GENESIS User Guide

Release 1.0

RIKEN AICS

2015/05/08

GENESIS release 1.0

Project Leader: Yuji Sugita (RIKEN)
Main Developers: Jaewoon Jung (RIKEN), Takaharu Mori (RIKEN)
Developers: Chigusa Kobayashi (RIKEN), Yasuhiro Matsunaga (RIKEN), Takashi Imai
(RIKEN), Takao Yoda (Nagahama Institute of Bio-Science and Technology)
Contributors: Naoyuki Miyashita (RIKEN), Ryuhei Harada (RIKEN), Wataru Nishima
(RIKEN), Raimondas Galvelis (RIKEN), Yasuaki Komuro (RIKEN)
Acknowledgments: Norio Takase (Isogo Soft), Hikaru Inoue (Fujitsu), Tomoyuki Noda
(Fujitsu)
Copyright 2014 RIKEN. All Rights Reserved

Citation Information
J. Jung, T. Mori, C. Kobayashi, Y. Matsunaga, T. Yoda, M. Feig, and Y. Sugita, "GENESIS: A hybridparallel and multi-scale molecular dynamics simulator with enhanced sampling algorithms for biomolecular and cellular simulations", WIREs Computational Molecular Science (DOI: 10.1002/wcms.1220)

Copyright Notice
GENESIS is distributed under the GNU General Public License version 2.
Copyright 2014 RIKEN.
GENESIS is free software; you can redistribute it and/or modify it under the terms of the
GNU General Public License as published by the Free Software Foundation; either version
2 of the License, or (at your option) any later version.
GENESIS is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with GENESIS
see the file COPYING. If not, write to the Free Software Foundation, Inc., 59 Temple
Place - Suite 330, Boston, MA 02111-1307, USA.
It should be mentioned this package contains the following softwares for convenience. Please note that
these are not covered by the license under which a copy of GENESIS is licensed to you, while neither
composition nor distribution of any derivative work of GENESIS with these software violates the terms
of each license, provided that it meets every condition of the respective licenses.

Mersenne Twister: A random number generator

A Mersenne Twister random number generator was originally written in C by Makoto Matsumoto and
Takuji Nishimura, and later translated into Fortran by Hiroshi Takano and Richard Woloshyn. This
routine is distributed under the GNU General Public License version 2.
Copyright 1997 Makoto Matsumoto and Takuji Nishimura.
Copyright 1999 Hiroshi Takano.
Copyright 1999 Richard Woloshyn.

This library is free software; you can redistribute it and/or modify it under the terms of the
GNU Library General Public License as published by the Free Software Foundation; either
version 2 of the License, or (at your option) any later version.
This library is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR
A PARTICULAR PURPOSE. See the GNU Library General Public License for more details. You should have received a copy of the GNU Library General Public License along
with this library; if not, write to the Free Foundation, Inc., 59 Temple Place, Suite 330,
Boston, MA 02111-1307 USA

FFTE: A Fast Fourier Transform Package

FFTE (http://www.ffte.jp/) is written by Daisuke Takahashi (Tsukuba University).
Copyright 2000-2004, 2008-2011 Daisuke Takahashi (Tsukuba University).
You may use, copy, modify this code for any purpose (include commercial use) and without
fee. You may distribute this ORIGINAL package.

Complementary error function: erfc04

A Complementary error function routine (erfc04) is written by SunSoft, a Sun Microsystems, Inc. business.
Copyright 1993 Sun Microsystems, Inc.
Developed at SunSoft, a Sun Microsystems, Inc. business. Permission to use, copy, modify,
and distribute this software is freely granted, provided that this notice is preserved (see
at_energy_pme.fpp, and sp_energy_pme.fpp).

Note:
"GENESIS (General Neuron Simulation System)" (http://www.genesis-sim.org/) simulator, which has
much longer history in the development.
[The Book of GENESIS]
http://www.amazon.com/
The-Book-GENESIS-Exploring-SImulation/dp/0387949380/
ref=sr_1_1?ie=UTF8&qid=1394425468&sr=8-1&keywords=bower+beeman
Our "GENESIS" has been developed independently to investigate conformational dynamics of proteins,
nucleic acids, biological membranes, and other biomolecules.

CONTENTS

Introduction

Getting Started
2.1 How to install GENESIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.2 Basic Usages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3
3
7

Available Programs
13
3.1 Simulators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2 Analysis tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3 Parallel I/O tools . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

Input and Output files

15
4.1 Input files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.2 Output files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

Energy
5.1 General keywords . . . . . . . . . . . .
5.2 Non-bonded interaction related keywords
5.3 Particle mesh Ewald related keywords . .
5.4 Lookup table related keywords . . . . . .

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

18
18
19
20
21

Dynamics
23
6.1 General keywords . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
6.2 Simulated annealing related keywords . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Minimize

25

Constraints

26

Ensemble

27

10 Boundary

29

11 Selection

30

12 Restraints

32

13 REMD

34

14 Tutorial 1: Building and Simulating BPTI in Water

37

ii

14.1
14.2
14.3
14.4
14.5
14.6

Building a simulation system . . . . . . .

Minimization with restraints on the protein
Heat-up with restraints on the protein . . .
Equilibration simulation . . . . . . . . . .
Production simulation . . . . . . . . . . .
Analysis: RMSD calculation . . . . . . . .

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

37
41
43
45
46
47

15 Tutorial 2: Simulating with Go-model

49
15.1 Building a simulation system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
15.2 Production simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
15.3 Analysis: RMSD calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
16 Tutorial 3: REMD Simulation of Alanine Dipeptide
16.1 Minimization . . . . . . . . . . . . . . . . . . .
16.2 Equilibration . . . . . . . . . . . . . . . . . . .
16.3 Replica temperatures . . . . . . . . . . . . . . .
16.4 Production simulation . . . . . . . . . . . . . .
16.5 Analysis of REMD simulation . . . . . . . . . .

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

56
56
58
61
61
64

17 Tutorial 4: REUS Simulation of Alanine Dipeptide

71
17.1 Replica temperatures and umbrella potentials . . . . . . . . . . . . . . . . . . . . . . . 71
17.2 Production simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
17.3 Analysis of REUS simulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
18 Tutorial 5: Parallel Input/Output Scheme
18.1 Minimization with restraints on the protein
18.2 Heat-up with restraints on the protein . . .
18.3 Equilibration . . . . . . . . . . . . . . . .
18.4 Production simulation . . . . . . . . . . .
18.5 Analysis: RMSD calculation . . . . . . . .
Bibliography

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

81
81
85
88
90
91
93

iii

iv

CHAPTER

ONE

INTRODUCTION
GENESIS (Generalized-Ensemble Simulation System) is a suite of computer programs for carrying out
molecular dynamics (MD) simulations of biomolecular systems. MD simulations of biomolecules such
as proteins, nucleic acids, lipid bilayers, N-glycans, and so on, are used as important research tools
in structural and molecular biology. Many useful MD simulation packages [1] [2] [3] [4] [5] are now
available together with accurate molecular force field parameter sets [6] [7] [8] [9] [10]. Most of the MD
software have been optimized and parallelized for distributed-memory parallel supercomputers or PCclusters. Therefore we can use hundreds of CPUs or CPU cores efficiently for a single MD simulation
of a relatively large biomolecular system, typically composed of several hundreds thousand of atoms. In
recent years, the number of available CPUs or CPU cores is rapidly increasing, and thereby, modern and
more efficient parallel schemes are necessary to be implemented in the MD simulation programs.
One of our major motivations is to develop MD simulation software whose performance is scalable on
such modern supercomputers. For this purpose, we have developed the software from scratch, introducing the hybrid (MPI + OpenMP) parallelism and several new parallel algorithms [11] [12]. Another
major motivation is to develop a MD simulation code, which can be easily understood and modified
for methodological development. The two policies (high parallel performance and simplicity) usually
conflict with each other in computer software. Therefore we considered to develop two MD simulators
simultaneously.
They are SPDYN (Spatical decomposition dynamics) and ATDYN (Atomic decomposition dynamics).
Although these two MD codes share almost the same data structures, subroutines, and modules, a different parallelization scheme is introduced in each simulator. In SPDYN, the spatial decomposition
scheme is implemented with new parallel algorithms [11] [12]. Its performance is therefore better than
ATDYN and most of other previously developed MD simulators. In ATDYN, the atomic decomposition
scheme is introduced aiming for simplicity in the source code; and enhanced conformational sampling
algorithms such as the replica-exchange molecular dynamics (REMD) method is available.
Due to the simple parallelization, the performance of ATDYN is worse than SPDYN. However, ATDYN
is simpler and, thereby, easier to modify for development of new algorithms or novel molecular models.
We hope, ambitious users will try to develop new methodologies in ATDYN at first and, eventually,
move to SPDYN for the better performance. As we try to maintain consistency between the source
codes of ATDYN and SPDYN, switching from ATDYN to SPDYN is possible by ambitious users.
Other features in GENESIS are listed as follows:
Not only atomistic molecular force field (CHARMM) but also Go-model is available in ATDYN.
For extremely large biomolecular systems (more than 10 million atoms), parallel input/output
(I/O) scheme is implemented and available.
GENESIS is optimized for K computer (developed by RIKEN and Fujitsu company), but it is
available on Intel-based supercomputers or PC-clusters.
1

GENESIS User Guide, Release 1.0

GENESIS is written in modern Fortran 90/95/2003 using modules and dynamic memory allocation. No common blocks are used!
GENESIS is free software under the GNU General Public License (GPL) version 2 or later. We
allow any uses to use/modify GENESIS and redistribute the modified version under the same
license.
And so on.
This manual consists of 18 chapters including information how to get started, explanation of each keyword in control files, and tutorials for MD simulations, coarse-grained (CG) MD simulations with Gomodel, REMD simulations, and so on. We recommend new users of GENESIS to start from the next
chapter, Getting Started, to learn a general idea, installation, and work flow of the program.
Compared to other MD software like AMBER, CHARMM, NAMD, and so on, GENESIS is a very
young MD simulator. Before releasing the program, the developers and contributors in GENESIS development team worked hard to kill all bugs in the program, and performed a bunch of test simulations.
Still, there is possibility of several defects or minor bugs in GENESIS. Since we cannot bear any responsibility for the simulation results produced by GENESIS, we recommend new users to check their
results carefully, if necessary, by comparing with other MD programs.
We, GENESIS development team, have a lot of plans for future development of methodology and
molecular models. Some of them are already on-going projects by us. We would like to grow GENESIS
toward one of the most powerful and feasible MD software package, contributing to computational
chemistry and biophysics. We believe that the current status of computational studies in life science
area is still in the very early stage (like GENESIS) compared to established experimental researches.
We hope, GENESIS can push forward the computational science and contribute to life-scientific and
medical applications in the near future.

CHAPTER

TWO

GETTING STARTED
GENESIS consists of two simulators and several analysis tools. The simulators, called ATDYN
and SPDYN, can perform minimization, molecular dynamics, and other advanced simulations of
biomolecules. The analysis tools, trj_analysis, crd_convert, pcrd_convert, remd_convert, are used
for post-processing trajectories produced by the simulators. prst_setup generates GENESIS original
multiple restart I/O data from the conventional CHARMM input data.
A description of each program is given in the next chapter (Available Programs), and a detailed usage
(including references for input parameters) is explained in Tutorials (from Chapter 14). This chapter
is devoted to orient new users who have just downloaded GENESIS package. In the first half of this
chapter, compilation and installation of GENESIS is described. In the later half, we give the users a
general idea of how to use GENESIS for their own purposes.

