Microarray data analysis for dummies…and experts too?

Microarray data analysis for dummies…and experts too?

Forum TRENDS in Biochemical Sciences Vol.27 No.8 August 2002 433 Book Review Throwing light on photosynthesis Molecular mechanisms of photosynthes...

33KB Sizes 0 Downloads 12 Views

Forum

TRENDS in Biochemical Sciences Vol.27 No.8 August 2002

433

Book Review

Throwing light on photosynthesis Molecular mechanisms of photosynthesis by Robert E. Blankenship, Blackwell Science, 2002. US$49.95/£29.95 (352 pages) ISBN 0 632 04321 0

It has long been known that light can be converted into useful chemical energy. The importance of this photosynthetic process has been recognized in different ways throughout history, and was elegantly acknowledged by Julius Robert Mayer in 1845 when he wrote: ‘Nature set herself the task of capturing the light flooding towards the earth and storing this, the most elusive of all forces, by converting it into an immobile force…the plant world constitutes a reservoir in which the fleeting sun rays are fixed and ingeniously stored for future use, a providential measure to which the very existence of the human race is inescapably bound’. In his recently published book, Bob Blankenship replaces Mayer’s flowery definition of photosynthesis with ‘a process in which light energy is captured and stored by an organism and the stored energy is used to drive cellular processes’. Behind this simple and rather bland modern definition of photosynthesis lies a vast base of knowledge, encompassing disciplines ranging from photophysics and photochemistry to molecular biology and physiology. The reactions of photosynthesis are initiated by the absorption of quanta by pigments, which occurs in the femtosecond time domain, followed by reactions that proceed in a time-scale stretching to days or even years. Therefore, to write a comprehensive text book on the molecular mechanisms of photosynthesis is a daunting task. Blankenship has taken up this challenge and, in my view, has produced a first class product. His approach is sensitive to the student reader and, with this in mind, he has included a http://tibs.trends.com

47-page appendix in which the appropriate basic principles of photophysics are well described. The remaining 257 pages comprise 11 chapters, in which Blakenship leads the reader through a logical progression of photosynthesis. The first three chapters set the scene, giving basic concepts in terms of reactions, where they take place and how they were discovered. This foundation is then built upon with a thorough coverage of structural and spectral properties of the photosynthetic pigments, and of the means by which they are associated with proteins. Over the past few years, X-ray crystallography has provided spectacular insights into how these pigment proteins are arranged, and has also given a structural basis to experimental and theoretical considerations of the energy transfer mechanisms. All this is dealt with rigorously by Blankenship. His account of reaction centre complexes in Chapter 6 deserves equal praise. He could not be more up-to-date in his content, providing the reader, for example, with a discussion of the latest X-ray structures of photosystems (PS) I and II. Chapter 7 is a bit of a hodgepodge. Quite rightly, Blankenship moves on to other components of the photosynthetic electron transport chain. He gives details of the similarities of the bacterial cytochrome bc and chloroplast cytochrome b6f complexes, and discusses the role of diffusable redox cofactors such as plastoquinone, plastocyanin, ferredoxin and NADPH. However, the discussion of turnover of the D1 protein of PSII and chlorophyll fluorescence as a probe of PSII function seems, to me, to be misplaced in this chapter. The conversion of ADP to ATP, the role of proton gradients, and Mitchell’s chemiosmotic hypothesis are detailed in a useful and readable way in Chapter 8. In addition, this chapter focuses on speculations about the rotatory ATP synthase mechanism by drawing on the most recent structural information available. It has been traditional to consider the processes of photosynthesis as consisting of two stages: light and dark reactions. Blankenship has continued this tradition and uses Chapter 9 to discuss biochemical details of the dark reactions; namely

carbon fixation via C3, C4 and CAM metabolism. He also includes the CO2 concentrating mechanism found in aquatic photosynthetic systems, and the biochemical reactions involved in longterm storage of photosynthate as sucrose and starch. The photosynthetic apparatus, whether it be of anaerobic bacteria or aerobic cyanobacteria, algae and higher plants, is complicated, being composed of many hundreds of different proteins, enzymes and cofactors. How these complex systems assemble themselves and how they evolved are the subjects of the last two chapters, respectively. Blankenship has a keen interest in the evolution of photosynthesis and his chapter on this subject makes good reading. Although there are major gaps in our knowledge, recent advances in gene sequencing and structural analyses of photosynthetic proteins are providing fodder for fascinating and realistic hypotheses; these provide a grand finale for this excellent text book. I will have no hesitation in recommending it to my students and colleagues alike. James Barber Ernst Chain Professor of Biochemistry, Dept of Biological Sciences, Imperial College of Science Technology & Medicine, London, SW7 2AY, UK e-mail: [email protected] .uk

Microarray data analysis for dummies … and experts too? Microarray Data Analysis and Visualization edited by Arun Jagota Bioinformatics By The Bay Press, 2001. US $29.95 (101 pages) ISBN 097002973X

When I first encountered microarray technology back in 1999, I was fascinated by its revolutionary role in aiding our understanding of complex biological systems and by its immense potential for the post-genomics era. I started to read as many references on this technology as possible and was soon startled by my unfamiliarity of the data analysis described by a few key publications. I was

0968-0004/02/$ – see front matter © 2002 Elsevier Science Ltd. All rights reserved.

