for the time required to reach equilibrium as well as the equilibrium concentrations themselves (21).
- 23. 6 Markham and Zuker It is important to emphasize that the computations used by the UNAFold software are based on a number of assumptions. 1. The simulations are for molecules in solution. For microar- rays, computing an effective concentration of the immobilized oligos is necessary. One needs to estimate the “surface phase concentration,” c, in units such as “molecules/cm2 ,” and the solution volume per cm2 of surface area, n (L/cm2 ) (21). 2. The system contains one or two different molecules. If, for example, a true system contains three different oligos, A, B and C, then some common sense must be used. If A and B do not hybridize with each other over the temperature range of interest, then A and C could be simulated separately from B and C if, in addition, the concentration of C is a couple of orders of magnitude greater than that for both A and B. 3. The computations assume that the system is in equilibrium. Thus, the melting profile predictions assume that tempera- ture changes are slow enough so that the system is always at equilibrium. When this is not true, observed melting profiles depend on the rate of temperature change. In particular, a hysteresis effect can be observed. For example, even if the temperature is raised at a uniform rate from T0 to T1 , and then brought down to T0 at the same rate, the measured UV absorbance profiles for melting and cooling may differ. The UNAFold package compiles and runs on many platforms. Table 1.1 lists a few operating system/architecture combinations under which the package is known to work. In fact, the UNAFold package is not known not to run on any platform. The software should function on any system having the following basic tools: ● A Unix-style shell. ● Make. ● A C compiler. ● A Perl interpreter. As noted, UNAFold requires only some very basic tools to build and run. However, there are several optional libraries and pro- grams that provide additional functionality when installed: ● gnuplot: If gnuplot is detected, the hybrid2.pl script will use it to produce several plots in Postscript format. ● OpenGL/glut: If OpenGL and the glut library (22) are detected, the hybrid-plot program will be built. 2. Materials: The UNAFold Software 2. Materials: The UNAFold Software 2.1. Supported Platforms 2.1. Supported Platforms 2.2. Dependencies 2.2. Dependencies
- 24. UNAFold 7 ● gd: If the gd library is detected, the hybrid-plot-ng program will use it to create images in GIF, JPEG, and/or PNG for- mats directly. (With or without this library, hybrid-plot-ng creates Postscript files which can be converted to virtually any other image format.) The UNAFold software is available for download at http:/ /www. bioinfo.rpi.edu/applications/hybrid/download.php. Binaries for Linux (in RPM format) and Windows (in EXE format) are available along with the source code. After downloading and unpacking, building the software consists of three steps: ● Configuration: Usually this is as simple as typing ./config- ure. Running ./configure --help lists options that may be used to configure installation locations, specify locations of libraries, and set compiler options. ● Compilation: A single command—make—compiles and links all programs. ● Installation: Typing make install copies the programs, scripts, documentation, and data files to the location set with the configure script. The core programs in UNAFold are written in C and are opti- mized when compiled since they are computationally intensive. Man pages exist for all of these programs. In addition, invoking any program with the --help option will generate an abbreviated set of instructions. Some programs have counterparts formed by 2.3. Downloading and Installing 2.3. Downloading and Installing 2.4. Core Programs 2.4. Core Programs Table 1.1 Some platforms supported by the UNAFold package Operating system Architecture Linux x86 Linux x86_64 FreeBSD x86 SunOS sparc IRIX mips IRIX64 mips AIX powerpc MacOS powerpc MS Windows x86
- 25. 8 Markham and Zuker adding the suffix -same. When simulating a one-sequence ensem- ble (in which the dominant dimer is a homodimer rather than a heterodimer) the -same version replaces the regular one. ● Folding: — hybrid-ss: Computes full partition functions for folding RNA or DNA. It may be run using the --energyOnly option, in which case the probabilities of base pairs and of single-stranded nucleotides and dinucleotides will not be computed, saving significant time and memory. It may also be run with the --tracebacks option to gen- erate a stochastic sample of foldings. UNAFold also includes two simplified versions of hybrid-ss: hybrid- ss-simple and hybrid-ss-noml. hybrid-ss-simple assigns fixed entropic costs to multibranch loops and ignores single-base stacking, while hybrid-ss-noml does not allow multibranch loops at all. — hybrid-ss-min: Computes minimum energy foldings. It can predict a single, minimum free energy folding or, using an extended algorithm, generate an mfold-like collection of foldings and an EDP. For fast folding of many sequences with the same parameters, the --stream option reads sequences, one at a time, from standard input and writes free energies to standard output. ● Hybridization: — hybrid: Computes full partition functions for hybridiz- ing RNA or DNA (without intramolecular base pairs). The --energyOnly and --tracebacks options function as in hybrid-ss. — hybrid-min: Computes minimum energy hybridiza- tions. The --mfold and --stream options function as in hybrid-ss-min. ● Ensemble computations: These programs rely on the output of the core programs. — concentration(-same): Computes the mole fractions of different molecular species using the already com- puted free energies for individual monomer (folded) and dimer (hybridized) species. — ensemble-dg(-same): Computes ensemble free ener- gies using species free energies and computed mole fractions. — ensemble-ext(-same): Computes UV absorbance at 260nm using computed probabilities for single-strand- edness of nucleotides and (optionally) dinucleotides, computed mole fractions, and published extinction coefficients (23).
