American Journal of Epidemiology Advance Access originally published online on February 8, 2006
American Journal of Epidemiology 2006 163(7):684-685; doi:10.1093/aje/kwj102
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Book Review |
Statistical Methods in Genetic Epidemiology By Duncan C. Thomas
ISBN 0-19-515939-X, Oxford University Press, New York, New York (Telephone: 800-445-9714, Fax: 919-677-1303, Website: http://www.oupusa.org), 2004, 464 pp., $65 (Hardcover)
The Seeing Eye, Inc., Morristown, NJ 07960
(e-mail: eleighton{at}cox.net)
Genetic epidemiology is rapidly evolving as a cross-discipline between genetics and epidemiology, served up with a healthy dose of statistics. It attempts to understand the genetic etiology of disease and, to do so, applies old statistical tools in new ways or builds new tools from more traditional ones common in genetics or epidemiology. The jargon can sometimes be daunting to define. Because this is a very young field, precious few books have been written specifically for newcomers. In his preface, Professor Thomas declares his objective to present "a broad overview written at a level that should be accessible to graduate students in epidemiology, biostatistics, and human genetics." In this reviewer's opinion, Professor Thomas has, for the most part, reached his objective.
The book is organized into 12 chapters that conceptually comprise two parts. Part 1, chapters 1–4, presents background material and reviews tools needed to understand part 2, chapters 5–12, where the real concepts of genetic epidemiology are explained. Chapter 1 is a basic introduction to the field. Chapter 2 (contributed by Dr. Sue Ingles) introduces molecular genetics. Chapter 3 presents fundamental principles of Mendelian inheritance, while chapter 4 covers basic principles of epidemiology and statistics. Part 1, then, provides the foundation for understanding part 2.
Part 2 begins with chapter 5, which discusses the various tests that can be used to discern whether disease aggregates in families. Chapter 6 explores segregation analysis, which attempts to determine the most likely genetic mode of inheritance given the data at hand. The very important topic of linkage analysis is covered in broad strokes in chapter 7, while chapter 8 discusses concepts of population genetics as they relate to genetic epidemiology. Chapter 9 looks at procedures used to search for associations between a candidate gene and a disease phenotype, including a discussion of study design options such as cohort studies or case-control designs, that will be familiar to traditional epidemiologists. Chapter 10 covers how linkage disequilibrium mapping can be used to test hypotheses about associations between a specific allelic form of a gene and a disease phenotype. Chapter 11, entitled "Gene Characterization," explores techniques for estimating genetic risk, again using familiar study designs such as cohort or case-control designs with unrelated subjects. With chapter 12, the book concludes by reviewing the genetics of colorectal cancer.
Clearly, this book is intended to introduce genetic epidemiology to graduate students or trained professionals working in related disciplines such as genetics, statistics, or traditional epidemiology. To understand much of the material in part 2, the reader should already have a good foundation in basic statistics, including maximum likelihood theory and general linear models. Chapter 4 reviews these procedures along with Markov chain Monte Carlo methods and randomization procedures, but, although well written, this review would likely be an insufficient introduction to the techniques for all but the most mathematically astute geneticists or epidemiologists.
Many of the chapters in part 2 are worthy of a book in their own right, so readers of this book should expect to obtain only an overview of topics such as linkage analysis, population genetics, or methods of testing for candidate gene associations. However, these chapters are also well written, and they do provide a clear introduction to these very important topics, including very substantial mathematical development of major concepts and tools. Nevertheless, scientists working with animal populations should know that the concepts covered are discussed from the human genetics perspective; there is almost no mention of the additional complexities involved in solving a problem that includes the presence of inbreeding loops.
An important theme used throughout the chapters in part 2 is the generous use of examples, some based on simulated data and some taken from the human genetics breast cancer literature. In addition to the material actually covered in the book, also included is a bonus section with a 19-page glossary and 40 pages of references cited in the book. Indeed, the book may be worth purchasing just for the reference list alone.
Even though this book is very well written and logically organized, there are, as expected, a few shortcomings. Lecturers looking for a textbook will not find any problem sets at the end of each chapter. Likewise, there is no errata page, either online or available by download or e-mail from the author, unlike the outstanding online errata page for Genetics and Analysis of Quantitative Traits (1
). There are also some mistakes. For example, while discussing a recessive X-linked gene in chapter 3 (page 57), Professor Thomas states that "only males are affected" even though a supporting figure shows an example pedigree that contains an affected female. Technically, it is possible for a female to be affected (2), but this situation happens so rarely that the condition is often described as the author states.
In summary, Statistical Methods in Genetic Epidemiology provides a clearly written, well-organized introduction to the field of genetic epidemiology. It is a great place for graduate students or scientists working in related fields to start learning about the topic.
ACKNOWLEDGMENTS
Conflict of interest: none declared.
References
- Lynch M, Walsh B. Genetics and analysis of quantitative traits. Sunderland, MA: Sinauer Associates, Inc, 1998. (Errata: http://nitro.biosci.arizona.edu/zbook/vol1errors.html).
- Refer to the topic "Sex Linked Recessive" (http://www.nlm.nih.gov/medlineplus/ency/article/002051.htm).
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||