434

Forum

totally lost when I tried to understand the exploratory analysis algorithms such as cluster [1] and self-organizing maps (SOMs) [2]. At that time, there was not even a single book about microarrays, not to mention a book dedicated to microarray analysis. Eventually, the first two microarray books, edited by Mark Schena, were published [3,4]. These primarily focused on experimental approaches and on various commercial arraying and scanning platforms. Only a few sections described data analysis and its challenges superficially. As a result, I had to find information bit by bit from the internet, and from publications from other fields. Usually, such sources are not written for a beginner nor in a context respective to the biological problems; this kind of learning is, therefore, not efficient for biologists who are outsiders to data analysis. Microarray Data Analysis and Visualization is the first book dedicated to microarray data analysis [5]. Written by Arun Jagota, the book comprises material originating from his bioinformatics course and is, therefore, concise and straight to the point. The book starts with a brief introductory chapter on microarray data structure and examples of the biological questions that could be answered by such data. This is followed by a chapter discussing how statistics can help our understanding of microarray data. Indeed, the best feature of the book is that every type of data analysis situation is described in a clear statistical context. This is crucial for the readers because one should not blindly use analytical tools in an experiment. On the contrary, to interpret data correctly, one should have a thorough understanding of the inherent statistical ideas and their limitations in analysis. Chapter 3 discusses various data preprocessing techniques such as ratioing, log-transforming and handling missing data. In Chapters 4–6, commonly used data analysis methods such as hierarchical clustering, K-means clustering, self-organizing maps (SOMs) and principle components analysis (PCA), are discussed. Substantial emphasis is put on the various distance metrics and dissimilarity measures for gene expression vectors. Each of these methods has strengths and weaknesses in interpreting the linkage between gene expression vectors, and Jagota describes them explicitly. http://tibs.trends.com

TRENDS in Biochemical Sciences Vol.27 No.8 August 2002

Perhaps the most attractive discussions for biologists are in Chapters 7–12. These chapters categorize different experimental designs or questions of relevance to the planning of experiments. Associated statistical tests are thoroughly discussed. The topics covered include: (1) identifying genes expressed significantly in a population by t-test and non-parametric tests; (2) identifying genes expressed differentially in two populations and paired samples by t-test and non-parametric tests; (3) classifying two samples from two populations by various machine learning methods (including a detailed discussion on Support Vector Machines); and (4) identifying genes expressed differentially in more than two populations by ANOVA. Potential problems and some suggested solutions of multiple hypothesis testing in t-test, and overfitting in machine learning, are mentioned repeatedly during the course of discussion. This is a useful reminder for biologists because we often apply such tests without paying attention to these potential complications. In Chapter 13, gene network inference from microarray data is briefly discussed. Although still in its infancy, I believe the next big impact on functional genomics will come from this gene network inference and from subsequent studies of the dynamic behavior of biological systems. The author has successfully incorporated this pioneering development into his book. Unfortunately, the book contains limited examples, and some statistical ideas are mentioned just briefly in a few sentences or formulae. This makes it difficult for beginners to understand, especially those with no background in data analysis, or those who are weak in mathematics or statistics; unfortunately, this describes many biologists. A few examples of using the methods discussed (e.g. clustering or SOMs) to analyze published microarray data would have been beneficial for everyone to acquire hands-on experience and thus build up their confidence in microarray data analysis. Some of the cutting-edge developments in microarray analysis, such as spatial- and intensity-dependent normalization, and empirical bayes analysis of replicated microarray data, are missing. However, this is acceptable given the facts that the book originates

from class materials and that the analysis field is evolving rapidly. Finally, there are some misaligned figures, although this does not detract from the readability. There is urgent need for an aid that helps outsiders pick up the essence of analysis quickly. An ideal guidebook and reference should share the same philosophy as the ‘For Dummies’ series by John Wiley & Sons: ‘…relate to the anxiety and frustration that people feel about technology by poking fun at it with books that are insightful and educational and make difficult material interesting and easy…’ (http://www.dummies.com/ about/success_story.html). Putting this in the context of microarray analysis, I believe the book should contain: (1) background of the various data analysis methods with clear reference to the experimental designs and statistical ideas involved, their advantages and limitations; (2) vivid examples for applying these methods in biological study, both simple ones to illustrate the idea and real ones quoted from published papers; and (3) a step-by-step guide for applying the respective analysis softwares in the examples. Microarray Data Analysis and Visualization has successfully fulfilled the first requirement and partially addressed the second. It is a handy reference for those who have certain experience. Future expansion of the book to fulfill the remaining requirements would definitely make it helpful for both dummies and experts in microarray data analysis. References 1 Eisen, M.B. et al. (1998) Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sci. U. S. A. 95, 14863–14868 2 Tamayo, P. et al. (1999) Interpreting patterns of gene expression with self-organizing maps: methods and application to hematopoietic differentiation. Proc. Natl. Acad. Sci. U. S. A. 96, 2907–2912 3 Schena, M. (1999) DNA Microarrays: A Practical Approach, Oxford University Press 4 Schena, M. (2000) Microarray Biochip Technology, Eaton Pub. Co., Natick, MA, USA 5 Jagota, A. (2001) Microarray Data Analysis and Visualization, Bioinformatics By The Bay Press (http://bioinformaticsbythebay.hypermart.net/)

Yuk Fai Leung Dept of Ophthalmology & Visual Sciences, The Chinese University of Hong Kong, Hong Kong. e-mail: [email protected]