- 26. Discovering Diverse Content Through Random Scribd Documents
- 27. resembled an arsenal. The Asiatics lolled about the steps and slept in the hall, and barely moved for us to pass. We picked our way among the reclining forms, climbed the steep steps, and stalked through a broad bare corridor, where our footfalls sounded like thunderclaps, to a reception-room, of which the only furniture was several small round coffee-stools. The walls were hung with Turkish rugs, of an indifferent quality, behind the usual divans, which were part of the construction of the building. The Turks, as is their way, and the other occupants of the house because the bishop was taking a siesta, walked the bare boards shoeless. It was not necessary to inform him of our arrival. A tousled head poked itself out of a door ready to say something a bishop shouldn’t, but, spying us, jerked itself back. We were required to wait fifteen minutes for his holiness to don his robes. Then he appeared in a flutter of excitement. Pouring out unintelligible apologies, he rushed up to my fat friend, being the elder, threw his arms around him, and smacked him twice on each round cheek. I saw I was to be treated likewise—there was no hope of escape—so I bent to the ordeal, to save the bishop the trouble of mounting a stool in all his robes. After he had finished with me the loving soul stooped and gave even the little dragoman four resounding kisses. The Metropolitan was a man of about sixty years of age, with pronounced Hellenic features. His beard and hair were almost entirely grey, but both were full and abundant still. He wore no hat, and his long hair was drawn straight back and done in a knot, like a woman’s. The bishop was alive to opportunities, and the unexpected arrival of two newspaper correspondents was a great chance for him. It quite caused him to lose his dignity for the time being in an effort to do the cause he espoused a service. He explained the presence of the soldiers below; he had received a letter from the insurgents telling him they would kill him unless he desisted from thwarting their diabolical propaganda. Then, as a preliminary to a lengthy discourse
- 28. on Bulgarian atrocities, the bishop cautioned us to believe every word he said. Indeed, we could take his word as we could that of an English gentleman, and we could publish everything he said, even if the committajis slew him for it. The old man here paused, at our request, for the interpreter to translate his remarks, and while interrupted, he called several attendants and despatched them in different directions—two to the Greek school for ‘professors,’ another to the kitchen for coffee and jelly, and still a fourth on another mission—all for our enlightenment and material benefit. Then he resumed his lecture, during the course of which the professors began to arrive, and with them came also a member of the Greek community, who, the bishop proposed, should lodge us that night. The professors joined the bishop in blaspheming the Bulgars, but our host-to-be only substantiated accounts of atrocities at the appeal of the others. Three little girls, who had to be dressed, were sent into the room. They courtesied as they entered and kissed our hands. These were the orphans of a man who had been assassinated by the committajis because he refused to contribute to their revolutionary fund. These ‘brigands’ had murdered several priests in the district, mutilated their bodies in a shocking manner, and laid them in the high-roads or before their churches as a warning to their compatriots. No punishment, said the Metropolitan, was too severe for such fiends, and, questioned by us, he declared that he informed the authorities whenever he learnt that there was a band in the district. We asked the bishop for some information of the affair at Armensko, but this was not in the line of his discourse, and he evidently did not care to complicate the Balkan question for our uninitiated minds. The great question was the Bulgarian propaganda. He dispensed with the massacre as a ‘mistake of the Turks; they should not have done what they did,’ and returned to the insurgent question. We took notes of the Metropolitan’s remarks, but he was dissatisfied that we should permit any to go unrecorded. Finally, as we started to leave, the old man said, with a touch of resentment in his voice, ‘I
- 29. wish I knew English; I would write letters to the Times and let the world know the truth.’ We went home with the Greek to whose tender mercy the bishop had consigned us for the night. A meal was already served when we arrived at his house, and his daughter, a pretty girl about twelve years of age, attired in her newest native frock, stood ready to wait on us, trembling at the honour. But the old man drove her from the room, closed and bolted the door, and cautiously approached our dragoman. ‘Tell the Englishmen,’ he said in a whisper, ‘that the bishop is a terrible liar!’ The interpreter was an English boy, whom we had picked up at Salonica, and the peasants were not afraid to talk to him, as they would have been to another native. It was obvious that the old man had more to say, but we put him off until we had eaten. Then, again carefully ejecting his gentle offspring, he proceeded to inform us that the father of the little orphans we had seen had joined an insurgent band, and then informed the bishop of the band’s plans; and the bishop had transmitted the information to the authorities. The traitor was discovered, hence his death. When the Metropolitan was in Armensko, the Greek said, he told the people that if the Turks came they should go out and meet them and tell them they were Greeks. The Turks came, the peasants went out to meet them, but the Turks did not give them time to announce their national persuasion.
- 30. ORTHODOX PRIESTS. The troops who destroyed Armensko were commanded by Khairreddin Bey, a man already notorious for his methods. According to a report of the committee, the Turks had met a body of 400 insurgents at Ezertze and been defeated. At any rate, the Turks turned back towards Florina, and on their way passed through Armensko, a village of about 160 houses. Without warning they fell upon the inhabitants, slaughtered about 130 men, women, and children, and plundered and burned the houses. Some Roman Catholic sisters of charity, who conduct a free dispensary at Monastir, secured permission from the Governor-General to proceed to Armensko and relieve the wounded. They arrived a week after the affair, and found as many as sixty living creatures huddled together in the two churches, the Greek and the Bulgarian, which, though plundered, had not been destroyed. The human bodies had all been buried, but the carcases of burned pigs, horses, and cows were still lying among the ruins, decomposing and befouling the atmosphere. The sisters, whom we saw after their return, said that some
- 31. revolting crimes had been committed upon the women. They gave the foreign Consuls at Monastir details of the affair, and the Governor-General was indignant, and permitted them to go to the relief of no more massacred villages. The sisters brought the survivors to Florina, and those severely wounded they took on to Monastir. The peasants were all the same people; the same blood coursed through their veins, and they spoke the same language, a corrupted Bulgarian, their vocabularies containing some Greek and many Turkish words; but some were ‘Greeks,’ and some were ‘Bulgarians.’ The ‘Greeks’ were received by the Greek hospital, but admittance was refused those who had rejected the offer of the Metropolitan of Florina to become ‘Greeks,’ and there was nowhere else to take them but to the Turkish hospital. The subjects of the Sultan do not love one another. The rivalry between the racial parties—they cannot be defined as races—works death and disaster among the Macedonian peasants. Bulgarian and Greek bands commit upon communities of hostile politics atrocities less only in extent than the atrocities of the Turks. Sometimes Servian bands enter the field. But the propagandas also greatly benefit the people. The Bulgarian, Greek, Servian, and Rumanian schools—tolerated by the Government because they divide the Macedonians—give the peasants an education which they would not acquire at the hands of the Turkish Government. In the large centres the ‘gymnasiums’ offer the inducements of higher education, and in some cases music and art, for which professors are brought from Budapest and Vienna. Children are often supplied with clothes, boarded, and lodged without charge. All this effort is to possess the greatest share of the community when the division of the country comes. As far as the peasants are concerned, I believe it would make very little difference whom the country goes to, as long as the Government is liberal and equitable.