2.1 How to install GENESIS

2.1.1 Requirements
The source code of GENESIS is written in Fortran 90/95/2003 and organized into modules. The preprocessing directives (such as #ifdef) are used to adapt the source code for different architectures and
options. Fortran compiler and its preprocessor is the minimum requirement to compile GENESIS.
The simulators (ATDYN and SPDYN) of GENESIS are highly parallelized in their own decomposition
schemes (atomic decomposition and spatial decomposition, respectively). These schemes are implemented by using both of de facto standard parallel programming models: MPI (distributed-memory)
and OpenMP (shared-memory). In a nutshell, MPI is used for the communication between the different machines (nodes) or processes, while OpenMP is used within a single machine (node) or process.
Whereas OpenMP is supported by many of modern Fortran compilers, MPI needs to be installed in your
environment if necessary. In particular, SPDYN is designed for large-scale parallel application.
Although GENESIS is tested on various platforms, at present, validated combinations of CPU and
compiler are limited to those provided by Intel and Fujitsu. Thus, the combinations of Intel CPU and
compiler, or SPARC CPU and Fujitsu compiler is recommended. As for the MPI library, we have
thoroughly tested with OpenMPI [13], which is another recommendation.
The recommendations for compiling GENESIS are summarized below. Please make sure that at least
one of them in each section is installed on your system.
Fortran compiler
Intel compiler ifort (Recommended)

GENESIS User Guide, Release 1.0

Fujitsu compiler frtpx (Recommended)

GCC compiler gfortran (4.4.7 or higher version required; some features do not work)
Preprocessor
fpp supplied with Intel compiler (Recommended)
Fujitsu compiler frtpx supports preprocessing (Recommended)
GCC preprocessor cpp
MPI implementation
OpenMPI (Recommended)
Fujitsu MPI (Recommended)
Intel MPI (Recommended)
MPICH (Tested, but not recommended)
Operating system
Linux (Recommended)
Mac OSX (Not recommended, but may work)

2.1.2 Download
GENESIS package is available at http://www.riken.jp/TMS2012/cbp/en/research/software/genesis/index.html
The files are organized according to their purpose.
User Guide (genesis.pdf): this file
Source code (genesis.tgz): source code for the simulators and the tools
Test files (test.tgz): script and input files for regression tests (see below)
Tutorials (tutorial.tgz): input files for tutorials, see Tutorials (from Chapter 14)

2.1.3 Installation
First, extract the archive of the source code (genesis.tgz).
# untar the package file and change the working directory
$ tar xvfz genesis.tgz
$ cd genesis/
$ ls -lF
total 8
-rw-r--r-1 user staff
79 9 12 22:27 README # README file
-rw-r--r-1 user staff
79 9 12 22:27 COPYING # License agreement
drwxr-xr-x 10 user staff 340 10 20 18:16 src/
# Source codes

GENESIS uses GNU autotools build system. Change the current directory to src/, and run
./configure script to create a Makefile for your system. ./configure script tries to detect all
of the requirements needed to compile GENESIS:

2.1. How to install GENESIS

GENESIS User Guide, Release 1.0

# change the working directory to src/ and invoke ./configure

$ cd src/
$ ./configure

It may take a while ./configure script to complete. While running, it prints messages telling which
things are checked.
If you need to add compilation options and/or library paths, you can set the following options:
options
prefix=PREFIX
exec-prefix=EPREFIX
program-prefix=PREFIX
program-suffix=SUFFIX
enable-debug=LEVEL
disable-mpi
disable-openmp
enable-fft3d
with-fftw[=PATH]
with-lapack
host=host
FC
FCFLAGS
LAPACK_PATH
LAPACK_LIBS

meaning
install architecture-independent files in PREFIX
install architecture-dependent files in PREFIX
prepend PREFIX to installed program names
append SUFFIX to installed program names
enable debugging (time consuming)
disable MPI parallelization
disable OpenMP parallelization
enable FFT3D calculation (see [ENERGY] section)
use FFTW library instead of embedded FFTE routine
use LAPACK
disable compiler check (Required in cross compiler system)
Fortran compiler command
Fortran compiler flags
LAPACK library path (recommended when given with-lapack)
LAPACK library path (recommended when given with-lapack)

# example of high performance options on non-cross-complation system

$ ./configure
# example of enable "LAPACK"
$ ./configure --with-lapack
# example of high performance options on K computer
$ ./configure --host=k
# example of high performance options with FFT3D option on K computer
$ ./configure --enable-fft3d --host=k
#
#
#
#
#
$

example of debugging mode with higher level (time consuming)

enable-debug=LEVEL: 1 = without optimization
2 = with debug information (-g) & -DDEBUG
3 = LEVEL=2 & memory check
if LEVEL is not defined explicitly, LEVEL is 1.
./configure --enable-debug=3

# example of disabled parallelization (serial)

$ ./configure --disable-mpi --disable-openmp
# example of cross-compilation options (i.e. many computer centers)
$ ./configure --host=host

Once ./configure successfully finished, Makefile is created for your system. To compile and
install GENESIS, type make install command:

2.1. How to install GENESIS

GENESIS User Guide, Release 1.0

$ make install

make install compiles all of the GENESIS programs (the simulators and the analysis tools). The
compiled binary files are copied to bin/ directory. If the compilations are successfully finished, the
following binary files has to be in your bin/:
$ ls -lF ../bin/
total 57840
-rwxr-xr-x 1 user
-rwxr-xr-x 1 user
-rwxr-xr-x 1 user
-rwxr-xr-x 1 user
-rwxr-xr-x 1 user
-rwxr-xr-x 1 user
-rwxr-xr-x 1 user
-rwxr-xr-x 1 user

staff
staff
staff
staff
staff
staff
staff
staff

6271072
1804980
1911360
6575348
1946552
1180328
7259252
1472264

10
10
10
10
10
10
10
10

20
20
20
20
20
20
20
20

18:19
18:24
18:24
18:23
18:24
18:24
18:23
18:24

atdyn
crd_convert
pcrd_convert
prst_setup
remd_convert
rst_convert
spdyn
trj_analysis

2.1.4 Regression Tests

Regression tests are prepared for ATDYN, SPDYN and prst_setup (parallel I/O) to check if these
programs work correctly.
# untar the package file and change
$ tar xvhz tests.tar.bz2
$ cd tests/regression_test
$ ls -lF
total 88
drwxrwxr-x 5 user staff 4096 11 29
drwxrwxr-x 3 user staff 4096 11 29
drwxrwxr-x 5 user staff 4096 11 29
drwxrwxr-x 5 user staff 4096 11 29
-rwxrwxr-x 1 user staff 8655 11 29
-rwxrwxr-x 1 user staff 8514 11 29
-rwxrwxr-x 1 user staff 29434 11 29
drwxrwxr-x 5 user staff 4096 11 29
drwxrwxr-x 2 user staff 4096 11 29

the working directory

10:29
10:29
10:29
10:29
10:29
10:29
10:29
10:29
10:32

test_spdyn
test_parallel_IO
test_common
test_atdyn
test.py
genesis.py
charmm.py
build
param

The tests can be run by executing a Python script (test.py):

$ ./test.py *genesis_command* *[parallel_io (optional)]*

Here, genesis_command is a command line for executing ATDYN or SPDYN (default: mpirun -np
8 atdyn). parallel_io needs to be appended to check the parallel I/O facility.
# execute atdyn test
$ ./test.py "mpirun -np 8 /path/to/atdyn"
# execute spdyn test
$ ./test.py "mpirun -np 8 /path/to/spdyn"
# execute spdyn test on FUJITSU compiler
$ ./test.py "mpirun -np 8 /path/to/spdyn" 0.03

2.1. How to install GENESIS

GENESIS User Guide, Release 1.0

# execute spdyn, parallel_io test

$ ./test.py "mpirun -np 8 /path/to/spdyn" parallel_io

Note: The regression tests are designed to be executed with 8 MPI processes. In particular, the parallel
I/O test will be failed with other MPI conditions. As for ATDYN and SPDYN tests, other MPI conditions
may give invalid results.

Note: Since the optimization scheme of FUJITSU compiler is different from those of other compilers,
tolerance of spdyn test should be increased in FUJITSU compiler.

Note: prst_setup does not work using the Fujitsu compiler. There is no problem in running SPDYN
by reading parallel restart files generated by prst_setup with other compilers.

2.2 Basic Usages

GENESIS programs use a simple command line style, where a program always interprets the first argument as an input file. The input file, control file, describes the parameters for controlling the program.
A typical usage of GENESIS program is as follow:
# serial execution
$ *program_name* *control_file*
# parallel execution with 8 MPI processes
$ mpirun -np 8 *program_name* *control_file*

For example, SPDYN can be called like this:

$ mpirun -np 8 /path/to/spdyn ctrl.inp

GENESIS program prints a templates of the control file then executed with -h ctrl template_name
option. A list of template_name names can obtained by running it with -h option only. For example,
SPDYN prints the template names for md (molecular dynamics), min (minimization), and remd (replica
exchange molecular dynamics):
$ spdyn -h
# normal usage
% ./spdyn INP
# check control parameters of md
% ./spdyn -h ctrl md
# check control parameters of remd
% ./spdyn -h ctrl remd

2.2. Basic Usages

GENESIS User Guide, Release 1.0

# check control parameters of min

% ./spdyn -h ctrl min
(skipped..)

For example, the template for minimization is shown by issuing the following command:
$ spdyn
[INPUT]
topfile
parfile
psffile
pdbfile

-h ctrl min
=
=
=
=

sample.top
sample.par
sample.psf
sample.pdb

[ENERGY]
forcefield
electrostatic
switchdist
(skipped...)

= CHARMM
= PME
= 10.0

#
#
#
#

topology file
parameter file
protein structure file
PDB file

# [CUTOFF,PME]
# switch distance

If you are interested in all of available options including detailed parameters for advanced users, add
_all to template_name. For example, running with -h ctrl min_all prints all of available options for minimization.

2.2.1 Information Flow in GENESIS

In general, simulations of biomolecules require several types of input files in addition to the control file.
For example, files for atomic coordinates, force field parameters, and molecular topology (connectivity
of molecules) are necessary.
The coordinates can be given by PDB, CRD, or GENESIS restart files (pdbfile, crdfile, and rstfile,
see Input and Output files). As the current version of GENESIS supports CHARMM force fields only,
CHARMM parameter and topology files (parfile and topfile, see Input and Output files) are necessary
along with PSF (psffile, see Input and Output files).
GENESIS does not include any programs for generating these input files. However, the setup of a
simulation system could be done by using CHARMM [14] or psfgen [15] (supplied with NAMD [16]
and VMD [17]). Setup procedures are demonstrated in the Tutorial chapters. The information flow of
input files in GENESIS are summarized in a chart below.

