ncreasing the affinity of your drug for the target doesn't help you if it also increases its nonspecific binding. Since our knowledge of the therapeutic context of most drugs is sketchy at best, as Ariel Fernandez says in this short but densely written book, this has created a bottleneck in drug design. Fernandez hopes to create a paradigm shift by proposing that we should model cooperativity, which he defines as "the concurrent participation of different regions of the biomolecule to promote and sustain intramolecular or intermolecular interactions." Sounds great, but how? By introducing the concept of "wrapping," where the candidate covers up a desolvation domain or some other part of the microenvironment surrounding the molecule. The goal is to eliminate "dehydrons," which are water-exposed intramolecular hydrogen bonds that weaken the structure. Fernandez says that the energy reduction caused by eliminating dehydrons is instrumental in favoring protein-ligand interactions. This is a many-body problem, but Fernandez says it is computationally tractable.
Fernandez also applies his theory to protein folding. Here he says that partially wrapped H-bonds promote further desolvation, which in turn promotes chain compaction.
The book is nicely published, with outstanding quality color graphs and ribbon structures. May be a bit technical for the casual reader.
his collection of articles by different authors provides an overview of the two basic approaches used in protein engineering today: phage display and computational modeling. The articles fall into two categories. In some articles there is little detail on the method being discussed, and the contributors are more interested in reviewing the literature than explaining the technology. In others, the theory and general principles of the practice are clearly and lucidly explained. None of the articles contain detailed protocols.
Phage display, one of the older techniques, is still the most often used. But it's not a panacea. The best fully random phage display libraries in the world, with 1011 clones, still only code for eight amino acid peptides--barely enough to form a binding motif. This is still useful for optimizing variant Fab fragments or single chain variable fragments (scFvs) specific to one antigen. In such experiments, even though the entire protein is expressed on the surface of the phage or bacterium, only a small portion of the protein of interest is randomized. This crucial point is never mentioned by any author, and it might take a newcomer quite a while to realize this. Indeed, figuring out which residues to mutate and which to leave intact can itself be a challenge.
Newer techniques are ribosome display and mRNA display. These systems are simpler, but their main advantage is that the library is hundreds of times bigger, and it can be re-diversified at each step to allow for an almost unlimited amount of directed evolution. This section starts out with articles on cell-free display systems and principles of library construction and gradually proceeds to more specialized topics. The section on computational approaches gives overviews of protein folding and computational protein design. The average chapter is 21 pages long, giving the authors space to explain their subject in depth.
Contains gray-scale diagrams and graphs, with a bunch of color plates stuck in the middle of the book.
feb 06, 2011
tepwise protocols and tips for making and using antibody phage libraries. Each of the 18 chapters by different authors give an overview followed by detailed procedures.
n outstanding introduction to computational molecular modeling. Has chapters on small molecules, protein, virtual screening and docking, and chemogenomics and rational drug design. Researchers first starting out discover that there is a bewildering array of algorithms and software packages, some of which will not work, some of which will take forever to run, and some of which will just bankrupt your lab. This is the book to read if you want to get up to speed, and more importantly, if you want to know what will and what will not work, and why. Not enough math for someone writing their own software, but has numerous color illustrations and references.
oct 07, 2012
t first glance this book seems like an ideal textbook on molecular modeling. Unlike many other books that are nothing more than a collection of random articles, this is a real book, written by the authors of FluX, a de novo drug design tool. That alone makes it a good way to get started in the field.
The focus is small molecules (drugs) that bind to proteins and the use of structure-activity relationships and computer modeling in drug discovery. Chapters progress from basic concepts to computer modeling algorithms. They give an overview of major computer tools and their underlying principles. But the level of detail is not nearly enough for a reader to jump up and start discovering drugs. There's not enough detail to give the reader a solid understanding of how the algorithms work. It's not even enough to start selecting the most appropriate software. One reason is that there's too much emphasis on programs of historical importance that are no longer in actual use, and not enough on teaching the current state of the art.
It is nicely published, with color illustrations, color molecular diagrams, and text boxes, but my copy seems to have become oxidized somehow: only three years after publication the paper is already starting to turn brown around the edges. The writing style is somewhat dry. The latest reference dates from 2007. A background in biochemistry or medicinal chemistry is recommended.
feb 22, 2011
he editors of this two-volume, 714-page multi-author treatise are the creators of a software package called PROSPECT for calculating protein folding. However, the focus of this book is not just folding, but every aspect of protein modeling, including protein-ligand docking, structure-based drug design, structure prediction, and the pros and cons of different approaches to modeling molecular dynamics. The contributions are uniformly well-written. Different software packages and the mathematical theories behind them are well described. The coverage is of sufficient detail and accuracy that beginners and advanced practitioners will all find it useful. I keep a copy of this one on my desk next to my copy of Molecular Modeling by Hans-Dieter Höltje, Wolfgang Sippl, Didier Rognan and Gerd Folkers. That's how good it is.
ost of the 83 pages in this book are taken up by full-page Ramachandran plots and poor quality grayscale images of molecules. Nothing about the software; in fact, very little information about molecular modeling.