This review was originally published in SciTech Lawyer, an American Bar Association Publication, in February 2008.
Understanding Bioinformatics
written by Marketa Zvelebil and Jeremy O. Baum
published by Garland Science, 2008
ISBN 0-8153-4024-9 (10 digit) or 978-0-8153-4024-9 (13 digit)
When I volunteered to write a book review in the field of bioinformatics, I couldn't exactly shop at the local bookstore. Being both intimidated and in a big hurry, I scanned Amazon's choices and I chose the one that sounded easiest: Understanding Bioinformatics, a recent paperback written by Marketa Zvelebil and Jeremy O. Baum. The title reminded me of Essentials of Molecular Biology by David Freifelder, the 1985 condensation with pretty pictures for budding biotech patent attorneys whose college papers were typed on an actual typewriter. However, shopping online and taking the easy route is risky; too many karmic variables. It turns out that the book weighs about four pounds; and even though it has plenty of pictures, it also has plenty of calculus.
Even so, for those of you who are fairly up-to-date on this subject, you will find this book comprehensive and current. It is loaded with information, and seems to cater to someone who would sit down at a computer with the book on the edge of her desk, picking through for pointers. Anyone who masters this text can, without cracking even a small smile, consider himself an expert on the subject.
Understanding Bioinformatics is, however, written for a variety of audiences, with each chapter formatted such that a reader can choose how technical to get. For right-brainers, the prose is easy to read and the graphics are great for memory retention. The book ramps up from fairly easy to extremely technical quickly but smoothly, with an increasing need for calculus as the book progresses. I will admit that the later chapters were slow-going. However, the authors are superb communicators and offered the information such that the concepts are attainable, even to someone who graduated in 1991, practiced biotech patent law for ten years, took a detour to make artisan cheese, and recently rejoined the lawyer game. In other words, no matter what your technical background, this book is written in a style that makes the basic topics simple, and the difficult topics relatively easy. Even though the authors mine the depths of bioinformatics, they never get distracted or off-course. This text is probably the only one you would need on the subject, with the next step being journal articles, conferences, or discussions with other bioinformatics enthusiasts.
For left brainers, there is plenty of math in this book. To be honest, I flashed the equations around my office and home, just to prove that my job is really, really hard. People were suitably impressed. I'm not sure that the authors had "fodder for patent lawyers to show off to immigration lawyers" as a goal when writing it, but I can vouch that it is a decent enough reason to buy the book.
The authors assume a college level understanding of molecular biology, evolutionary biology and genomics, but they are kind enough to give gentle reminders if you need a refresher. Happily, interesting pieces of trivia are sprinkled throughout, in separate, somewhat stylish boxes. The glossary is adequate. The index is five pages, but the subjects are printed in a tiny font, making it difficult to read. The small font size in an informatics book made me snicker, but just a bit.
My favorite part of this book is that the authors admit that you are a reader of the book, and that you have expectations and desires. For instance, at the beginning of each chapter, the authors promise that after you read the chapter, you should be able to do and know certain things. Upon reading Chapter 11, you should be able to "discuss the statistical scoring of patterns." With whom, I wonder. But, their unending confidence in my abilities encouraged me to give each new chapter a try.
Another fantastic feature is the "Mind Map," a diagram at the beginning of each chapter, with color-coded ideas that make literal line connections between concepts. I loved those Mind Maps! I also got a kick out of the "Flow Diagrams" that they throw in with the start of new concepts within the chapters. The Flow Diagrams tell you, for instance, "The key concept introduced in this section is that multiple sequence alignments can be produced by a variety of closely related techniques, which are based on the pairwise sequence alignment methods." In other words, if you don't understand the sentence in the Flow Diagram, make sure you read the section, or skip it, if you do.
If you have read this far, you are obviously genuinely interested in bioinformatics, so I will give you the details.
Part 1 is Background Basics. Chapters 1 and 2 cover molecular biology basics, and are nice to have in the book, but most biotech patent lawyers already know this stuff. Every now and then, the authors remind the reader that science is science, and that you should not assume that things are really as you have been taught. For instance, Box 1.2 header screams: "Things are usually not that simple!" and strongly warns the reader not to assume that everyone agrees what the definition of a gene is. Refreshing, really. Chapter 3 covers database basics, and I was able to understand it even though I do not consider myself a computer person. For those of you who are computer people, this is stuff that you would know off the top of your head, and would not bother reading. The authors acknowledge that the first three chapters are warm-ups, and not intended to be the way you learn this information if you are serious about it, and offer further reading suggestions. Plus, they throw in several appendices for good measure, on subjects such as: probability, information, and Bayesian analysis; molecular energy functions; and function optimization.
Part 2 is Sequence Alignments. Chapter 4 discusses producing and analyzing sequence alignments, and covers the basics, in a regimented, but conversational, manner. Biotech patent attorneys should definitely know this information, and the authors deliver it painlessly. Chapters 5 and 6 dig deeper into alignments, including how to do multiple sequence alignments, as well as presentation and critical analysis of the results.
Chapter 7 kicks off Part 3, on Evolutionary Processes, with phylogenetic tree reconstruction and some information on molecular evolution. Along with Chapter 8, which teaches specifically how to build phylogenetic trees, this section strikes me as most often applicable to scholarly endeavors. Even so, I enjoyed it immensely. One reason to read this chapter is to learn how each type of tree is flawed in its own way, so that you stop and consider what you are seeing, for instance, in an invention disclosure.
Part 4 is Genome Characteristics, with Chapter 9 covering gene prediction, promoter and splice site detection, and statistical analysis. Chapter 10 compares various computer programs for gene detection, and provides tips for predicting eukaryotic gene signals and exon/intron structure. The authors once again warn readers against relying too heavily on the technology and concepts of the day, and remind scientists to question their methods and delivery of results. This theme underscores the fluidity of bioinformatics, and, for patent attorneys, the need to know what you are talking about when writing a specification or arguing a motion.
Part 5 deals well with Secondary Structures, including, in Chapter 11, laying out the types of prediction methods, making sure the reader knows the tools available and ways to use them, and following up with statistical analysis and caveats. Prediction of transmembrane protein structure is also reviewed, including a selection of prediction programs for helices, and when to choose which program or technique. This chapter also provides coil and RNA secondary structure prediction tools and considerations. Chapter 12, Predicting Secondary Structure, has a lot of calculus and is very intimidating if you thought you would actually need to prove mathematically that, as Box 12.1 states, "Neural networks must be parameterized by training before use for prediction." If you stick to the text and the figures, however, most of the concepts and vocabulary are attainable to even the oldest and slowest among us. Moreover, if you ever need to know the math details, this is the text you need by your side.
Part 6, Tertiary Structure takes Part 5 to another dimension and complexity. I have to say that anyone who can teach this Part is Nobel Prize material. My brain chugged through it the way my Chevy Vega took to the Bloomington hills, slowly but surely. I mostly credit the authors for their patience and understanding in my conquering the material, but Starbucks certainly contributed. Chapter 13 digs deep into potential energy functions and force fields, obtaining a structure by threading, principles of homology modeling, steps in homology modeling, automated homology modeling, and then, to give you a break, provides a concrete example of the PI3 Kinase p110á. Chapter 14, Analyzing Structure-Function Relationships, an inherently interesting subject to me (no kidding), was no disappointment, although I was surprised and reluctantly happy to find information that I didn't know. There are more programs out there than I previously thought, and the authors describe each with precision and clarity.
Read more ›
