Abstract

Computational Molecular Biology: An Algorithmic Approach By Pavel Pevzner MA: MIT Press (2000). 314 pp. $44.95As we swim in a sea of data—both genomic and microarray—we need good computational tools to understand the biological significance of the information we generate. One such tool that has emerged from the field of computational molecular biology and is used widely by biologists is the sequence comparison tool BLAST. Other tools include multiple sequence alignment software such as ClustalX and contig assembly software such as Phrap. With the draft of the human genome sequence available now and the mouse genome sequence available shortly, we must increasingly turn to the field of computational molecular biology to build additional tools to help us make sense of these data.As an undergraduate mathematics major at the University of Wisconsin in the 1970s, I was totally enthralled with population and quantitative genetics after taking an introductory genetics course from Professor James F. Crow. At the time, this area presented a way to apply my mathematical interest in the exciting field of genetics. Now, the potential areas of research for mathematicians and computer scientists interested in the field of molecular biology are equally rich and much more diverse.The question does arise, however, how best to generate interest and train the next generation of computational biologists? There is a strong need to entice computer scientists into the field of computational biology to solve such problems as, for example, promoter recognition in genomic sequence, analytical tools for understanding data from microarray experiments, and accurate prediction of protein folds from sequences. Many universities and colleges are requiring all undergraduates to take an introductory course in biology. If you teach such a course, Computational Molecular Biology by Pevzner provides a useful high-level introduction to selected computational problems and solutions in molecular biology, which could be useful for those trained in computer science or mathematics who want to become familiar with the problems that interest biologists.Conversely, there is also a need for biologists to understand the theory behind the tools that they use. When doing a BLAST search, a biologist should understand scoring matrices, probability distributions and alignment scores. The web-based forms for access to computational biology tools make it easy to just paste in some data and get back an answer never having to know what algorithm is used and how changes in parameters may affect the results. Unfortunately, Computational Molecular Biology will not help the lab biologist as it is not a cookbook for applied bioinformatics. Biologists are more likely to benefit from more application-oriented books such as Baxevanis and Ouellette, Bioinformatics: A Practical Guide to the Analysis of Genes and Proteins, 1998. (Note: a second edition of this book is scheduled to be published in the spring of 2001.)Computational Molecular Biology is based on a course that Pevzner has taught at the Pennsylvania State University and University of Southern California for a number of years to advanced undergraduate and graduate students in computer science and mathematics. Before reading this book, you would want to have some background in computational algorithms and combinatorial theory. If so, you will see familiar problems and algorithms such as backtracking, Hamiltonian path, and traveling salesman. A little background in molecular biology would also be helpful. There is a brief chapter titled “All You Need to Know about Molecular Biology,” but its first sentence is “Well, not really, of course, see Lewin, 1999 [Genes VII], for an introduction” (p. 271).Each chapter in Computational Molecular Biology begins with an introduction to the computational and biological ideas without any formulas. For example, to introduce the computational problems associated with physical mapping, Pevzner describes the experiments used in the physical mapping of cystic fibrosis. To motivate the problem, he uses an analogy of having “… several copies of a book cut by scissors into thousands of pieces. Each copy is cut in an individual way such that a piece from one copy many overlap a piece from another copy…” (p. 5). This is characteristic of the style of the book. For each computational problem, he gives an analogy that requires no biological knowledge and then describes the biological problem for which a computational solution is required. This is a good style and makes the introductory section of each chapter accessible to biologists interested in learning about some of the computational challenges in the field.Pevzner covers problems drawn primarily from genomics and sequence analysis. Included are chapters on Computational Gene Hunting, Restriction Mapping, Map Assembly, Sequencing, Sequence Comparison, DNA Arrays, Multiple Alignment, Finding Signals in DNA, Gene Prediction, Genome Rearrangement, and Computational Proteomics. Readers interested in structural biology-related topics including such topics as predicting structure from sequence or other topics not covered by Pevzner will need to turn to other books. See for example Bioinformatics: The Machine Learning Approach (Adaptive Computation and Machine Learning) by Pierre Baldi and Soren Brunak, 1998; Algorithms on Strings, Trees, and Sequences: Computer Science and Computational Biology by Dan Gusfield, 1997; or Computational Methods in Molecular Biology edited by Steven Salzberg, David Searls, and Simon Kasif, 1998.Despite the limited breadth of Pevzner's book, I believe Computational Molecular Biology will be, to a limited audience, a useful companion to other computational biology books currently available. Parts of the book read quite well although, given its nature, it gets quite dense in places. This is an exciting time to move into the field of computational molecular biology, motivate others to move into this field, or simply keep an eye on the practical developments emerging from this rapidly evolving field. Reading all or parts of this book can help meet some of these goals.Computational Molecular Biology is the first in a series of books on computational biology published by the MIT Press. Additional titles forthcoming within the next year or two include Computational Modeling of Genetic and Biochemical Networks; Gene Regulation and Metabolism: Post-Genomic Computational Approaches; Current Topics in Computational Biology; Comparative Genomics: The Domains of Life; and The Handbook of Computational Molecular Biology and Bioinformatics. This is a series worth watching for anyone interested in the growing field of computational biology.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call