Genetic Analysis of Complex Traits Using SAS.
The title leads the reader to expect a book that demonstrates how to use SAS in the analysis of molecular genetic data, and certainly, this is one of the book's objectives. But surprisingly, and no doubt to the delight of plant and animal breeders and traditional quantitative geneticists, the book also contains a very comprehensive treatment of traditional quantitative genetic data analyses using SAS. As the editor notes in his introduction, these conventional quantitative SAS applications are long overdue. While quantitative genetics textbooks have continued to be published, the theory they present is typically unaccompanied by statistical software applications that take advantage of modern computational techniques and hardware. Many who have taught quantitative genetics are painfully aware of this fact, and they will cheer the arrival of this volume.
The book is divided into 11 chapters written by experts in their fields. Chapter 1 provides a very brief introduction to the book and to SAS protocols. The remainder of the book is divided into two parts: Part 1 focuses on classical quantitative genetics (Chapters 2-7) and Part 2 (Chapters 8-11) covers molecular genetics. Chapter 2 gives an excellent treatment of estimation of genetic variances and covariances using PROC MIXED. For those who are still estimating variance components by equating mean squares to their expectations, this chapter provides a nice stepping-off point into a more powerful, robust methodology. Chapter 3 covers regression approaches to estimation of genetic gain and realized heritability, among other parameters. In Chapter 4 there are SAS applications for estimation of genetic gain from single trait selection, indirect selection and index selection as well as selection based on best linear unbiased predictors. Chapter 5 provides very strong coverage of genotype x environment interactions. Although volumes have already been written on this topic, this chapter provides a succinct summary of different statistical approaches including stability analysis, biplot analysis, and smoothing spline genotype analysis. Again, what sets this book apart from the rest of the literature is the presentation of actual SAS code for each statistical procedure. The focus of Chapter 6 is the application of repeated measures analysis to growth and lactation curves, and Chapter 7 treats quantitative genetic analysis from a Bayesian perspective. Chapter 8 covers gene frequencies, linkage disequilibrium, and the use of association mapping techniques in natural populations. Chapter 9 describes the use of the extended sib--pair method of mapping quantitative trait loci (QTL) in outbred populations in which a fixed model approach is inappropriate. In Chapter 10 there is extensive coverage of mapping methods from the Bayesian perspective, and Chapter 11 provides a mixed model method for analysis of microarray gene expression data.
Although writing styles vary, the book is generally quite readable. The chapters on Bayesian approaches are not for the faint of heart, but breeders and traditional quantitative geneticists will be very much at home in Chapters 2 through 6. Each chapter presents a sample problem which includes some background theory, SAS code and expected output. The data files and SAS code can be downloaded in .zip format from http://support.sas.com/publishing/bbu/companion_site/ 59454.html; verified 12 July 2005. Many who have taught quantitative genetics courses over the years will look forward to using this book as an excellent software supplement to their courses. Now one can teach diallel analysis using a mixed model approach and have the SAS code at hand to complement the theory with actual data analysis.
There is no doubt that other software exists for each type of data analysis that is discussed in this book. It is remarkable that this set of software applications which cover the waterfront of quantitative genetic analysis can be built on one platform. This begs the question, though, "will the mappers and gene expression researchers really use SAS for their data analysis?" If, as the authors of Chapter 9 suggest, existing QTL mapping software, which employs a fixed model approach, is inadequate for mapping in outbred populations, then the extended sib analysis presented in this book may be just the solution. For most breeders and traditional quantitative geneticists, SAS is already the software of choice, and the applications presented in Chapters 2 to 6 will be reason enough to buy the book.
In summary, this book will be a valuable addition to the library of plant and animal breeders as well as quantitative geneticists. The presentation of SAS applications that can be used for traditional quantitative genetic analysis fills a longstanding need in the literature. The editor recognizes this fact and appeals to readers to send their favorite SAS applications so they can be made available to a wider audience. The utility of the newer applications geared towards molecular data will depend on the software preferences and needs of that user community. SAS is clearly reaching out to that community, but it is too early to predict the success of this effort.
David A. Van Sanford
Department of Plant and Soil Sciences
University of Kentucky
Lexington, KY 40546-0312
|Printer friendly Cite/link Email Feedback|
|Author:||Van Sanford, David A.|
|Article Type:||Book Review|
|Date:||Nov 1, 2005|
|Previous Article:||Crops and Environmental Change: An Introduction to Effects of Global Warming, Increasing Atmospheric CO2 and O3 Concentrations, and Soil Salinization...|
|Next Article:||Measuring the Natural Environment, 2d ed.|