- 32. Indeed, I found sympathy with the Bulgarian cause among many Greeks, Vlachs, and Servians, simply because the Bulgarians are fighting the Turks. The Greek clergy and other propagandists worked hard to influence us. They brought documents to prove their contentions. But figures lie in Turkey. A little thing like figures never bothers one of the ‘elect’; a Turk can supply official documents proving anything—a map coloured red as far as Vienna, or a census of the population showing more Mohamedans in the land than there are inhabitants. And the other races to some extent copy the Turk. Some of the Greek partisans contended that the major part of the country was peopled by Greeks, but wiser men explained that many members of the Greek community spoke Slav languages and Vlach, but that they are Greeks, nevertheless, because their sympathies are Greek. ‘The inhabitants of Normandy are not British,’ they said. ‘But is not this sympathy unnatural—the work of your clergy, by means not wholly righteous?’ They said the adhesion of the other races to the Patriarchate was entirely natural; the Bulgarians converted artificially with brigand bands. The Greeks fear that an autonomous Macedonia—for which the Bulgarian committees are striving—would be annexed by Bulgaria, as in the case of East Rumelia. The Greeks, therefore, support the Turks, until such time as Macedonia becomes Hellenic. They have been at work for a century converting the country. Before the creation of the Exarchate, when there was but one Orthodox Church in European Turkey, they strove to destroy the Bulgarian language, abolishing it from the schools and churches. When the new Church was established they stamped it schismatic; and many Bulgarians were afraid to leave the old Church, and remain to-day faithful to the Patriarchate—and members of the Greek community. Some Greek partisans claim also the Servian communities of Macedonia because the Servians have no autocephalous church, and
- 33. all Greeks claim the Vlach communities. The Kutzo-Vlachs, or Wallachians, are a people akin to the Rumanians. They speak a language similar to that of the Rumanians, evidently a Latin tongue. The kingdom of Rumania claims these people, and conducts a propaganda among them to retain them, in the hope of securing territorial compensation—a corner of Bulgaria, perhaps—at the division of Macedonia. Until 1905 the Vlach churches were also under the direct control of the Patriarchate; but Rumanian influence at Constantinople then obtained their independence. The Greeks contested the separation violently, and sought to prevent by force the installation of the Vlach clergy. Rumania, not being contiguous to Turkey, was unable to give battle with armed bands, and declared a civil war upon Greece. Diplomatic connections were severed, trade treaties abolished, and Greek shipping in the Danube was severely taxed.
- 34. CHAPTER IX ACROSS COUNTRY Travel in Turkey is severely restricted. If a native succeeds in obtaining a teskeré, or the visé thereto, necessary for making a journey, there is still the deterring danger of arrest on suspicion at his destination or en route, in spite of his papers. If he is a non- Moslem he is suspected of nothing worse than being a revolutionist, and is only set upon by polite police officers; but if he be Mohamedan, he is required to deal with the spies of the Sultan. I once witnessed in Salonica the impressive military funeral of a pasha who had been in high favour at Court. So highly was the pasha esteemed that the Sultan sent one of his own physicians, a Greek, from Constantinople to attend him—though, incidentally, the doctor arrived after the pasha’s death. But the unfortunate Turk had not possessed sufficient of Abdul Hamid’s confidence to secure for him permission to visit Constantinople—for which he had applied several months before—in order to have an operation performed there by competent surgeons. Foreigners fare better. They may travel to the limits of the few railway lines without serious annoyance—if they confine their stops to Consular towns. To enter the ‘interior,’ however, permission is seldom given, and Europeans (in Turkey the name includes Americans) are never allowed to leave the railways without an escort. Only on one occasion did we get away from the railways with the consent of the authorities. This was at the instance of a certain Consul, a man who demanded things and got them. The journey was across a section of Macedonia from Monastir, the terminus of one railway, to Veles, an intermediary point on the north-and-south line. As might be supposed, the country was comparatively quiet at the time, the crops were being gathered, and the authorities
- 35. informed us (the Englishman and me) that all insurgents had been ‘suppressed.’ We rode out of Monastir perched high on Turkish saddles, at a dizzy distance above our diminutive steeds. At first we sought to secure our lofty positions by a tight grip of the reins, but they pulled on curb bits, and so tortured our poor little ponies that we soon sacrificed our pride, gave the animals their heads, and ‘gripped leather’ until we learned to balance. Just outside the town our escort, six mounted men, awaited us and fell in with us without so much as a salaam. They were the usual ragged beggars, much patched where they sat, tied up in places, and generally off colour. Across their faded chests stretched many yellow stripes—in lieu of gold braid—which designated them of the corps of Zaptiehs. Three of them wore shoes of the regulation order issued by the Imperial Ottoman commissary department, but the others were more fortunate. Of these latter two possessed native woollen stockings and charruks, and the third had a high boot on one foot and a shoe and leather legging on the other. The leather legging hardly met about the calf to which it was applied, and lacing was necessary to fill a slight breach, while the boot was large enough to admit a long, flute-like cigarette-holder, a tobacco-pouch, and a flint. The fezzes of this brigade were the one uniform thing other than their guns; they were all good, possessed tassels, and one even showed signs of having been pressed at a not far distant date—unlike those which sat upon Christian heads. We discovered early that our escort were very poor horsemen. They did not seem to understand their animals; for though the ponies they rode could have been managed without any bit at all, yet they all kept a heavy hand on a cruel curb. The ponies were small, and had none but natural gaits, and the short trot was most uncomfortable unless one rose in the saddle. This the Zaptiehs were unable to do. In consequence the horse suffered. Two at a time they took turns at riding with us at a steady trot, while the others galloped and walked alternately, thereby covering the same distances we did in the same time.