2.2. Basic Usages

GENESIS User Guide, Release 1.0

2.2.2 Control File

The input files and control parameters of GENESIS are specified in a control file. The control file
consists of several sections (i.e. [INPUT], [OUTPUT], [ENERGY], etc.), with each section containing
a set of keywords.
Note: Each parameter should be less than MaxLine (Default : 1000) characters. MaxLine is defined in
src/lib/string.fpp. If a statement is too long, it can be continued with a backslash at the end of the
line.
Examples of the sections and their important keywords are given below:
[INPUT] section: input files (see Input and Output files)
topfile: topology file
parfile: parameter file
psffile: protein structure file
pdbfile: PDB file
[OUTPUT] section: output files (see Input and Output files)
dcdfile: trajectory file in DCD format
rstfile: restart file
[ENERGY] section: energy and force evaluations (see Energy)
forcefield: type of force field (CHARMM/KBGO)
electrostatic: type of long-range electrostatic interactions (CUTOFF/PME)
switchdist: switch distance (default: 10.0)
cutoffdist: energy/force cutoff distance (default: 12.0)
pairlistdist: cutoff distance for Verlet pair-list (default: 13.5)
2.2. Basic Usages

GENESIS User Guide, Release 1.0

[DYNAMICS] section: integrator and timestep (see Dynamics)

integrator: integrator (LEAP/VVER)
nsteps: number of MD steps (default: 100)
timestep: time interval of MD step (default: 0.001)
nbupdate_period: interval (in MD steps) of a non-bonded pair-list regeneration (default:
10)
[MINIMIZE] section: minimization (see Minimize)
method: minimization algorithm (SD)
nsteps: number of minimization steps (default: 100)
nbupdate_period: interval (in minimization steps) of a non-bonded pair-list regeneration
(default: 10)
[CONSTRAINTS] section: constraints (see Constraints)
rigid_bond: enable constraints (YES/NO)
[ENSEMBLE] section: temperature and pressure controls (see Ensemble)
ensemble: type of ensemble (NVE / NVT / NPT)
tpcontrol: type of thermostat and barostat (NO / Berendsen / Langevin)
temperature: initial and target temperature (default: 298.15)
pressure: target pressure (default: 1.0)
[BOUNDARY] section: system size and boundary condition (see Boundary)
type: type of boundary (default: PBC)

2.2.3 Minimization
A control file for minimization of a molecule is shown below. [INPUT] section contains the input file
names, and the parameters of [ENERGY] specify energy and force evaluation. [MINIMIZE] section
enables the minimization algorithm.
[INPUT]
topfile
parfile
psffile
pdbfile

=
=
=
=

top_all27_prot_lipid.top
top_all27_prot_lipid.par
mol.psf
mol.pdb

[OUTPUT]
dcdfile = min.dcd
rstfile = min.rst
[ENERGY]
forcefield
electrostatic
switchdist
cutoffdist
pairlistdist

2.2. Basic Usages

=
=
=
=
=

#
#
#
#

topology file
parameter file
protein structure file
PDB file

# DCD trajectory file

# GENESIS restart file

CHARMM
PME
10.0
12.0
13.0

#
#
#
#
#

[CHARMM,KBGO]
[CUTOFF,PME]
switch distance
cutoff distance
pairlist distance

10

GENESIS User Guide, Release 1.0

[MINIMIZE]
nsteps
eneout_period
crdout_period
rstout_period

=
=
=
=

[BOUNDARY]
type

= PBC

100
10
10
100

#
#
#
#

number of steps
energy output period
coordinates output period
restart output period

# [NOBC,PBC]

2.2.4 Molecular Dynamics

A control file for molecular dynamics of a molecule is shown below. In this case, instead of [MINIMIZE] section, [DYNAMICS] section enables the molecular dynamics algorithms. ATDYN and SPDYN give identical results for the same control file except a few cases (see the next sections for details).
For small simulation systems (total number of atoms including solvent is less than 1000), ATDYN is
recommended. In contrary, for large systems, SPDYN is faster than ATDYN as long as your machine
has many cores or multiple nodes.
[INPUT]
topfile
parfile
psffile
pdbfile
rstfile

=
=
=
=
=

top_all27_prot_lipid.top
top_all27_prot_lipid.par
mol.psf
mol.pdb
min.rst

[OUTPUT]
dcdfile = md.dcd
rstfile = md.rst

#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
restart file

# DCD trajectory file

# restart file

[ENERGY]
forcefield
electrostatic
switchdist
cutoffdist
pairlistdist

=
=
=
=
=

CHARMM
PME
8.0
12.0
13.0

#
#
#
#
#

[CHARMM,KBGO]
[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period

=
=
=
=
=
=

LEAP
1000
0.002
10
10
1000

#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
timestep (ps)
energy output period
coordinates output period
restart output period

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVE
= NO
= 300

# [NVE,NVT,NPT]
# no thermostat
# initial temperature

[BOUNDARY]

2.2. Basic Usages

11

GENESIS User Guide, Release 1.0

type

2.2. Basic Usages

= PBC

# [NOBC,PBC]

12

CHAPTER

THREE

AVAILABLE PROGRAMS
GENESIS consists of two simulators with different algorithms and several analysis and conversion tools
for trajectories.

3.1 Simulators
GENESIS has two simulators with different decomposition schemes, but Input/Output files adhering
to the common formats (except for the parallel I/O scheme). However, some of simulation options are
ATDYN or SPDYN specific.
ATDYN (ATomic decomposition DYNamics simulator)
Molecular dynamics, energy minimization, and replica exchange molecular dynamics
(REMD) [18] [19] are available. The simulator uses the atomic decomposition scheme
with hybrid (MPI/OpenMP) parallelization. The simulator is designed for an easy prototyping and implementation of new methods (force fields, generalized-ensemble, external force,
coarse-grained models, etc.).
SPDYN (SPatial decomposition DYNamics simulator)
Molecular dynamics, energy minimization, and replica exchange molecular dynamics
(REMD) are available. The simulator uses the spatial decomposition scheme with our new
algorithms for parallel scaling; and parallel I/O scheme is available. The simulator is more
complex and designed for high performance and good scalability on massively parallel computers.

3.2 Analysis tools

In addition GENESIS includes several tools to analyze trajectories, and to convert trajectories/restart
files to intended formats. The usage of the tools is introduced in the tutorial chapters. (Tutorial 1:
Building and Simulating BPTI in Water, Tutorial 2: Simulating with Go-model.)
trj_analysis
An utility to analyze trajectory files generated by the simulators, by computing the values
of distances/angles/dihedral angles. The utility is capable to aggregate results from separate
trajectories.
crd_convert

13

GENESIS User Guide, Release 1.0

An utility to convert trajectory files to PDB/DCD formats. In addition, it performs atomic

coordinate superimposition with several algorithms, and can extract coordinates of selected
atoms.
remd_convert
This utility has similar functions to crd_convert with an additional capability to handle
REMD trajectory files. It regenerates REMD trajectory files for each conditions by reading
the original trajectory files from all replicas.

3.3 Parallel I/O tools

In order to perform massive parallelization, SPDYN uses parallel I/O scheme, where each compute node
reads/writes trajectories (coordinates, velocities, etc.) independently. The following tools are used to
handle files from parallel I/O simulations. The usage of these tools is introduced in tutorial (chapter 16).
prst_setup
This utility provides input files for the parallel I/O simulation by processing huge input
files (PSF, PDB) and dividing them into multiple GENESIS restart files. The control file is
similar to that for SPDYN. The usage is shown in tutorial (in chapter 16).
pcrd_convert
This utility has similar functions to crd_convert. It is intended to handle parallel I/O trajectory files.

3.3. Parallel I/O tools

14

CHAPTER

FOUR

INPUT AND OUTPUT FILES

GENESIS uses the following input/output files. The input and output files are set in [INPUT] and
[OUTPUT] sections, respectively.

4.1 Input files

GENESIS requires topology, parameter, and psf files describing simulation system and potential energy
function. Atomic coordinates are read by either in PDB or CRD format files. If you set rstfile, the
coordinates, velocities and other properties of the system are read from the restart file only.

topfile
Topology file which contains information about atom connectivity in residues and monomer
unit. GENESIS reads topfile in CHARMM format. For details on the format, see the
CHARMM web site [14]. GENESIS can read multiple topfiles (see Note). (Required)
parfile
Parameter file which contains force field parameters, e.g., force constants and equilibrium
geometries. GENESIS reads parfile in CHARMM format. GENESIS can read multiple
parfiles (see Note). (Required)
strfile
CHARMM stream file which contains additional force field parameters. GENESIS can
read multiple strfiles (see Note). (Optional)
psffile
PSFfile contains system information about atomic mass, charge, atom connectivity, and etc.
GENESIS reads psffile in X-PLOR and CHARMM formats. (Required)
pdbfile
Atomic coordinates in PDB (Protein Data Bank) format. (Required)
crdfile
Atomic coordinates in CHARMM format. If crdfile is specified, coordinates in crdfile are
used as the initial coordinates prior to those in pdbfile. (Optional)
rstfile

15

GENESIS User Guide, Release 1.0

This file contains atomic coordinates, velocities, simulation box size and other simulation
variables with the double-precision floating-point number representation. Rstfile, which
is given in GENESIS-original format, is specified for restarting a simulation. If rstfile
exists, coordinates in pdbfile/crdfile and simulation box size in the control file are ignored.
(Optional)
reffile
Reference coordinates (PDB file format) for positional restraints. This file should contain
the same total number of atoms in pdbfile. (Optional)
localresfile (available for SPDYN only)
LocalRes file is used to applying external forces as Local restraint. (Optional)
Local restraint are the restraining forces implemented within the spatial decomposition scheme. Their are more computationally efficient than the ordinary ones, but the atoms
has to be located in the same cell.
These energy terms are harmonic potentials:
U (r) = k (r r0 )2 for bonds
U () = k ( 0 )2 for bond angles
U () = k ( 0 )2 for dihedral angles
Here, r, , and are bond distance, angle, and dihedral angles, respectively; subscript 0
denotes their reference values; and k is the force constant. The restraint energies are added
to the corresponding bond or angle/dihedral angle energies in the output file.
The file contains the following information;
[BOND/ANGLE/DIHEDRAL] atom atom [atom [atom]] k r0
atom indices start from 1.
Example for localres file
BOND
139 143
ANGLE
233 231 247
DIHEDRAL 22 24 41

43

2.0 10.0
3.0 10.0
2.0 10.0

Note: Ordinary restraint functions are set in [RESTRAINTS] section (see Restraints). The equivalent
calculations to local restraint can be performed with ordinary functions of ATDYN/SPDYN
without the spatial decomposition scheme.

Note: To specify multiple files for parfile, topfile, and strfile, comma-separated lists are written (e.g.,
parfile = par_all36_prot.prm, par_all36_na.prm, par_all36_lipid.prm). If the character string becomes
long, backslash can be used for line break and continuation.

