Over 50 years have passed since thousands of steel balls were poured into balloons to simulate molecular processes. In the 21st century, parallel computing systems perform numerical simulations of molecular dynamics (MD) with high levels of efficacy and accuracy. Rapid advancements in computational methodologies and an increasing interest in MD simulations across interdisciplinary fields of research (such as in silico pharmacology) suggest the need for a reference work that not only provides the current state of the art, but further reveals the physics behind the methods.

This review compares three possible candidate handbooks, in order of their publication dates: Understanding molecular simulation (2nd ed.) (2001), The art of molecular dynamics simulation (2nd ed.) (2004), and Introduction to practice of molecular simulation (2011).

Understanding molecular simulation, by Frenkel and Smit, brilliantly maintains a balance between explaining the physical phenomena and performing computations. Its marvelous writing style invites scientists and students to deepen their knowledge of MD simulations.

The book is composed systematically into four parts, ranging from basic to advanced techniques. The appendices provide background knowledge on the physics and computational mathematics. The book follows a clear strategy: as it introduces the underlying physics of molecular processes, it simultaneously provides detailed discussions of the stochastic and deterministic computational approaches. Both Monte Carlo and MD simulations are introduced, both in general and in the case of various ensembles, which gives the book a methodological comparative trait. In addition, numerous case studies, examples, and exercises help readers deepen their insights, making the book highly appropriate for graduate students.

In my opinion, the authors have achieved the highest standards of scientific writing. This excellent reference book on molecular simulations is appropriate for any interested scientist or graduate student.

The art of molecular dynamics simulation, by Dennis C. Rapaport, follows a slightly different strategy. While the general physics behind the molecular processes is introduced very thoroughly, the author elegantly keeps the reader in constant contact with the real issue of the book, namely, the implementation of physics in the programming language C.

Starting with a brief history of and introduction to MD, Rapaport addresses different cases in separate chapters. Each includes an introduction to the case-related physical phenomena; recipes on how to design numerical algorithms to simulate the processes; and a very welcome and challenging “further study” section. In contrast to the other handbooks, this book does not address Monte Carlo simulations, which is both an advantage and a drawback. By solely focusing on deterministic MD, the author is able to address a large variety of cases very thoroughly and consistently. However, in particular cases, the real power of a computational method can only be appreciated if it is discussed in light of an alternative technique.

I believe that this book is indeed the realization of what its author intended, that is, a recipe book for deterministic MD that will be highly appreciated by experienced practitioners.

Almost a decade after the publication of the above seminal handbooks on MD simulations, Akira Satoh produced Introduction to practice of molecular simulation, which accomplishes a very difficult task: it manages to address the same topic from yet another perspective.

Satoh provides MD-specific physics, but his major focus is the computational implementation and practice of different types of molecular simulations. Several chapters enable readers to improve their MD-related programming skills, after the essentials of the underlying physical phenomena are discussed. The book particularly discusses the dissipative particle dynamics method and the lattice Boltzmann method, which are novel techniques for addressing multi-body hydrodynamics interactions of dispersed particles in a particle suspension or of polymers in polymeric liquids. This well-written handbook provides an overview of the current developments in the practice of MD (both deterministic and stochastic), which makes it a useful reference book for software developers and practitioners of MD simulations.

In summary, one can easily observe a clear tendency of the handbooks in this field to become more specific and more focused on the computational and programming aspects, which is probably a consequence of the great quality of the first two handbooks discussed in this review. In my opinion, the chronological order of these books happens to reflect the various levels of scientific involvement and expertise of their readership in the fields of MD. Students and interested scientists who are looking for a solid reference on MD simulations should start with Frenkel and Smit. Experienced practitioners in MD who are interested in specific applications would benefit strongly from reading Rapaport. Finally, MD software and algorithm developers could benefit from the suggested implementation strategies and up-to-date character of Satoh’s handbook.