Quicke Donald - Practical R for Biologists

PRACTICAL R FOR BIOLOGISTS

An Introduction

PRACTICAL R FOR BIOLOGISTS

An Introduction

Donald L.J. Quicke

Buntika A. Butcher

and

Rachel A. Kruft Welton

CABI is a trading name of CAB International

Donald Quicke, Buntika A. Butcher and Rachel Kruft Welton 2021. All rights reserved. No part of this publication may be reproduced in any form or by any means, electronically, mechanically, by photocopying, recording or otherwise, without the prior permission of the copyright owners.

References to Internet websites (URLs) were accurate at the time of writing.

A catalogue record for this book is available from the British Library, London, UK.

Library of Congress Cataloging-in-Publication Data

Names: Quicke, Donald L.J., author. | Butcher, Buntika A., author. | Welton, Rachel A. Kruft, author.

Title: Practical R for biologists : an introduction / Donald L.J. Quicke, Buntika A. Butcher, and Rachel A. Kruft Welton.

Description: Wallingford, Oxfordshire ; Boston, MA : CAB International, [2021] | Includes bibliographical references and index. | Summary: A new textbook showing beginners how to use the free programming language R for fundamental biostatistical analysis, graphical display, and experimental design. The book takes a simple step-by-step approach to give a good grounding in the use of R for undergraduate/beginning postgraduate biology students-- Provided by publisher.

Identifiers: LCCN 2020025099 (print) | LCCN 2020025100 (ebook) | ISBN 9781789245349 (paperback) | ISBN 9781789245356 (ebook) | ISBN 9781789245363 (epub)

Subjects: LCSH: Biometry--Data processing. | R (Computer program language)

Classification: LCC QH323.5 .Q56 2021 (print) | LCC QH323.5 (ebook) | DDC 570.1/5195--dc23

LC record available at https://lccn.loc.gov/2020025099

LC ebook record available at https://lccn.loc.gov/2020025100

ISBN-13: 9781789245349 (paperback)

9781789245356 (ePDF)

9781789245363 (ePub)

Commissioning Editor: Ward Cooper

Editorial Assistant: Lauren Davies

Production Editor: Tim Kapp

Typeset by SPi, Pondicherry, India

Printed and bound in the UK by Severn, Gloucester

This book is dedicated to all the people who help look after street dogs throughout the world. This is the first author with Ma Pao, who is cared for by those working at noodles shop and a coconut stall near Saphan Taksin station, Bangkok.

Contents

Online Supplementary Appendices

1. Online Resources: Data Files

2. Online Resources: Complete R Codes Used for Graphs, Analyses and Simulations

3. Online Resource: Suggested Answers to Exercises

These Online Resources can be found at: www.cabi.org/openresources/45349

Prof. Dr Donald Quicke has had more than 40 years experience teaching undergraduate and postgraduate biology students, initially at Sheffield University, UK and then at Imperial College London.

Buntika Butcher gained her PhD at Imperial College and is currently Associate Professor in the Biology Department at Chulalongkorn University, Bangkok, with 20 years of teaching experience.

Dr Rachel Kruft Welton did her masters degree at Imperial College, London and a PhD at University of Birmingham, UK before qualifying as a teacher. She has been a professional biology and science tutor for nearly 20 years, including mentoring undergraduates as part of Birmingham Universitys alumni scheme.

Collectively the authors have a vast amount of teaching experience which they apply here to make the passage into R programming as gentle and easy as possible, whilst guiding the reader to tackle quite complicated programming.

There are many easy ways to create simple graphs using common computer software. The popular Microsoft Excel for example, offers a variety of choices that can be called simply by highlighting blocks of cells. It even offers some built-in basic statistical tests that are often introduced in biology A-level courses in some western countries (they certainly were in the senior authors day in the UK). However, there are very many limitations to just doing that. Not only do you have virtually no control over the presentation of the graphics, the simple stats provided will often give completely misleading answers. There are other potential problems too. With programs such as Excel it is very easy to mess things up; it is easy to fail to sort all the columns or to select all the rows. There is an infamous case of just that, the so-called Reinhart and Rogoff effect, a wrongly sorted spreadsheet that led to totally wrong economic conclusions (Herndon et al., ) that arguably led to the whole western austerity drive from 2010 onwards. Because R is an explicit, scripted language it is far harder to make such devastating mistakes with datasets.

What Is R?

R is an open-source statistical environment modelled after the previously widely used commercial programs S and S-Plus, but in addition to powerful statistical analysis tools, it also provides powerful graphics outputs and, as we are all taught at an early age, a picture is worth a thousand words. R users with expertise are constantly adding new associated packages, but the range already available is immense. R is now used by a very wide range of people, biologists (who are the main targets of this book) for sure, but also other scientists plus economists, market researchers, medical professionals, epidemiologists, historians, geographers and those working in numerous other fields.

In addition to its statistical and graphical capabilities, R is a programming language suitable for medium-sized projects. It is not suitable for things like CGI (computer generated imagery) or gaming for which very powerful specific software applications exist, but for modelling things like population dynamics, it offers a very fast and free solution.

In this book we will work our way through a set of studies that collectively represent almost all the R operations that beginners, analysing their own data up to perhaps the early years of doing a PhD, need. Although the chapters are organized around topics such as graphing, classical statistical tests, statistical modelling, mapping and text parsing, we have chosen examples based largely on real scientific studies at the appropriate level and within each we nearly always cover the use of more R functions than are simply necessary just to get a