4.1. Input files

16

GENESIS User Guide, Release 1.0

4.2 Output files

GENESIS yields trajectory files (coordinates and velocities) in DCD format. GENESIS also generates
restart file (rstfile) during simulations for restart. Output intervals of each file are set in [DYNAMCICS]
section.

dcdfile
Trajectory is written in DCD format (also used by X-PLOR and CHARMM). The
double-precision flouting-point number representation is used in the file (Required if
crdout_period > 0)
dcdvelfile
Velocities are also written in DCD format. (Required if velout_period > 0)
rstfile
ReSTart file contains coordinate, velocities, simulation box size and other dynamic informations. An MD simulation can be restarted using a rstfile generated by either a previous MD
or a MIN simulation; as an REMD simulation can be restarted from any (MIN/MD/REMD)
rstfile. (Required if rstout_period > 0)
remfile (generated in REMD simulations)
The file provides parameter ID for each replica to match replica ID and parameter ID. It is
used as an input file for remd_convert utility. (Required if exchange_period > 0)

4.2. Output files

17

CHAPTER

FIVE

ENERGY
In [ENERGY] section, there are several options to set keywords related to energy and force evaluation.

5.1 General keywords

Force field consist of a potential energy function and a set of parameter. In the case of all-atom force
field with explicit water, the potential energy function:
E(r) =

Kb (b b0 )2 +

Kub (S S0 )2 +

UB

bond

nonbond

K ( 0 )2

improper

dihedral

K ( 0 )2

angle

K (1 + cos (n )) +

"


Rmin,ij
rij

12


2

Rmin,ij
rij

6 #
+

X
nonbond

qi qj
1 rij

where Kb , Kub , K , K , and K are force constants of bond, UB 1-3 distance, angle, dihedral angle,
and improper dihedral angle potentials, respectively; b0 , S0 , 0 , and 0 are corresponding equilibrium
values; and is a phase shift of the dihedral angle potential.  is Lennard-Jones potential well depth,
Rmin,ij is a distance of the Lennard-Jones potential minimum, qi is an atomic charge, 1 is an effective
dielectric constant, and rij is a distance between two atoms. The parameters are set according to an
atom type.
The form of potential energy function is force field dependent. In GENESIS, TIP3P explicit water [20]
with CHARMM22 and CHARMM27 force fields are available (proteins: [7][8], nucleic acids: [21][22],
and lipids: [23][24]). Recently developed CHARMM36 force fields [25] [26] are also available.
As for a coarse-grained simulations, a model proposed by Karanicolas and Brooks [27] [28] is supported.
It is mainly based on the Go-like model [29] with some sequence-based parameters incorporated. The
potential energy function of the model is:

18

GENESIS User Guide, Release 1.0

E(r) =

Kb (b b0 )2 +

bond

K ( 0 )2

angle

K (1 + cos(n ))

dihedral

"
+

 13

nativecontacts

X
nonnativepairs

"




 #


Rmin,ij 12
Rmin,ij 10
Rmin,ij 6
18
+4
rij
rij
rij
#
12

Rmin,ij
rij

where the last two terms are non-bonded interactions between native contact pairs and non-native contact
pairs, respectively. Roughly speaking, the native contacts are defined for the pairs close to each other in
a PDB structure of a target molecule. Rmin,ij is a reference distance between a pair atoms in the PDB
structure.

forcefield CHARMM / KBGO

Type of force field used for energy/force evaluation (Default : CHARMM).

5.2 Non-bonded interaction related keywords

The non-bonded interaction is the most time consuming part in MD simulations. Computational time of
the non-bonded interactions without any approximation is proportional to O(N 2 ). To reduce computational cost, the cut-off approximation is introduced where energy/force is truncated at the given cut-off
value (keyword cutoffdist).
Simple spherical truncation at the cut-off value leads to discontinuities of energy and forces. So it is
necessary to introduce a polynomial function (so called switching function) that smoothly turn off the
interaction from another given value (so called switch cut-off ) for van der Waals interactions (keyword
switchdist). In general, there are two kinds of switching functions: switching function that smoothly
turns off (1) potential energy, and (2) forces. To enable the switch function that smoothly turns off
forces [30], vdw_force_swith=YES should be set.
The spherical truncation of electrostatic energy is different from van der Waals due its long-range nature.
To truncate electrostatic energy at the cut-off value, a shift function is introduced which is enable by
Electrostatic=Cutoff in GENESIS. In case of the complete electrostatic energy, the smooth
particle Mesh Ewald (PME) scheme is used (it is explained in the PME-related keyword section). Please
note that the type of electrostatic energy calculation is relevant to CHARMM all-atom force field only.

electrostatic CUTOFF/PME
Type of long range electrostatic energy/force evaluations (Default : PME)
switchdist Real
Switch cut-off distance (unit : Angstrom) (Default : 10.0)
5.2. Non-bonded interaction related keywords

19

GENESIS User Guide, Release 1.0

cutoffdist Real
Potential energy/force cut-off distance (unit : Angstrom) (Default : 12.0)
pairlistdist Real
Cut-off distance of Verlet pair list for non-bonded energy/force evaluations [31] (unit :
Angstrom) (Default : 13.5)
dielec_const Real
Dielectric constant (Default : 1.0)
vdw_force_switch YES / NO
Usage of the force switch function for van der Waals interactions (Default : NO)

5.3 Particle mesh Ewald related keywords

The following keywords are relevant if electrostatic=PME. Electrostatic energy in the smooth
particle mesh Ewald (PME) scheme [32][33] is:
Eelec =

X qi qj erfc(rij ) 2 X exp(|G|2 /42 ) X qi qj

X qi qj

+
cos(G

r
)

ij
1
rij
V
1
1
|G|2
2
ij
i<j
ij
|G| 6=0

Here, the first term is calculated with cut-off because it is decreased rapidly then pair distances increase.
The third term is so called self-energy, calculated only once. The second term is rewritten as:
X exp(|G|2 /42 )
|S(G)|2
2
|G|
2

|G| 6=0

where
S(G) =

qi exp(iG ri ) b1 (G1 )b2 (G2 )b3 (G3 )F(Q)(G1 , G2 , G3 )

For the evaluation of S(G); b1 (G1 ), b2 (G2 ), and b3 (G3 ) are approximated by the cardinal B-splines of
order n; and F(Q) is calculated by the fast Fourier transformation (FFT) of charge values.

pme_alpha Real
Exponent of complementary error function (Default : 0.34)
pme_ngrid_x Integer
Number of FFT grid points along x dimension (Required if PME is used)
5.3. Particle mesh Ewald related keywords

20

GENESIS User Guide, Release 1.0

pme_ngrid_y Integer
Number of FFT grid points along y dimension (Required if PME is used)
pme_ngrid_z Integer
Number of FFT grid points along z dimension (Required if PME is used)
pme_nspline Integer
B-spline order of charges for the evaluation from b1 (G1 ) to b3 (G3 ) (Default : 4)
pme_multiple YES/NO
Enable partitioning of processors for real and reciprocal PME term computation (available
in ATDYN only) (Default : NO)
pme_mul_ratio Integer
Ratio of process numbers for real and reciprocal PME term computation (available in ATDYN only and used when PME_multiple=YES only) (Default : 1)

5.3.1 Number of PME grid points

In ATDYN and SPDYN, the number of PME grid points should be multiples of 2, 3, and 5. Moreover,
in SPDYN, there are several notes to define PME grid numbers according to processor numbers. In
SPDYN, we first define domain numbers in each dimension such that product of them equals to the total
number of MPI processors. Let us assume that the domain numbers in each dimension are domain_x,
domain_y, and domain_z. If you execute configure with --enable-fft3d option, there are following
restriction in deciding PME grid numbers:
1. pme_ngrid_x should be multiple of (2*domain_x)
2. pme_ngrid_y should be multiple of (domain_y*domain_z)
3. pme_ngrid_z should be multiple of (domain_y*domain_z) and (domain_x*domain_z)
If you do not add --enable-fft3d option, the restriction condition of the grid numbers are changed
as following :
1. pme_ngrid_x should be multiple of (2*domain_x)
2. pme_ngrid_y should be multiple of (2*domain_y)
3. pme_ngrid_z should be multiple of domain_z
In the case of SPDYN, parallelization of FFT is written inside the program and both FFTE [34] and
FFTW [35] libraries are used for the one-dimensional FFT. As for ATDYN, FFTE library is used for
FFT only.

5.4 Lookup table related keywords

The following keywords are relevant if electrostatic=PME. For a linearly-interpolating lookup
table (table_order = 1), table points are assigned at the unit interval of cut-off2 /r2 and energy/gradients are evaluated as a function of b2 (G2 ) [11].

5.4. Lookup table related keywords

21

GENESIS User Guide, Release 1.0

F (r2 ) Ftab (L) + t(Ftab (L + 1) Ftab (L))

where
L = INT(Density rv2 /r2 )

and
t = Density rv2 /r2 L

Density is the number of points per a unit interval. Lookup table using cubic interpolation is different
from that of linear interpolation. In the case of cubic interpolation, monotonic cubic Hermite polynomial
interpolation is used to impose the monotonicity of the energy value. Energy/gradients are evaluated as
a function of r2 [36] using four basis functions for the cubic Hermite spline : h00 (t), h10 (t), h01 (t),
h11 (t)
Ftab (L 2) + Ftab (L 1)
h10
2
Ftab (L 1) + Ftab (L)
h11 (t)
+ Ftab (L)h10 (t) +
2

F (r2 ) Ftab (L 1)h00 (t) +

where
L = INT(Density r2 )

and
t = Density r2 L

table YES / NO
Enable the lookup table (Default : YES)
table_density Integer
Number of table points per unit interval (Default : 20)
table_order 0 / 1 / 3
Order of interpolation used in the lookup table (Default : 1 (Electrostatic=PME), 3 (Electrostatic=Cutoff)
water_model TIP3 / NONE
Type of water model used for the lookup table approach (Default : TIP3)
5.4. Lookup table related keywords

22

CHAPTER

SIX

DYNAMICS
In [DYNAMICS] section, we choose integrator and time step options for MD simulations.

6.1 General keywords

In GENESIS,two integrators are available:velocity Verlet and leapfrog. Velocity Verlet (VVER) has two
stages to update velocities v(t) and coordinates r(t) at time t with time step t. The update procedure
is as follow:
Stage 1 :
t
1
F(t)
v(t + t) = v(t) +
2
2m
1
r(t + t) = r(t) + tv(t + t)
2

where m is the mass.