- 36. A ride across Macedonia affords a wealth of interest. Your escort is a study in Turk; every peasant you meet is a new picture; the mud- brick houses of the Christians and the Mohamedan chiflics are curious and picturesque, and you must stop at times and absorb the scenery. You can sympathise on a journey like this with the small boy who cried because he had so many sweets he could not eat them all. Our route the first day lay through open country, and our escort was therefore quite small. We traversed the length of the Monastir valley and stayed the night at Prelip. It should be a happy, prosperous valley, for Nature smiles on it, but it is desolate and almost deserted. The cornfields hug the towns, and the villages hide themselves in obscure corners of the mountains. The ‘high road,’ a waggon-track, which we followed, skirted one village and passed through another, but they were made up of such huts as brigands would not stoop to enter. A sheep-dog, big framed and thick coated —but a bread-fed, skinny animal, with an uncertain lope and an unsound bark—came at us. One of the Zaptiehs drew his sword and gave it a trial swing at a low bush near his horse’s feet; but a peasant came crying after the dog, and called the brute off before it got within reach of the Turk’s blade. This was a Turk of less religious fervour than his fellows. The Zaptiehs smoked continually as they rode, and rolled cigarettes for us. They gave us lights from their cigarettes, but only the irreligious fellow would accept the same favour from us, for which I asked the reason. ‘They will not take fire from a giaour,’ he said. The insurgents had boasted that the crops would not be harvested this year, but the corn and the tobacco were already on their way to market. We passed Christian caravans which took the fields to give us the road, and Mohamedan carts which made us give them the right of way. The former were unarmed and most meek, doffing their dejected fezzes and standing abject with hands clasped on their stomachs as we passed. The others, down to the half-grown boys, carried pistols and guns, and bore themselves like a ruling race. The Turks, however, appeared to be as poor as the Christians, and once two veiled women, gathering their faded rags about them,
- 37. even to covering their henna-tipped fingers, came up to our horses to beg. Nevertheless, their husband, riding a dwarfed donkey, carried a revolver. The lot of the animals in Macedonia is similar to that of the people. The one survives on grass as the other lives ‘by bread alone.’ The peasant lies down to sleep at night in his clothes, and the heavy- saddled pack-animals are relieved only of their loads. The long, latticed saddle, reaching from before the animal’s shoulders to his haunches, is seldom removed. It becomes in time an integral part of the animal, it conforms somewhat to his shape, and he gives way in places to its lines; and when it does leave a back it often brings hair, and sometimes skin, with it. The animals are not pegged out or tied together when the caravan halts. The system practised is to lock their fore feet with short-chained iron cuffs, or else to tie them with a bit of rope. There are various means of propelling the beasts of burden, but only the carriage-driver uses the Western lash. A donkey is generally sat upon sideways, not astride, and continually beaten with the heels; the horseman wears heavy spurs; the driver of pack- trains, oxen and buffalo teams, carries a pointed stick or a staff with a nail in the end. These last instruments are gently pressed against the hind quarters, and the pressure is kept on till the animal attains the required speed. The buffalo, which is a heavy creature and unable to acquire speed rapidly, lifts his long, snake-like tail and veritably twists it about the tantalising stick. These pitiful-eyed, straight-necked, knock-kneed creatures are larger and more powerful than the ox, and the buffalo cow gives considerably more and richer milk than the domestic variety. But the buffalo is an exceedingly delicate creature, and requires constant care. His hair is long, but thin and scant, and he is addicted to early baldness on the back. In this condition his skin resembles the hide of a rhinoceros. When the weather is warm he drags his slow way along the roads, covered with soft, slimy mud. The driver walks beside him with a crude, long-handled dipper, and at every puddle replenishes the supply of cooling mud. In the winter the black beast maintains the same measured pace, but then he
- 38. wears a different covering. His thick, coarse blanket protects him from the cold—a thing of broad stripes, brown and white, made of the same material of which his master’s cloak is woven, spun by the peasant wife, probably in the same piece of cloth. At several places at which we stopped the peasants came to us to ask medical advice for themselves and their animals, and we were exceedingly sorry that we could not prescribe for either; for their own ideas of doctoring border on superstition, and seem to follow the plan of killing pain by pain. At one village we witnessed (and protested against) the treatment of an unfortunate horse which had, by strange mishap, swollen to an abnormal size. A stout cord was put around its tail close to the root and twisted with a stick until all circulation in the tail was stopped. Then, when the appendage had become numb, a wire nail was driven into it in four places. The horse died of complications, including lockjaw. A