Customer Reviews

Methods for Computational Gene Prediction

5 star:		(7)
4 star:		(0)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

If you are interested in bioinformatics and the art of gene prediction or genome annotation, you MUST buy this book!! Methods for Computational Gene Prediction was written with both molecular biologists and computer scientists in mind. Although those with training in math and statistics will find some of the material easier to grasp, the book starts out with both a math primer and background on molecular biology to bring both target audiences up to speed. The author, Bill Majoros, did a fantastic job at describing gene prediction methods and walks the reader through examples by providing numerous illustrations and blocks of pseudocode. My personal favorite is the chapter on the `Toy Exon Finder', which really sets the stage for studying the art of gene prediction. The book is structured so that it can be used as a textbook for a course in bioinformatics, with problem sets provided at the end of each chapter. Although there are many books that focus on topics in bioinformatics, this is most definitely one of my favorites, alongside Durbin and Eddy's Biological Sequence Analysis, another essential text for anyone in this field.

This is a fantastic reference on a complex, yet fundamental, problem in genomics. Computational gene prediction is a foundational aspect of modern genomics, yet it is hard to find a one-stop-shop to learn what it is all about. This book provides that all-in-one resource, covering the general biological, mathematical, and machine learning backgrounds alongside the detailed and clear explanations of gene finding. The book also includes many helpful examples, with pseudocode, making it ideal for a class textbook. After working through the examples I came away not only with a clear understanding, but with my own functioning gene finder! This is an excellent book that provides the kind of detail not found in the journals, along with lots of tips and tricks accumulated by the author that can't be found elsewhere. I highly recommend it.

I wanted to learn about implementing HMMs to a particular problem and this book was extremely useful for that. The explanations, though mathematical are simple and straightforward. It is an easy read and provides a great reference for various bioinformatic tools.

Well-written and at the right level - not so simple that you are bored and not so complex that you cannot follow. Enjoyable to anyone interested in science.

I recently sat down to tackle a problem in bioinformatics that was just begging for a hidden markov model with Baum Welch to estimate the emissions and transition probabilities. I had two books at my fingertips: this one and Durbin. I have to say, I found this the more helpful of the two. The explanations are on par with each other but this book takes a slightly more computationally focused approach and provides very clear pseudo-code for all of its more complicated algorithms (Baum Welch, forward, backward, viterbi, etc.). I found it indispensable as a computational reference.

I don't want to give the impression that all this book is good for is hidden markov models, either, though a hefty portion of the book is dedicated to them. It also contains reviews of probability theory, statistics and various machine learning techniques aside from HMMs that are equally clear. I'm not saying that this replaces Durbin, since the focus isn't the same, but it's a great bioinformatics book nonetheless and has won a permanent spot on my shelf.

"Methods for Computational Gene Prediction" is a comprehensive tome
covering different methods for gene prediction from the basics to
state-of-the-art models that use more exotic approaches. The
information is meticulously organized so that new ideas are introduced
in a logical and almost transparent manner. Thus, it is easy to read,
and quickly provides the reader with an expert understanding of a
complex topic.

This book is appropriate for anyone interested in understanding how
the A, C, G, T letters comprising the genome is decoded into genes,
the first step in decoding the book of life. For people studying
bioinformatics, it should definitely be on your reading list. No
background in computer science or biology is required, although a
familiarity of math at the high school level is needed.

A bonus: the book also provides clear and concise descriptions of
relevant numerical and machine learning methods, e.g. basic
probability, information theory, bayesian networks, etc.

This book is an excellent overview of gene finding techniques. The author takes the reader from simple models to the most current techniques. Throughout the book, the author keeps track of model predictive accuracy to help the reader understand the real improvements of using more advanced methods.

The book also has a website at [...] which contains source code, data sets from the book, course materials, and additional chapters.