Force calculation F(t + t)
Stage 2 :
v(t + t) = v(t) +

t
[F(t) + F(t + t)]
2m

Leapfrog (LEAP) integrator has one stage. After force evaluation, velocities at half time step and coordinates at next integer time step are updated as follow:
1
1
t
v(t + t) = v(t t) +
F(t)
2
2
m
1
r(t + t) = r(t) + tv(t + t)
2

integrator LEAP / VVER

23

GENESIS User Guide, Release 1.0

Type of integrator (Default : LEAP).

nsteps Integer
Number of MD steps (Default : 100)
timestep Real
Time step (unit : ps) (Default : 0.001)
eneout_period Integer
Period of energy outputs in time steps (Default : 10)
crdout_period Integer
Period of coordinates outputs in time steps (Default : N/A)
velout_period Integer
Period of velocity outputs in time steps (Default : N/A)
rstout_period Integer
Period of restart files updates in time steps (Default : N/A)
stoptr_period Integer
Period of translation and rotation motion removal in time steps (available in ATDYN only)
(Default : 0 (i.e. No translational rotational motion is removed.))
nbupdate_period Integer
Period of updating the non-bonded pair list in time steps (Default : 10)
pairalloc_perid Integer
Period of memory reallocation for Verlet pair list in time steps(available in ATDYN only)
(Default : 0)
iseed Integer
Seed of pseudo-random number generator (if not set, it is decided from current date and
time) (Optional)

6.2 Simulated annealing related keywords

The following keywords are relevant for MD simulations only.

annelaing YES/NO
Enable simulated annealing (Default : NO)
annelaing_period Integer
Period of temperature steps in time steps (Default : N/A)
dtemperature Real
Temperature change per temperature step (unit : Kelvin) (Default : 0.0)

6.2. Simulated annealing related keywords

24

CHAPTER

SEVEN

MINIMIZE
In [MINIMIZE] section, we choose options for energy minimization. Currently, the steepest descent
algorithm is available only.

method SD
Algorithm of minimization (Default : SD).
nsteps Integer
Number of minimization steps (Default : 100)
eneout_period Integer
Period of energy outputs in minimization steps (Default : 10)
crdout_period Integer
Period of coordinates outputs in minimization steps (Default : N/A)
rstout_period Integer
Period of restart file updates in minimization steps (Default : N/A)
nbupdate_period Integer
Period of non-bonded pair list updates in minimization steps (Default : 10)
Note: If the initial structure is very deviated from the equilibrium and there are several non-bonded
interactions at short distances, it is recommend to assign table_order=0 to avoid large energy/force
values. Please note, constraints are not available in minimization.

25

CHAPTER

EIGHT

CONSTRAINTS
In [CONSTRAINTS] section, constraint keywords are described. SHAKE scheme is used for bonds
between heavy and hydrogen atoms [37]. For velocity Verlet integrator, RATTLE constraint is also
available [38]. Bond constraints between heavy atoms are not imposed. If explict water molecules
(TIP3P) are present, SETTLE algorithm is enabled automatically [39].

rigid_bond YES/NO
Enable constraints (not availalbe for minimization) (Default : NO).
shake_iteration Integer
Number of iterations for coordinates/velocities updates. If it does not converge within the
given number of iterations, the program terminates with an error message. (Default : 500)
shake_tolerance Real
Tolerance of SHAKE convergence (unit : Angstrom) (Default : 1.0e-10)
faster_water YES/NO
Usage of SETTLE algorithm for constraints in water molecules (Default : YES)
(fast_water=NO is not available in SPDYN).

26

CHAPTER

NINE

ENSEMBLE
In [ENSEMBLE] section, type of ensemble, temperature and pressure control algorithm, and parameters used in the algorithms are specified.
In the Langevin thermostat algorithm (ensemble=NVT with tpcontrol=LANGEVIN), every particle is coupled with a viscous background and a stochastic heat bath [40]:
dv(t)
F(t) + R(t)
=
v(t)
dt
m

where is the thermostat friction parameter (gamma_t keyword) and R(t) is the stochastic force. In
the Langevin thermostat and barostat method (ensemble=NPT with tpcontrol=LANGEVIN), the
equation of motions are given by [41]:
dr(t)
= v(t) + v r(t)
dt
dv(t)
F(t) + R(t)
3
=
[p + (1 + )v ]v(t)
dt
m
f
dv (t)
3K
= [3V (P (t) P0 (t)) +
p v + Rp ]/pmass
dt
f

where K is the kinetic energy, p is the barostat friction parameter (gamma_p keyword), Rp is the
stochastic pressure variable.

ensemble NVE / NVT / NPT / NPAT / NPgT

Type of ensemble (Default : NVE).
NVE: Microcanonical ensemble
NVT: Canonical ensemble
NPT: Isothermalisobaric ensemble
NPAT: Constant area, normal pressure, temperature
NPgT: Constant surface-tension, normal pressure, temperature
temperature Real
27

GENESIS User Guide, Release 1.0

Initial and target temperature (unit : Kelvin) (Default : 298.15)

pressure Real
Target pressure in NPT or target normal pressure in NPAT and NPgT (unit : atm) (Default
: 1.0)
gamma Real
Target surface tension in NPgT (unit : dyn/cm) (Default : 0.0)
tpcontrol NO / BERENDSEN / LANGEVIN
Type of thermostat and barostat (Default : NO). Available options are listed in Table I.
NO: Do not use temperature/pressure control algorithm (for NVE only)
BERENDSEN: Berendsen thermostat/barostat
LANGEVIN: Langevin thermostat/barostat
tau_t Real
Temperature coupling time in Berendsen thermostat (unit : ps) (Default : 5.0)
tau_p Real
Pressure coupling time in Berendsen barostat (unit : ps) (Default : 5.0)
compressibility Real
Compressibility parameter in Berendsen thermostat (unit : 1/atm) (Default : 0.0000463)
gamma_t Real
Friction parameter in Langevin thermostat (unit : 1/ps) (Default : 1.0)
gamma_p Real
Friction parameter in Lengevin barostat (unit : 1/ps) (Default : 0.1)
isotropy ISO/ANISO/SEMI-ISO/XY-FIXED
Isotropy of the simulation system (Default : ISO). This parameter specifies how X, Y, Z
dimensions of the simulation box change in NPT, NPgT, NPAT ensembles.
ISO: X, Y, and Z dimensions are coupled together
ANISO: X, Y, and Z dimensions fluctuate independently
SEMI-ISO: X, Y, and Z dimensions fluctuate, where the ratio of X and Y dimensions
are kept constant, and Z dimension changes independently
XY-FIXED: X and Y dimensions are fixed, while Z dimension changes (for NPAT
only)

Table I. Combinations of integrators and thermostats/barostats

tpcontrol
Berendsen
Langevin

LEAP NVT
O
O

LEAP NPT
O
O

VVER NVT
O
O

VVER NPT
X
O

28

CHAPTER

TEN

BOUNDARY
In [BOUNDARY] section, boundary conditions of the system such as simulation box size are specified.

type PBC/NOBC
Type of boundary condition (Default : PBC).
PBC: Periodic boundary condition (rectangular or cubic box)
NOBC: Non boundary condition (vacuum system) (available in ATDYN only).
box_size_x Real
Box size along the x dimension (unit : Angstrom) (Required if PBC is used)
box_size_y Real
Box size along the y dimension (unit : Angstrom) (Required if PBC is used)
box_size_z Real
Box size along the z dimension (unit : Angstrom) (Required if PBC is used)
domain_x Integer
Number of domains along the x dimension (Optional; available in SPDYN only)
domain_y Integer
Number of domains along the y dimension (Optional; available in SPDYN only)
domain_z Integer
Number of domains along the z dimension (Optional; available in SPDYN only)
Note: If number of domains (domain_x, domain_y, and domain_z) are not specified in the control file,
they are automatically determined based on the number of MPI processes. If the user wants to specify
number of domains explicitly, the product of domain numbers (domain_x x domain_y x domain_z) must
be equal to the total number of MPI processes.

Note: If the simulation system has a periodic boundary condition, the user must specify the box size
initially. During simulations, box size is saved in a restart file. If the restart file is used as an input of the
subsequent restart MD, the initial box size is replaced with it, even if another value is specified in the
control file.

29

CHAPTER

ELEVEN

SELECTION
To perform MD simulations with restraints like umbrella sampling, the user must select atoms to be
restrained. Once the group of selected atoms is defined in this section, the group index can be called
from [RESTRAINTS] section. Nothing is selected as default.
[SELECTION] section is also included in a control file of some analysis tools, in which selected atoms
are used as analysis and fitting atoms.

group expression
Select atoms by expression and define them as a group. Available keywords and operators in expression are listed in Table II. Note that mname (or moleculename, molname) in
expression is a molecule name that is defined by mol_name below.
mole_name molecule starting-residue ending-residue
Define molecule by starting-residue and ending-residue. Those residues are defined by
[segment id]:residue number:residue name

Table II. Available keywords and operators in group.

expression
an:name
ai:number[-[number]]
atno:number[-[number]]
rnam:name
rno:number[-[number]]
mname:name
segid:ID
water
hydrogen
heavy
backbone
all
and
or
not
()

meaning
atom name
atom index
atom number
residue name
residue number
molecule name
segment index
water molecule
hydrogen atoms
heavy atoms
protein backbone
all atoms
conjunction
logical add
negation
assemble

example
an:CA
ai:1-5
atno:6
rnam:GLY
rno:1-5
mname:molA
segid:PROA

other available expression

atomname, atom_name
atomindex, atomidx
atomno
residuename, resname
residueno, resno
moleculename, molname
segmentid, sid
hydrogenatom
heavyatom
backboneatom
*
&
|
!

30

GENESIS User Guide, Release 1.0

Note: ai, atomindex, and atomidx indicate the index of atom that is sequentially renumbered over
all atoms in the system, while atno and atomno are the index of atom that is assigned to each atom in
PDB file. Atom index in PDB file (column 2) is not always starting from 1 or numbered sequentially. If
the user want to use such PDB file as an input file, although it shoule be a rare case, atno and atomno
are useful to select atoms.

Example of [SELECTION] section

[SELECTION]
group1
=
group2
=
mole_name1 =
group3
=

resno:1-60 and an:CA

(segid:PROA and not hydrogen) | backbone
molA PROA:1:TYR PROA:5:MET
mname:molA and backbone

31

CHAPTER

TWELVE

RESTRAINTS
In [RESTRAINTS] section contains keywords to set external harmonic restraint functions. The restraint
functions are applied to the selected atom groups in [SELECTION] section.
External harmonic restraint functions are applied to the selected atom groups in order to restrict the
motions of the atoms. The potential energy of a restaint is:
U (x) = k (x x0 )2

where x is a variable (see bellow), x0 is a reference value, and k is a force constant.