horse which, at a stage of the journey, carried our luggage, possessed but one ear. We asked what had become of the other, and were told that it had been cut off piece by piece to cure repeated fits. There is often to be seen in Macedonia, especially in the Monastir district, a thing resembling a big bird’s-nest built on stilts. The nestling wears a soldier’s costume and carries a gun. He is a field guard, an institution of the Government designed to ‘protect’ Christian peasants from ‘brigands,’ Albanian and Bulgarian. This he often accomplishes by becoming a member of a band of the former. The Governor-General will show you yard-long petitions stamped with many tiny seals, the marks of the peasants, pleading that no Christians be put to guard them, as the Austro-Russian reform scheme provides. The signatures to these petitions are not secured in the general way, by a Turk with a loaded gun; they are bona fide. The peasants really do not want the protection of a half-hearted Christian, who has probably never before handled a gun, and who will only bring disaster upon them. The Turkish guard is a contemptuously tolerant creature. His band is strong enough to defend the peasants from other marauders, and so long as they pay the annual tribute of so many sheep or goats, and so much grain,
- 39. there is no other call upon them—except for the needs of the bird in the nest. The committee’s agents, when laying their cause before Europeans, will designate this bird a vulture, and tell you how he exacts maidens of the peasants; but the Greeks, who claim to be the enlightened people of the country, explain that this, to a Macedonian peasant, is not what it is to an Englishman or an American. There are always two sides to a question. Albanians. Bulgarians. CAPTIVES. Though the revolution had not yet occurred, and the peasant population was still engaged in peaceful pursuits, the country swarmed with soldiers. Cavalry and infantry patrols, Turks, Albanians, and Asiatics, passed us by. Occasionally we met a guard with handcuffed prisoners, Bulgarians and sometimes Albanians. Now and then a member of our escort would meet a long-lost friend, and the old comrades would drop from their horses and embrace each other, pressing cheeks first one side and then the other. We were yet an hour off from Prelip when the white tents about the town came into view. Soon we came to the cornfields. The corn was
- 40. ripe and glowing under the slanting rays of the evening sun, and here and there red poppies had wandered in to stud the golden fields. Once the road led by a milk-white field, most innocent in appearance, but covered with the deadly blooms of opium. Many houses on the edge of the town, and some in the narrow streets, were hung from roof to ground with strings of tobacco leaves, changing colour in the sun. When we entered Prelip the natives were gathered at their gates preparatory to withdrawing for the night. It was too late for Christians to follow, and the Turks are too dignified to do more than bestow a casual glance at any traveller. But in the morning our appearance caused a commotion in the town. Greeks left their shops, Bulgarians deserted the market-place, Vlachs followed us with their pack-animals, Jews and gypsies came after us, the one to sell, the other to beg of us; men, women, and children joined in our train. They followed us until we crossed a narrow street, at the other side of which only a few veiled women were visible; then the whole throng came to an abrupt stop. ‘What is the matter with the crowd?’ I asked one of our guards. ‘They are like the dogs,’ he replied; ‘they have their boundaries. At this street begins the Turkish quarter.’ We walked on through the quiet, clean, Turkish quarter and came upon a group of bashi-bazouks, who had been called into service as village guards, squatting by the roadway smoking. They were kind enough to rise and permit me to photograph them standing. This was rather an exceptional case; the Mohamedans generally resented my camera. A gypsy minstrel, a thing of shreds and patches, on his way to a wedding feast, protested that the Evil Eye would be upon him if I took his likeness, but I ‘snapped’ him while he argued. It would have been unkind to inform him.
- 41. TURKISH WEDDING FESTIVITIES. We then followed the Tzigane to the wedding, of which, of course, we were permitted to witness only the street celebrations, those of the male side of the house. This took the form of an almost uninterrupted dance to the monotonous music of two reed flutes and two crude bass drums. The flutes had a range of about three shrill chords, and the drums had two notes apiece. With the right hand and a heavy stick the drummers beat a slow, steady boom, while with a lighter stick in the other hand they kept up a rapid tattoo. They played by ear, of course, and the strain of a single bar of music went for hours. Monotony is bliss to the Mohamedan. A long mixed line of men gave the dance. There were Turks with red fezzes, Albanians with white skull-caps, soldiers, and bashi-bazouks. The leader of the line, swinging a red handkerchief, led the way round a circle formed by the crowd and set the figures, which varied little more than the music. The dance was evidently copied from the Bulgarian horo. Sometimes the leader withdrew in favour of the second man, and now and then a man in the line would fall out, to have his place filled sooner or later. But on went the dizzy dance to the doleful sound all the afternoon.
