Fragmentation: Toward Accurate Calculations on Complex Molecular Systems introduces the reader to the broad array of fragmentation and embedding methods that are currently available or under development to facilitate accurate calculations on large, complex systems such as proteins, polymers, liquids and nanoparticles. These methods work by subdividing a system into subunits, called fragments or subsystems or domains. Calculations are performed on each fragment and then the results are combined to predict properties for the whole system. Topics covered include: Fragmentation methods Embedding methods Explicitly correlated local electron correlation methods Fragment molecular orbital method Methods for treating large molecules This book is aimed at academic researchers who are interested in computational chemistry, computational biology, computational materials science and related fields, as well as graduate students in these fields.

Answering the need to facilitate quantum-chemical calculations of systems with thousands of atoms, Kazuo Kitaura and his coworkers developed the Fragment Molecular Orbital (FMO) method in 1999. Today, the FMO method can be applied to the study of whole proteins and protein-ligand interactions, and is extremely effective in calculating the propertie

This book covers recent advances of the fragment molecular orbital (FMO) method, consisting of 5 parts and a total of 30 chapters written by FMO experts. The FMO method is a promising way to calculate large-scale molecular systems such as proteins in a quantum mechanical framework. The highly efficient parallelism deserves being considered the principal advantage of FMO calculations. Additionally, the FMO method can be employed as an analysis tool by using the inter-fragment (pairwise) interaction energies, among others, and this feature has been utilized well in biophysical and pharmaceutical chemistry. In recent years, the methodological developments of FMO have been remarkable, and both reliability and applicability have been enhanced, in particular, for non-bio problems. The current trend of the parallel computing facility is of the many-core type, and adaptation to modern computer environments has been explored as well. In this book, a historical review of FMO and comparison to other methods are provided in Part I (two chapters) and major FMO programs (GAMESS-US, ABINIT-MP, PAICS and OpenFMO) are described in Part II (four chapters). dedicated to pharmaceutical activities (twelve chapters). A variety of new applications with methodological breakthroughs are introduced in Part IV (six chapters). Finally, computer and information science-oriented topics including massively parallel computation and machine learning are addressed in Part V (six chapters). Many color figures and illustrations are included. Readers can refer to this book in its entirety as a practical textbook of the FMO method or read only the chapters of greatest interest to them.

The fragment molecular orbital (FMO) method is a fast linear-scaling quantum-mechanical method employed by chemists and physicists all over the world. It provides a wealth of properties of fragments from quantum-chemical calculations, a bottomless treasure pit for data mining and machine learning. However, there is no user-friendly description of its usage in the widely employed quantum-chemical open-source software GAMESS, nor is there any book covering the usage of GAMESS in general. This leaves very many interested users to their own devices to get through a variety of problems with very cryptic descriptions of keywords in the program manual and no guide whatsoever as to what options should be set for particular scientific tasks. This book is the panacea to many frustrations.The main focus of the book is to build a solid bridge connecting FMO users to GAMESS, by giving a helpful introduction of various FMO methods as needed for particular problems found in computational chemistry, and describing in detail how to do these simulations and understand the results from the output of the program. The book also covers parallelization strategies for attaining high parallel efficiency in massively parallel computations, and provides means to analyze performance and design a solution for overcoming performance bottlenecks. A special section is devoted to dealing with problems in executing GAMESS, arising from computational environment and user errors. Finally, 14 carefully selected types of applications are discussed in detail, describing the input keywords and explaining where to find the main results in the text-based output.

The fragment molecular orbital (FMO) method is a fast linear-scaling quantum-mechanical method employed by chemists and physicists all over the world. It provides a wealth of properties of fragments from quantum-chemical calculations, a bottomless treasure pit for data mining and machine learning. However, there is no user-friendly description of its usage in the widely employed quantum-chemical open-source software GAMESS, nor is there any book covering the usage of GAMESS in general. This leaves very many interested users to their own devices to get through a variety of problems with very cryptic descriptions of keywords in the program manual and no guide whatsoever as to what options should be set for particular scientific tasks. This book is the panacea to many frustrations.The main focus of the book is to build a solid bridge connecting FMO users to GAMESS, by giving a helpful introduction of various FMO methods as needed for particular problems found in computational chemistry, and describing in detail how to do these simulations and understand the results from the output of the program. The book also covers parallelization strategies for attaining high parallel efficiency in massively parallel computations, and provides means to analyze performance and design a solution for overcoming performance bottlenecks. A special section is devoted to dealing with problems in executing GAMESS, arising from computational environment and user errors. Finally, 14 carefully selected types of applications are discussed in detail, describing the input keywords and explaining where to find the main results in the text-based output.