Note: In this section, ordinary restraint functions can be set. If you want to apply local restraint
functions, please use localresfile in [INPUT] (see Input and Output files).

nfunctions Integer
Number of ordinary restraint functions. Note that number of local restraint is not included.
The following parameters is set with serial number (Maximum is nfunctions). (Default: 0)
function POSI /DIST[MASS] /ANGLE[MASS] /DIHED[MASS] /RMSD[MASS]
Type of harmonic restraint. (Default: N/A)
Each keyword is used with serial number up to numfuncs POSI: positional restraint. The
reference coordinates are set by reffile in [INPUT]. (see Input and Output files)
DIST[MASS]: distance restraint.
ANGLE[MASS]: angle restraint.
DIHED[MASS]: dihedral angle restraint.
RMSD[MASS] (available in ATDYN only) : RMSD restraint. MASS means mass-weighted
RMSD. The calculation is done without superimposing the reference coordinates.
DIST, ANGLE, DIHED are capable to calculate distance/angle/dihedral among representative points of the groups. MASS indicates that the force is applied to the center of mass
of a selected group. The force without MASS is applied to the arithmetic average of the
coordinates.
constant Real
32

GENESIS User Guide, Release 1.0

Force constant of a restraint function. (Default: 0.0)

Unit for DIST/RMSD is kcal/mol/Angstrom2 and that for ANGLE/DIHED is kcal/mol/rad2 .
The keyword is capable to accept multiple numbers. By setting two values for DIST, ANGLE, DIHED, and RMSD, a constraint with different upper/lower force constants can be
set.
reference Real
Reference value of a restraint function. For the position restraint, the value is ignored.
(Default: 0.0)
Unit for DIST is Angstrom. Note, the unit in ANGLE/DIHED is degree (NOT radian) even
though constant is kcal/mol/rad2 . The keyword is capable to accept multiple numbers.
By setting two values in DIST, ANGLE, DIHED, and RMSD, a constraint with different
upper/lower references can be set. (i.e. you can use a flat-bottom harmonic potential.)
select_index Integer
Index of a atom group, where restraint forces are applied. The index is set in [SELECTION]. (see Selection) (Default: N/A)

Example for [RESTRAINTS]

[RESTRAINTS]
nfunctions
function1
reference1
constant1
select_index1

=
=
=
=
=

1
DIST
10.0
2.0
1 2

33

CHAPTER

THIRTEEN

REMD
In [REMD] section, the users can specify keywords for Replica-Exchange Molecular Dynamics
(REMD) simulation. REMD method is one of the enhanced conformational sampling methods used
for systems with rugged free-energy landscapes. The original temperature-exchange method (T-REMD)
is the most widely used in simulations of bio-molecules [18] [42]. Here, replicas (or copies) of the
original system are prepared, and different temperatures are assigned to each replica. Each replica is run
in a canonical (NVT) or isobaric-isothermal (NPT) ensemble, and target temperatures are exchanged
between a pair of replicas during a simulation. Exchanging temperature enforces a random walk in
temperature space, resulting in the simulation surmounting energy barriers and the sampling of a much
wider conformational space of target molecules.
In REMD methods, the transition probability for the replica exchange process is given by the usual
Metropolis criterion,
w(X X 0 ) = min(1,

P (X 0 )
) = min(1, exp()).
P (X)

In the T-REMD method, we have

n
o
= (m n ) E(q [j] ) E(q [i] ) ,

where E is the potential energy, q is the position of atoms, is the inverse temperature defined by
= 1/kB T , i and j are the replica indexes, and m and n are the parameter indexes. After the replica
exchange, atomic momenta are rescaled as follows:
[i]0

r
=

Tn [i]
p ,
Tm

[j]0

r
=

Tm [j]
p ,
Tn

where T is the temperature and p is the momenta of atoms.

The transition probability should be independent of constant temperature and constant pressure algorithms, whereas the momenta-rescaling scheme depends on the algorithm used in the simulation. If
thermostat and barostat momenta are included in the equations of motion, these variables should be also
rescaled after replica exchange. In GENESIS, thermostat and barostat momenta are rescaled in the case
34

GENESIS User Guide, Release 1.0

T-REMD is performed with the Langevin method in NPT, NPAT, and NPgT ensembles. For other cases,
only atomic momenta are rescaled.
In GENESIS, not only Temperature REMD but also pressure REMD, surface-tension REMD, REUS (or
Hamiltonian REMD), and their multi-dimensional version are available in both ATDYN and SPDYN.
REMD simulations in GENESIS require a MPI environment. At least one MPI processors must be
assigned to one replica. For example, in the case that the user wants to prepare 32 replicas, (32 x n) MPI
processors are needed.

dimension Integer
Number of dimensions (i.e. number of parameter types to be exchanged) (Default: 1).
exchange_period Integer
Period of replica exchange in time steps (Default: 100). If exchange_period = 0 is
specified, REMD simulation without parameter exchange is executed.
iseed
Random number seed in the replica exchange scheme (Default: 3141592).
type TEMPERATURE / PRESSURE / GAMMA / RESTRAINT
Type of parameter to be exchanged in a dimension (Default: TEMPERATURE).
TEMPERATURE: Temperature REMD
PRESSURE: Pressure REMD
GAMMA: Surface-tension REMD
RESTRAINT: REUS (or Hamiltonian REMD)
nreplica Integer
Number of replicas (or parameters) in a dimension (Default: 0).
parameters real / (real, real)
List of parameters in a dimension. Parameters are separated by white spaces, and the total number of parameters must be equal to the above nreplica. In REUS (type =
RESTRAINT), parameters should be specified in parentheses, where the first and second
values separated by a comma correspond to a force constant and reference value in the
restraint potential function, respectively. Even if force constants and reference values are
specified here, dummy values must be set in the corresponding restraint function in [RESTRAINTS] section.
rest_function (only available for REUS)
Index of restraint function, which is pointing to the restraint function defined in [RESTRAINTS] section (see Restraints).
If rest_function is a positinal restraint, multiple PDB files with different coordinates
should be prepared by specifying reffile = filename{}.pdb in [INPUT] section.
In this case, reference values in the above parameters are ignored but should be specified
as a dummy.
cyclic_params YES / NO

35

GENESIS User Guide, Release 1.0

If cyclic_params = YES is specified, the first and last parameters in parameters

are considered as neighbouring parameters (Default: NO). Cyclic parameter is useful when
REUS in dihedral angle space is carried out.
Note: When one dimensional T-REMD, P-REMD, and surface-tension REMD is carried out, parameters specified by temperature, pressure, and gamma in [ENSEMBLE] section are ignored,
respectively. For example, in the case T-REMD in the NPT ensemble is carried out, target temperature in each replica is set to the parameters in [REMD] section, while the target pressure is set to
pressure in [ENSEMBLE] section

Note: When multi-dimensional REMD is carried out, parameters are exchanged alternatively. For
example, in TP-REMD (type(1) = TEMPERATURE and type(2) = PRESSURE), temperature is exchanged at the first exchange trial, and pressure is exchanged at the second trial. This is repeated during
the simulations.

Note: All parameters except for exchange_period in [REMD] section should not be changed
before and after restart run.

Example of conventional T-REMD using 4 replicas

[REMD]
dimension
exchange_period
type(1)
nreplica(1)
parameters(1)

=
=
=
=
=

1
1000
TEMPERATURE
4
300 310 320 330

Example of two-dimensinal REMD (T-REMD/REUS) using 32 replicas

[REMD]
dimension
exchange_period
type(1)
nreplica(1)
parameters(1)
type(2)
nreplica(2)
parameters(2)
rest_function(2)

=
=
=
=
=
=
=
=
=

[SELECTION]
group1
group2

= ai:1
= ai:2

[RESTRAINTS]
nfunctions
function1
reference1
constant1
select_index1

=
=
=
=
=

2
1000
TEMPERATURE
8
300 310 320 330 340 350 360 370
RESTRAINT
4
(2.0,10.0) (2.0,10.5) (2.0,11.0) (2.0,11.5)
1

1
DIST
10.0
2.0
1 2

# dummy
# dummy

36

CHAPTER

FOURTEEN

TUTORIAL 1: BUILDING AND

SIMULATING BPTI IN WATER
In this chapter, a step-by-step tutorial for a molecular dynamics simulation with GENESIS is provided.
The aim is to guide a new user through the process of building and simulating a system containing a
protein (Bovine pancreatic trypsin inhibitor: BPTI) in a box with water molecules and ions.
BPTI, a 58-residue globular protein, is a popular molecule in the simulation community because of
its small size and stability. Indeed, it is the first protein for which a molecular dynamics simulation
study was performed [43] and also a 1-millisecond simulation was recently achieved with a specialpurpose supercomputer [44]. In this tutorial, we examine its stability by a 1-nanosecond simulation
with GENESIS.
The input files for this chapter are supplied in tutorial/bpti/ of the GENESIS package. Before
you start the tutorial, please note the following things:
Running all of the simulations requires several hours (or a whole day, depending on a machine
specifications).
VMD [17] is required for building the simulation system and the visualization of results.
gnuplot [45] is required for making plots.
PATH environment variable includes a directory with GENESIS binary.
If you are a Mac OSX user, it is recommended to make an alias for VMD:
$ alias vmd="/Applications/VMD\ 1.9.app/Contents/MacOS/startup.command"

14.1 Building a simulation system

We build a simulation system containing a BPTI in a simulation box with water molecules and ions
using VMD and its plugins, and generate two files required for a simulation with GENESIS:
1. psffile, containing the connectivity, mass, and charges of the molecule;
2. pdbfile, containing the initial structure of the molecule.
For the details about the files, see Input and Output files.
To build a simulation system, we download a crystal (or NMR) structure of BPTI from the RCSB Protein
Data Bank (PDB) [46]. You can download the PDB file via a web browser, or directly get it by using
curl command:
37

GENESIS User Guide, Release 1.0

#
$
#
$
#
$

change to a setup directory

cd tutorial/bpti/1_setup/
download the PDB file (PDB code 4PTI)
curl -o 4pti.pdb http://www.rcsb.org/pdb/files/4pti.pdb
check the downloaded file
less 4pti.pdb

Lets look at the header entries of the file. It can tell a lot about this molecule, for example, the structure
was determined by X-ray diffraction (described in EXPDTA entry in the file), there is no missing residues
(no MISSING entries), and there are three disulfide bonds between 5-55, 14-38, 30-51 cystein residue
pairs (SSBOND entries).
Since atom the naming conventions of PDB and CHARMM are slightly different, we need manually to
edit the file to comply with the CHARMM convention with a text editor (such as vi or Emacs). We need
to replace O and OXT1 atoms of the C-terminal residue (ALA58) with OT1 and OT2, respectively. After
the edit, ALA58 should look like follow:
$ less 4pti.pdb
...skipped...
ATOM
449 N
ATOM
450 CA
ATOM
451 C
ATOM
452 OT1
ATOM
453 CB
ATOM
454 OT2
TER
455
...skipped...

ALA
ALA
ALA
ALA
ALA
ALA
ALA

A
A
A
A
A
A
A

58
58
58
58
58
58
58

25.146
25.617
25.248
24.962
27.160
24.919

29.681
30.840
30.735
31.791
30.980
29.594

-6.493
-7.256
-8.729
-9.369
-7.146
-9.172

1.00
1.00
1.00
1.00
1.00
1.00

46.21
45.05
46.90
39.78
50.07
43.54

N
C
C
O
C
O

Since GENESIS does not provide any programs for building simulation systems, we use other packages
such as CHARMM [14] or psfgen [15] (supplied with NAMD [16]). In this chapter, we use VMD
[17], and psfgen is called via the plug-in interface within VMD. Other VMD plug-ins (solvate [47] and
autoionize [48]) are also used for solvation and neutralization, respectively.
A VMD script for building the simulation system containing a BPTI molecule in a box with water
molecules and ions is at tutorial/bpti/1_setup/setup.tcl. We can run it with VMD:
$ vmd -dispdev text <setup.tcl | tee run.out

Here, with -dispdev text option, we inhibits any graphical display windows from opening (we do
not need any visualization for this set-up). The redirection <setup.tcl instructs VMD to process the
script. The content of the script, written in Tcl languages, is shown below.
The script consists of four parts. In the first part, the crystal water molecules are removed by using the
atom selection facility of VMD. The protein structure without crystal water molecules is written in a pdbfile (4pti_protein.pdb). In the second part, psfgen plugin is called. psfgen matches the residues in
the original file with the residues defined in topfile (top_all27_prot_lipid.rtf), and then creates disulfide bonds by patching special residues for them. Also, the coordinates of missing hydrogen
atoms are guessed from the topology. Then, a psffile (protein.psf) and a pdbfile (protein.pdb)
of BPTI without solvent are written. In the third part, solvate plugin is invoked to solvate BPTI with
TIP3P bulk water molecules and fills the box for simulations under the periodic boundary conditions.
The resulting system is written in solvate.psf and solvate.pdb. In the fourth part, counter ions
are added to neutralize the system using autoionize plug-in. The final results are saved as ionize.psf
and ionize.pdb.