- 42. My companion trounced a Greek barber at Prelip, and I had my hair cut by accident. We had begun to look like Bulgarian insurgents, with full crops of hair and unshaven faces, and, resolving here to abolish the dangerous likeness in so far as our beards were concerned, we repaired forthwith to the nearest barbers’. The Englishman chose a Greek barbershop, and was shaved by a man with a characteristic nose of large proportions. At the conclusion of the ordeal he inquired the price, and was told that he owed the sum of two piastres. He handed the Greek a mijidieh, which is worth nineteen piastres in Prelip, and received five piastres in change. At this the Englishman protested, and the Greek yielded up another small coin. But more than this no gentle persuasion could move him to give. Among the crowd which had gathered to see the ‘Frank’ shaved was one accommodating individual who spoke a garbled French. The Englishman enlisted his services to make known to the man with the nose that, unless he produced the proper change forthwith he would have his olfactory organ promptly and vigorously pulled. This had no effect, and the threat was put into execution, to the wonderment and increase of the crowd. But nobody protested, and the Greek produced another insignificant coin. Again the interpreter was employed, and again without result. So again the Englishman laid his hands on the Greek, and this time so ill-used the poor man that he handed the key to him and told him to help himself with piastres from the money drawer. The Englishman took the proper change and departed. My experience was less thrilling, but the disfiguring was of me. I discovered a Turkish barbershop, consisting of a Turk and a towel, a cane-bottomed stool, and some utensils made in Austria. The shop occupied the narrow pavement with the dogs, out of the way of the pedestrians. After shaving me with a heavy weapon, the Turk held up a formidable pair of scissors by way of asking if I wished to have my hair cut. For the moment I forgot that a shake of the head in Turkey means ‘yes,’ and a nod means ‘no’—and I shook my head. I was rescued from the wall against which I had been reclining during the process of shaving, and straightened up for the purpose, I
- 43. thought, of having my hair combed. But the Turk, with a single clip, took off a large bunch of hair, and left me, without alternative, to be barbered in the latest Prelip fashion. A GYPSY MINSTREL. A TURKISH TRUMPETER. The Turk does a great many things in an opposite way to which we do them. He writes backwards; the conductor on the horse-car at Constantinople and Salonica punches the tickets for the station at which one gets aboard instead of that to which he is destined; the wood-sawyer rubs the wood on the saw, which he holds between his legs; the sailor, feathering oars, turns the blades forward instead of backward; the officer salutes the soldier. In the interior of Macedonia it is not necessary for the authorities to preserve the same show of order that is required in Consular towns, and our escort for the next stage of the journey came to the khan
- 44. for us. There were a score of Zaptiehs in the charge of a fat but ragged sergeant, who gave me his name but could not write it. This is nothing extraordinary; one of the foreign officers of the reform scheme told me he had found but two sub-lieutenants in the whole Kossovo vilayet who could read and write. For several hours the road led along the sides of a stream winding between two ridges of mountains. The mountains were said to be infested with insurgents; this was a part of the country through which Sarafoff operated. Turks’ heads peered down at us, and silently assured us that the road was overlooked for miles beyond. Studded over the steep slopes, wherever a great boulder protruded far enough for a footing, soldiers were suspended between us and the clouds, which the mountains often pierced. Despite this survey of the route, five of our men straggled out to the front, the foremost a mile in advance. As we would descend one steep slope we could see the vanguard climbing the next. Whenever we came to a blockhouse, always pitched on the highest peak, one of the garrison would bring us cool water from the nearest fountain. The road was good for many miles; it had been constructed only a year before. But the contract had not called for bridges, so bridges there were none, and it was necessary for us to ford every stream. But a few months after this excursion a war-scare set the Government to honest work, and this and several other excellent roads, most of them leading towards the Bulgarian border, were hurriedly completed. Millions to retain, but not one cent to maintain. Not a single village did we pass this day, only one lone wayside khan. Macedonia is sparsely inhabited. Once we came over the crest of a hill and descried a gathering of twenty or thirty men far down in a valley below—a little island formed by a split in a thin stream. It took us an hour to get to the island, which lay in our route, and meanwhile men mounted their horses and rode away into the mountains, and others appeared from unseen places and came to the meeting. This was too open a spot—visible from any of the surrounding hills—for brigands to divide spoils; nevertheless the
- 45. business was illicit. We got off our horses and penetrated the crowd. In the centre sat a Turk with two sacks of cut tobacco. This he was selling direct to consumers, without paying the tax levied by the Turkish Regie. We filled pockets for two metaleeks—a penny between us—and proceeded on our way up the opposite mountain- side. OUR ESCORT FORDING A STREAM. This was a hard day’s ride. It would not be exact to say that we were in the saddle ten hours, for we dismounted and walked over many steep mountains, but we were on the road from six in the morning until six in the evening, allowing two hours for halts. We passed through the camp of an Anatolian regiment pitched beside the vast caverns of Veles, dropped down the Vardar, and crossed by the only bridge in view of many primitive wooden water-wheels. The bazaar began at the bridge and ended at a Turkish khan, at which we alighted. There was but one sleeping-room in the khan, and this chamber was equipped with six cots filled with loose cornshucks in lieu of mattresses; there was no other furniture in the room. We wanted to take the room and pay for all six beds, but the landlord preferred to accommodate two Turkish friends, and offered to let us have the other four beds.
- 46. We washed at the tap of the inevitable petroleum tin in the stable, and the proprietor’s son brought us clean but exceedingly rough towels. After our ablutions we repaired to the front of the house, where a dozen or more Turkish officers sat sipping coffee. The ranking man among them, an Albanian, rose as we appeared, and addressed us in French. A Turk would not have spoken without some substantial motive. The Albanian asked where we had come from, where going, how old we were, whether married or not, as rapidly as he could put the questions—which is polite in Turkey. We both understood that this was all in good taste, as was also the noise the other officers made drinking coffee. It was difficult for the Englishman, however, bound by the heavy fetters of British restraint, to reply to this interrogatory readily and with any marked show of pleasure, and quite impossible for him to sip his coffee in the manner of the company. But, having come in contact with many queer people in the course of my travels, I was experienced in such a situation, and not only answered all the Albanian’s questions with alacrity, but put them straight back to him, and while he was speaking I sucked coffee and sighed heavily after each mouthful as though in the height of bliss. This display of good manners met with a cordial reception by the Turks, and they invited us to dine with them at the officers’ mess—an exceptional invitation. We went with them to their quarters in a clean Turkish house, off a narrow street half covered by the extended second storey. We climbed a bare, ladder-like staircase and entered a small, unpainted room with many rugs on the rough boards. There was a long, covered thing like a mattress on one side, stretching from end to end of the floor, and a high divan, likewise stretching the length of the wall, on the other side. I was weary, and the long cushion offered more excuse for reclining, so I dropped myself upon it; but the other man got upon the divan and let his feet hang. We looked foreign to the place, I know; for when the officers were seated there were many pairs of shoes on the floor, but ours were the only feet to be seen, and ours were the only bare heads. Once in a while a Turk