14.1. Building a simulation system

38

GENESIS User Guide, Release 1.0

##### read pdb and remove crystal waters

mol load pdb 4pti.pdb
# invert the center of mass
set all [atomselect top all]
set protein [atomselect top protein]
$all moveby [vecinvert [measure center $protein weight mass]]
# write pdb of protein
$protein writepdb 4pti_protein.pdb
mol delete all
##### generate psf by using the psfgen plugin
package require psfgen
resetpsf
topology top_all27_prot_lipid.rtf
segment BPTI {
first nter
last cter
pdb 4pti_protein.pdb
}
patch DISU BPTI:5 BPTI:55
patch DISU BPTI:14 BPTI:38
patch DISU BPTI:30 BPTI:51
regenerate angles dihedrals
pdbalias atom ILE CD1 CD
coordpdb 4pti_protein.pdb BPTI
guesscoord
# write psf and pdb files of protein
writepsf protein.psf
writepdb protein.pdb
##### solvate with TIP3P waters by using the solvate plugin
mol delete all
package require solvate
solvate protein.psf protein.pdb -rotate -t 22.5 -o solvate

##### add counter ions by using the autoionize plugin

mol delete all
package require autoionize
autoionize -psf solvate.psf -pdb solvate.pdb -neutralize -cation SOD -anion CLA -seg IO
##### invert the center of mass and check boxsize
set all [atomselect top all]
$all moveby [vecinvert [measure center $all weight mass]]
$all writepdb ionize.pdb
set minmax [measure minmax $all]
foreach {min max} $minmax { break }
foreach {xmin ymin zmin} $min { break }
foreach {xmax ymax zmax} $max { break }
puts "Box size estimation:"
puts "boxsizex = [expr abs($xmin)+ $xmax + 1.0]"

14.1. Building a simulation system

39

GENESIS User Guide, Release 1.0

puts "boxsizey = [expr abs($ymin)+ $ymax + 1.0]"

puts "boxsizez = [expr abs($zmin)+ $zmax + 1.0]"
exit

It is always a good idea to check the final structure by a visual inspection. We can visualize the structure
with VMD by specifying psffile (ionize.psf) and pdbfile (ionize.pdb) together:
$ vmd -psf ionize.psf -pdb ionize.pdb

If the building has successfully finished, the structure looks like follows:

Note: The size (70 A x 83 A x 68 A) of the simulation box built at this step is rather large for a
typical globular protein, because we are going to use SPDYN. For the spatial decomposition scheme in
SPDYN, the simulation box size in each of direction has to be 5-times larger than the pair list distance
(pairlistdist=13.5 distance is in Angstrom). If you prefer a smaller simulation box, decrease the
padding distance (-t 22.5) in solvate plug-in, and use ATDYN for your simulations instead.

14.1. Building a simulation system

40

GENESIS User Guide, Release 1.0

14.2 Minimization with restraints on the protein

The initial structures often contain non-physical steric clashes or non-equilibrium geometries, especially
at the interface between the protein and solvent, because the solvent molecules are randomly placed
around the protein in the building process. Thus, before performing a molecular dynamics simulation,
it is imperative to remove such clashes by minimizing the potential energy of the system. ATDYN and
SPDYN, support the steepest descent algorithm for minimization, which moves atoms proportionally to
their negative gradients of the potential energy.
The following command performs a 1000-step minimization with SPDYN:
#
$
#
$
#
$

change to the minimization directory

cd tutorial/bpti/2_minimization/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform minimization with SPDYN by using 8 MPI processes
mpirun -np 8 spdyn run.inp | tee run.out

The control file (run.inp, shown below) consists of several sections, such as [INPUT], [OUTPUT],
and [ENERGY], etc., where we can specify the control parameters for the calculation.
In [INPUT] section, input files for the minimization are specified. topfile (topology file), parfile (parameter file), psffile (PSF file), pdbfile (PDB file with an initial structure) are always required. As an
optional input, reffile (reference file) is given as the reference structure for positional restraints (see Input
and Output files for an explanation of each input file).
In [OUTPUT] section, output files are specified. SPDYN does not create any output file unless we
explicitly specify the files. Here, rstfile (restart file) is used for the restart of the simulation (see Input
and Output files for an explanation of each output file).
In [ENERGY] section, we set the keyword related to the energy and force evaluation. Here, the particle
mesh ewald (PME) method is selected for computing electrostatic interactions, usually combined with
the periodic boundary condition (PBC) in [BOUNDARY] section. By default, an interpolation scheme
with the lookup table is used for the evaluations of non-bonded interactions.
[MINIMIZE] section turns on the minimization engine of SPDYN. Here, we specify 1000 steps of the
steepest descent algorithm (SD). See Minimize for details.
In [BOUNDARY] section, the boundary condition and the simulation box size are set. Here, we use the
periodic boundary condition (PBC). The values of the box size were taken from the previous building
step. See Boundary for details.
In [SELECTION] section, we define a group (group1) consisting of backbone atoms of the protein to
impose positional restraints. For the detailed expressions for selection, see Selection.
In [RESTRAINTS] section, positional restraints are specified for the group of backbone atoms defined
in the previous [SELECTION] section. The positional restraints are imposed with respect to the reference coordinates given by reffile in [INPUT] section. See Restraints for details.
[INPUT]
topfile
parfile
psffile
pdbfile
reffile

=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ionize.psf
../1_setup/ionize.pdb
../1_setup/ionize.pdb

14.2. Minimization with restraints on the protein

#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
reference for restraints

41

GENESIS User Guide, Release 1.0

[OUTPUT]
dcdfile = run.dcd
rstfile = run.rst

# DCD trajectory file

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=

[MINIMIZE]
nsteps
eneout_period
crdout_period
rstout_period

=
=
=
=

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z

=
=
=
=

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

PME
10.0
12.0
13.5
72
80
72

#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

1000
100
100
1000

#
#
#
#

number of steps
energy output period
coordinates output period
restart output period

PBC
70.8250
83.2579
69.0930

#
#
#
#

[PBC,NOBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

1
POSI
10.0
1

number of functions
restraint function type
force constant
restraint group

After the minimization, it is recommended to confirm the decrease of the potential energy. The output
format of GENESIS is simple, so we can easily extract the potential energy term with standard UNIX
tools:
$ grep "^INFO" run.out | tail -n +2 | awk {print $2, $5} >pot.dat
$ gnuplot
gnuplot> set xlabel "step"
gnuplot> set ylabel "potential energy [kcal/mol]"
gnuplot> plot "pot.dat" w lp

14.2. Minimization with restraints on the protein

42

GENESIS User Guide, Release 1.0

14.3 Heat-up with restraints on the protein

In this step, we heat up the system from 0.1 K to 300 K during a 10-picosecond molecular dynamics
simulation. During the simulation, the positional restraints are imposed again on the proteins backbone
not to disrupt the structure due to abrupt increase in temperature.
The following command performs a 10-picosecond molecular dynamics simulation, gradually heating
the system from 0.1 K to 300 K with SPDYN:
#
$
#
$
#
$

change to the heating directory

cd tutorial/bpti/3_heating/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform heat-up with SPDYN by using 8 MPI processes
mpirun -np 8 spdyn run.inp | tee run.out

In [INPUT] section of the control file (run.inp), rstfile is explicitly set for the restart from the previous
minimization.
[INPUT]
topfile
parfile
psffile
pdbfile
reffile
rstfile

=
=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ionize.psf
../1_setup/ionize.pdb
../1_setup/ionize.pdb
../2_minimization/run.rst

#
#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
reference for restraints
restart file

Major differences from the previous minimization are [DYNAMICS], [CONSTRAINTS], and [ENSEMBLE] sections with the parameters for molecular dynamics simulations.
[DYNAMICS] section turns on the molecular dynamics engine of SPDYN. In this section, the parameters related to molecular dynamics are specified. For the heating up of the system, a simulated annealing
protocol is enabled with annealing=YES. The simulated annealing algorithm increases the target
temperature by dtemperature K every anneal_period steps. In this case, the temperature is

14.3. Heat-up with restraints on the protein

43

GENESIS User Guide, Release 1.0

increased by dtemperature=3 K every anneal_period=50 steps. Thus, nsteps=5000 MD

steps results into temperature increase by 3*5000/50=300 K. See Dynamics for details.
[CONSTRAINTS] section is for constraints during a simulation. rigid_bond is a keyword for
the SHAKE algorithm. For TIP3P waters, a faster algorithm (SETTLE) is automatically applied if
rigid_bond=YES. See Constraints for details.
In [ENSEMBLE] section, we can choose a thermostat and a barostat from several standard options. For
typical simulations, LANGEVIN or BERENDSEN is recommended. See Ensemble for details.
In [BOUNDARY] section, we don not set the simulation box size, since the box size values are read
from the restart file.
[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period
annealing
anneal_period
dtemperature

=
=
=
=
=
=
=
=
=

[CONSTRAINTS]
rigid_bond

= YES

# constraints of all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 0.1

# [NVE,NVT,NPT]
# thermostat
# initial temperature (K)

[BOUNDARY]
type

= PBC

# [PBC,NOBC]

LEAP
5000
0.002
50
50
5000
YES
50
3

#
#
#
#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period
simulated annealing
annealing period
temperature change at
annealing (K)

After the heating up simulation, the increase in temperature can be confirmed in this way:
$ grep "^INFO" run.out | tail -n +2 | awk {print $3, $21} >temp.dat
$ gnuplot
gnuplot> set xlabel "time [ps]"
gnuplot> set ylabel "temperature [K]"
gnuplot> plot "temp.dat" w lp

14.3. Heat-up with restraints on the protein

44

GENESIS User Guide, Release 1.0

Also, it is recommended to visualize the trajectory to confirm that the protein structure is not disrupted.
We can visualize the trajectory with VMD by the following command:
$ vmd -psf ../1_setup/ionize.psf -dcd run.dcd

Note: For real applications, 10 ps may be too short for the heating up process. For your research, longer
runs than 1-nanosecond would be recommended. The same is true for the next equilibration step.

14.4 Equilibration simulation

In this step, we equilibrate the system under the condition of 300 K and 1 atm by a 10-picosecond
molecular dynamics simulation. Note that the ensemble is changed from NVT to NPT. During the
simulation, we no loner impose any restraints on the protein, and we relax the whole system for the next
production simulation.
The following command performs a 10-picosecond molecular dynamics simulation under the condition
of 300 K and 1 atm with SPDYN:
#
$
#
$
#
$

change to the equilibration directory

cd tutorial/bpti/4_equilibration/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform equilibration with SPDYN by using 8 MPI processes
mpirun -np 8 spdyn run.inp | tee run.out

In the control file, [SELECTION] and [RESTRAINTS] sections are removed, since we performs a
restraint-free simulation. Most of the control parameters are same as those in the previous heat-up
simulation, except for rstfile, as we specify the output of the previous simulation.