- 47. would remove his fez and rub his head, but generally the red cap sat somewhere on the skull of its owner. A strong native drink, which changed colour like absinthe when water was added—mastica it is called—was served by a Bulgarian boy, who shed his shoes at the door and entered in stocking feet. One of the officers made the boy tell us what good masters the Turks are. Radishes, sliced apple, roasted monkey-nuts, and a delightful little Turkish nut were served and left in the room an hour before dinner. The Englishman and I ate heartily of these, for we were ravenous, and it was well that we did. When the meal came on we all drew around a small wooden table. Six of us sat in so many chairs, and the others stood around behind us, and reached over our heads for their food. We were each supplied with a hunk of bread, a fork, a spoon, and a towel, but no plates were distributed. One dish at a time was placed in the centre of the table, and removed when it was empty. The meal varied from stewed lamb to little squares of lamb toasted on sticks, going through five courses of lamb. Then there was fruit and coffee. There was wine, and five of the Turks drank it; devout Mohamedans do not. At this meal I failed in Turkish manners, even as the Englishman had done previously. We were all required to stick our forks and spoons into the single dish and dig for ourselves, and when the meat was gone to sop our bread in the gravy. But we were both continually withdrawing our forks as another man advanced his, which the Turks did not understand. Of the first few courses we got very little, but then the Albanian caused the officers to give us a two minutes’ handicap at the succeeding dishes. After dinner there was Turkish music—which was not pleasant. The reed flute played in the Turkish street harmonises with the character of the country, and is not unattractive; but in a close room its monotony is inclined to put the weary travellers to sleep. The low wail of a Mohamedan priest calling the ‘faithful’ from a minaret is ‘like the sighing of the pines,’ but the whine of a Turk at close quarters, accompanied by the facial contortions necessary to his
- 48. nasal chant, is conducive to bad dreams. We had our revenge; the other man retaliated with ‘Alice, Ben Bolt.’ Several of the officers escorted us back to the khan through the silent street, answering the challenges of the night patrols. Two dark figures, which followed us from the officers’ quarters, entered the khan behind us and stretched themselves on the floor before the door of the general sleeping-room. There we found them when we emerged in the morning; they proved to be two soldiers to whom the authorities had assigned the duty of ‘shadowing’ us. They told us, with much amusement, of how they had lost us the night before. Arriving at the khan about nine o’clock, they were informed that we had ‘disappeared’; the khanji had not seen us leave with the Turkish officers. This alarmed the soldiers, and they started on a search for us. They were about to report our disappearance to headquarters, when, coming to the Turkish quarter, they heard strange sounds never before perpetrated in Veles. This was the song of ‘Sweet Alice.’ In the morning a negro merchant arrived at the khan from Istip and told us of a fight ‘in progress’ at Garbintzi, a little village about eight hours’ ride to the east. We had intended to take the train that afternoon for Uskub, but the chance of seeing a fight caused us to change our plans. We gathered as much hurried information as we could about the route, hired a Turkish guide, and set off for Garbintzi before noon. We planned to go unescorted, but this was not to be. Our guide, in pursuance of police orders, had informed the Konak of our sudden change of destination, and the kaimakam despatched four Zaptiehs to accompany us. We were surprised that they permitted us to proceed. Being anxious to reach the scene of the combat as quickly as possible, we rode rapidly over the mountains, and came to Istip about six o’clock. An officer came up as we entered the town and greeted us like long- lost brothers. He was a Turk, and had a mission to perform. He
- 49. informed us that the kaimakam had received a telegram from Veles advising him of our approach, and instructing him to see that we were treated in a manner befitting our exalted positions. The only place they could offer such worthy guests, who had so honoured Istip with a visit, was the kaimakam’s own house. The kaimakam, I may explain, lived above the gaol. We were presented to the kaimakam, and the official congratulated the Englishman on belonging to that great race which had so long befriended the Turks. To me he said he thought it wonderful that a great New York paper would send so youthful a man so many miles on so important a mission. ‘How old are you?’ he asked. ‘Twenty-five,’ I replied. ‘You look eighteen.’ He did not ask why I wore no moustache, probably fearing it was because I could not. The Turk is a gentleman. Information had evidently been given by our escort that we carried revolvers, for two officers entered the room through a door at the back, drew up chairs, and seated themselves immediately behind us. But we did not attempt to shoot the kaimakam. Another officer, perhaps the spy attached to the governor, also entered and occupied a seat beside his quarry. Then the kaimakam brought his compliments to an end and sat silent. Nobody spoke for forty seconds. We sought to end the uneasy interview, and informed the kaimakam, what we were sure he already knew, that we were on our way to Garbintzi. ‘The fight is over; the troops have just returned,’ he informed us. ‘That is unfortunate,’ I replied, ‘but as we have come this far I guess we’ll visit the scene.’ But the kaimakam guessed we wouldn’t.
- 50. ‘I have orders,’ he said, ‘to prevent you from going any further. You must return to Veles.’ We suggested that the Governor-General was making a mistake; if we were not allowed to visit Garbintzi we must conclude that the reports that massacre and arson had accompanied the fight were true. The Englishman added that, if the Turkish version were based on fact, it would be well to let us verify it. But the kaimakam shook his head; he had his instructions. We left the house extremely disappointed, and on the way to the khan—for he had said nothing about putting us up—began to think out a plan for getting to Garbintzi. We went to our guide, and, feigning extreme dejection, instructed him to saddle, and be ready himself at eight o’clock next morning; we were going back to Veles. An officer visited us during the evening to ascertain what time an escort should be ready to take us back. The information we gave him agreed with that we had given the Turkish guide—which had been imparted to him. Putting the question to us was only a point of politeness: the horses were being watched. We rose at five o’clock next morning, dressed hurriedly, and went to the stables. Two soldiers had slept there, and one set off at a run to the Konak. But the hour was early for the Turks, and we got out of town without a soldier on our heels. We passed the sentinels on the border of the town and rode hard in the direction of Veles until we had passed out of sight of a blockhouse which stood high on a hill a few miles beyond, and would, no doubt, report that we had fairly gone by towards the railway. It was a ride of barely ninety minutes from Istip to Garbintzi by road; with a good hour’s start, we calculated that we could get there before being overtaken, even though we went by a roundabout route. But we did not reckon with our guide. When we called a halt and asked him if there was not a road over the mountains to Garbintzi, he was frightened. He answered that there was a way, but the road was bad, and it would take four hours to go by it from the spot where we stood.
- 51. ‘Lead us over it,’ we said to the dragoman, who repeated the words to the guide. There was a parley of ten minutes, during which our nerves were at high tension. Every minute we expected to see a troop of cavalry coming after us. At last we got the information. ‘He won’t go.’ There was no time for argument, when it had taken so much time and all the Turkish which we had heard to convey that fatal negation. ‘How much does he want?’ the Englishman demanded. ‘He will not go at any price,’ came the reply. ‘He has a wife and children depending on him, and an officer has been to him last night and told him that he should lead us to Veles and nowhere else.’ It was no use arguing. We turned our horses’ heads towards a village of some ten houses a few miles off, half way up a mountain side. The dragoman followed. The guide would not leave the road to Veles, literally following instructions. It was Sunday, and the peasants were all in their brightest clothes. They were dancing a horo, but our appearance among them broke up the festivities. Every man, woman, and child in the village collected about these queer travellers. They understood the dragoman’s Bulgarian, as was apparent by the state of alarm into which they fell. Not for a hundred liras, said the headman of the village, would one of them guide us over the mountains. ‘Why?’ I asked. ‘Why!’ came the answer, ‘the man who should take you over those mountains would be shot by the committajis, for we have refused to arm. Were the Turks to find out that one of us had left here without a teskeré, and taken you to see a village which they had destroyed, they would come and do the same to this place.’ ‘Please leave us,’ they begged, as we still argued, ‘and get away before the Turks see you.’ Several old women began to cry. We returned to our guide, our last card played, and said demurely, ‘Lead us back to Veles.’
- 52. We made our way slowly, and waited at the next khan for a cloud of dust on our trail to develop into a troop of cavalry, who kept a close cordon about us for the rest of the journey back to the railway. Defeated we had been, but we had learned a lesson in the ways of the Turk, who thinks his intelligence is superior to that of a mere ‘giaour.’
- 53. CHAPTER X USKUB AND THE SERBS After our attempt to evade the authorities we were closely watched until we left Veles, the police, as is their way, pretending to wait upon us only for our convenience. When we departed two mounted gendarmes accompanied us to the railway station, though we needed no protection, and a careful sleuth, with painful politeness, assisted us in taking tickets for Uskub—an unnecessary courtesy— and went with us to the train to see, he alleged, that we secured a comfortable compartment. There was only one first-class compartment in the train, and this was occupied by a well-dressed officer whose trousers had been pressed inside out. The Turkish gentleman stood not upon ceremony, as does his admiring British contemporary on such occasions; he introduced himself before we had taken our seats, immediately inquired our life history, and soon divulged what purported to be his. He was no other than Hamdi Pasha, of Albanian extraction, the youngest general in the Turkish army, so he informed us, on his way to the Bulgarian border, of which he was military inspector. It was raining heavily when we arrived at Uskub; nevertheless, a picked company of Nizams (regulars) was drawn up in honour of our travelling companion, and presented arms as the train pulled in. The pasha alighted, saluted, and, with us on either side of him, sharing a great white umbrella, proceeded to the Hôtel Turati. Then the bedraggled band struck up one of several Sousa compositions which have been Orientalised for the Ottoman army, and the company marched away through the slush, doing the German ‘goose’ step, acquired from the Kaiser’s officers in the Sultan’s service, which showy effort spattered the mud on civil pedestrians on both sides of the narrow street.
- 54. Behind the soldiers straggled several hundred Albanians, raw Redifs (first reserves), who had come up on our train in cattle-cars marked in bold letters, in a language they knew not of, ‘8 Chevaux ou 48 Hommes.’ And behind the Arnauts trailed a score of prisoners protesting violently at being driven to gaol through the mire. These were Christians impregnated with the sense of free men’s rights. They were attired in ‘Francs,’ fezzes, and handcuffs—with the exception of one, a priest, who wore only the manacles in common with the others, apparently the conductors of a Bulgarian gymnasium temporarily out of business. Before the school teachers paraded a grinning gypsy bearing on his back a bundle of old muskets. ‘See, see!’ said the pasha. ‘They were captured in arms. There are the guns.’
- 55. ‘8 CHEVAUX OU 48 HOMMES’: ALBANIAN RECRUITS. But a foreign Consul, wise in the ways of the wily Government, told us that this gypsy and his parcel of rifles was the ostentatious advance guard of every detachment of Bulgarian prisoners. The manœuvre was designed to deceive those representatives of the Powers and newspaper correspondents who were particularly prying. Uskub is a stern place with a breath of the mountains upon it. It is but an eight hours’ journey from Salonica, but, thanks to the restrictions of travel and intercourse, wholly free of a Levantine atmosphere. It is peopled principally by Arnauts—as the Turks call the Albanians—and Slavs, both men of character, though their morals are of a peculiar code. These Albanians and Slavs are natural enemies, and of the Slavs again there are Bulgarians and Servians,
- 56. Welcome to our website – the ideal destination for book lovers and knowledge seekers. With a mission to inspire endlessly, we offer a vast collection of books, ranging from classic literary works to specialized publications, self-development books, and children's literature. Each book is a new journey of discovery, expanding knowledge and enriching the soul of the reade Our website is not just a platform for buying books, but a bridge connecting readers to the timeless values of culture and wisdom. With an elegant, user-friendly interface and an intelligent search system, we are committed to providing a quick and convenient shopping experience. Additionally, our special promotions and home delivery services ensure that you save time and fully enjoy the joy of reading. Let us accompany you on the journey of exploring knowledge and personal growth! ebookfinal.com