14.4. Equilibration simulation

45

GENESIS User Guide, Release 1.0

[INPUT]
topfile
parfile
psffile
pdbfile
reffile
rstfile

=
=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ionize.psf
../1_setup/ionize.pdb
../1_setup/ionize.pdb
../3_heating/run.rst

#
#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
reference for restraints
restart file

After the equilibration, it is always a good practice to check the convergence of quantities related to
thermodynamic conditions (such as temperature, pressure, or volume). This can be done in a similar
way described before.

14.5 Production simulation

Finally, we perform a production 1-nanosecond molecular dynamics simulation. Running such a relatively long simulation, we can analyze the stability of BPTI in solution.
Note: Depending on the machine specifications and the number of MPI processes used, this step could
take anywhere from a couple of hours to days. We can always check the progress by inspecting the
standard output of SPDYN.
The following commands perform a 1-nanosecond molecular dynamics simulation at 300 K and 1 atm
with SPDYN:
#
$
#
$
#
$

change to the production directory

cd tutorial/bpti/5_production/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform production with SPDYN by using 8 MPI processes
mpirun -np 8 spdyn run.inp | tee run.out

In [DYNAMICS] section of the control file, the total number of simulation steps is set as
nsteps=500000, and the time step is set as timestep=0.002 ps. This combination results in
the length of nsteps*time step = 1000 ps = 1 ns simulation. Since the output period for coordinates is crdout_period=500, we have the total of nsteps/crdout_period = 1000 snapshots
in the output trajectory.
[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period

=
=
=
=
=
=

LEAP
500000
0.002
500
500
500000

#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period

When the production simulation finishes, lets visualize the trajectory with VMD. From the visualization,
you may get some insights into the fluctuations of the protein structure. In the next step, we will quantify
the size of the fluctuations by analyzing the trajectory.

14.5. Production simulation

46

GENESIS User Guide, Release 1.0

$ vmd -psf ../1_setup/ionize.psf -dcd run.dcd

14.6 Analysis: RMSD calculation

In this step, we examine the stability of BPTI by analyzing the trajectory obtained in the previous production simulation. As a measure to examine the stability, we calculate mean square displacements
(RMSDs) from the reference structure. Roughly speaking, this measure monitors the size of the fluctuation from the reference structure. As the reference structure, we choose the crystal structure.
For the analysis, we use crd_convert program which is one of the post-processing programs in the
GENESIS package. crd_convert is an utility to extract various quantities from the trajectory The
following commands calculate the RMSDs from the trajectory of the 1-nanosecond simulation.
#
$
#
$
#
$

change to the analysis directory

cd tutorial/bpti/6_analysis/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform analysis with crd_convert
crd_convert run.inp | tee run.out

The control file for calculating the RMSDs is shown below. In [TRAJECTORY] section, we set the
input trajectory file as trjfile1=../5_production/run.dcd.
Note: crd_convert supports multiple input files, so we can specify another file as trjfile2=XXX.
In [SELECTION] section, we define a group (group1) of the C atoms of the protein, which are fitted
for the RMSD calculation. In the [FITTING] section, the fitting method is specified. In this case, the
translations TR and rotations ROT are allowed. Finally the RMSD output file (rmsfile=run.rms) is
set in [OUTPUT] section.
[INPUT]
psffile = ../1_setup/ionize.psf
pdbfile = ../1_setup/ionize.pdb

# protein structure file

# PDB file

[OUTPUT]
rmsfile = run.rms

# RMSD file

[TRAJECTORY]
trjfile1
md_step1
mdout_period1
ana_period1
trj_format
trj_type

=
=
=
=
=
=

[SELECTION]
group1

= backbone and an:CA

# selection group 1

[FITTING]
fitting_method
fitting_atom

= TR+ROT
= 1

# method
# atom group

../5_production/run.dcd
1000
1
1
DCD
COOR+BOX

14.6. Analysis: RMSD calculation

#
#
#
#
#
#

trajectory file
number of MD steps
MD output period
analysis period
(PDB/DCD)
(COOR/COOR+BOX)

47

GENESIS User Guide, Release 1.0

[OPTION]
check_only

= NO

# (YES/NO)

After running crd_convert, we get the RMSD file (run.rms). The first column of the file is for the
time step, and the second column is for the RMSD values in unit of Angstrom. This can be plotted with
gnuplot:
$ gnuplot
gnuplot> set xlabel "step"
gnuplot> set ylabel "RMSD [angstrom]"
gnuplot> plot "run.rms" w lp

The most of the RMSDs are lower than 1 Angstrom throughout the simulation, indicating that this
protein is very stable in solution.

14.6. Analysis: RMSD calculation

48

CHAPTER

FIFTEEN

TUTORIAL 2: SIMULATING WITH

GO-MODEL
Coarse-grained models are useful for long-term behavior of biomolecules with time scales (typically
microseconds to seconds) difficult to access by molecular dynamics simulations with atomistic models.
In this chapter, we provide a step-by-step tutorial for using a coarse-grained model (Go-model) with
GENESIS. Using a Go-model potential energy function developed by Karanicolas and Brooks [27], we
simulate folding/unfolding of protein G.
The input files of this chapter are in tutorial/go/ of the GENESIS package. Before you start,
please note the following things:
The production simulation may require a whole day.
VMD [17] is required for building the simulation system and the visualization of results.
gnuplot [45] is required for making plots.
PATH environment variable includes a directory with GENESIS binary.
If you are a Mac OSX user, it is recommended to make an alias for VMD:
$ alias vmd="/Applications/VMD\ 1.9.app/Contents/MacOS/startup.command"

15.1 Building a simulation system

The Go-model by Karanicolas and Brooks [27] [28] is based on a minimalistic representations of a
protein, where each amino acid residue is a bead located at the C atom position. The potential energy
function is given in Energy. Although the function is almost same as that of CHARMM force field,
there are several exceptions: (1) there are no electrostatic terms; (2) the vdW terms are divided into two
terms for native contacts and non-native contacts, where the native contacts have 12-10-6 type attractive
potential, while the non-native contacts have 12 type repulsive potential.
In this step, we generate parfile (parameter file) and topfile (topology file) for this Go-model using
the MMTSB web service [49]. Then, from generated files, we create psffile (PSF file) (required for
GENESIS simulations) using VMD.
To build a simulation system, we start from downloading a crystal (or NMR) structure of protein G from
the RCSB Protein Data Bank (PDB) [46]. You can download the PDB file via a web browser, or directly
by using curl command. The downloaded PDB file should be cleaned; the entries other than ATOM
lines for submission to the MMTSB web service.
49

GENESIS User Guide, Release 1.0

#
$
#
$
#
$

change to the set-up directory

cd tutorial/go/1_setup/
download the PDB file (PDB code 1PGB)
curl -o 1pgb.pdb http://www.rcsb.org/pdb/files/1pgb.pdb
"clean" the file by deleting the entries other than ATOM
grep -e "^ATOM" 1pgb.pdb >1pgb_edited.pdb

The submission to the MMTSB web service consists of three stages: (1) upload the cleaned PDB file
(1pgb_edited.pdb), (2) give a reference tag (i.e. 1pgb), and (3) enter your e-mail address. After
the submission, youll get an e-mail with a tar-ball file containing parfile and topfile.

After extracting the tar-ball file into the working directory, you can see parfile, topfile
(GO_1pgb.param and GO_1pgb.top, respectively) and other files.
# change to the set-up directory
$ cd tutorial/go/1_setup/
# extract the tar-ball file

15.1. Building a simulation system

50

GENESIS User Guide, Release 1.0

$ tar xvf /path/to/GO_1pgb.tar

A VMD script for building the Go-model is at tutorial/go/1_setup/setup.tcl. We can run

it with VMD:
$ vmd -dispdev text <setup.tcl | tee run.out

Here, with -dispdev text option, we inhibits any graphical display windows from opening (we do
not need any visualization for this set-up). The redirection <setup.tcl instructs VMD to process the
script. The content of the script, written in Tcl languages, is shown below.
The script consists of three parts. In the first part, the PDB file created by the MMTSB web service is
read, and the molecule is moved as the center of mass is at the origin. In the second part, residue names
are replaced with special names for the Go-model (G1, G2, G3,...). These replacements are required to
match the residue names to those defined in topfile (GO_1pgb.param). In the third part, psfgen plugin
is called, and psffile (go.pdb) and pdbfile (go.pdb) are generated.
##### read pdb and remove center of mass
mol load pdb GO_1pgb.pdb
##### replace residue names with G1, G2, G3, ...
set all [atomselect top all]
set residue_list [lsort -unique [$all get resid]]
foreach i $residue_list {
set resname_go [format "G%d" $i]
set res [atomselect top "resid $i" frame all]
$res set resname $resname_go
}
$all writepdb tmp.pdb
##### generate PSF and PDB files
package require psfgen
resetpsf
topology GO_1pgb.top
segment PROT {
first none
last none
pdb tmp.pdb
}
regenerate angles dihedrals
coordpdb tmp.pdb PROT
# invert the center of mass
$all moveby [vecinvert [measure center $all weight mass]]
$all moveby [vecinvert [measure center $all]]
# write psf and pdb files
writepsf go.psf
writepdb go.pdb
exit

It is always a good idea to inspect the final structure visually. We can visualize the structure with VMD
15.1. Building a simulation system

51

GENESIS User Guide, Release 1.0

by specifying psffile (go.psf) and pdbfile (go.pdb) together:

$ vmd -psf go.psf -pdb go.pdb

If the building has successfully finished, the structure looks like this:

15.2 Production simulation

Contrary to atomistic model simulations, coarse-grained simulations are not so sensitive to the initial
structure. Thus, we do not need to relax the system (as in the case of BPTI in the previous chapter)
before a production simulation. So, in this step, we are going to perform a production simulation without
minimization or equilibration.
The following command performs a 5*108 step production simulation with ATDYN:
#
$
#
$
#

change to the production directory

cd tutorial/go/2_production/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform production simulation with ATDYN by using 8 MPI processes

15.2. Production simulation

52

GENESIS User Guide, Release 1.0

$ mpirun -np 8 atdyn run.inp | tee run.out

The control file (run.inp, shown below) is different from those for atomistic simulations.
The control file (run.inp) contains several sections, such as [INPUT], [OUTPUT], and [ENERGY],
where we can specify the control parameters for the simulation. In [INPUT] section, topfile (topology
file), parfile (parameter file), psffile (PSF file), pdbfile (the initial structure) are set (see Input and Output
files for an explanation of each input file).
In [OUTPUT] section, output files are set. ATDYN does not create any output file unless we explicitly
specify the files. Here, rstfile (restart file) and dcdfile (binary trajectory file) are set (see Input and Output
files for an explanation of each output file).
In [ENERGY] section, we can specify the parameters related to the energy and force evaluation. Here,
KBGO is specfied for the Go-model of Karanicolas and Brooks (forcefield=KBGO). This value turns
on the special 12-10-6 type vdW interactions for native contacts. For the distances, very large values
are set (switchdist=997, cutoffdist=998, pairlistdist=999) to perform a non-cutoff
simulation.
[DYNAMICS] section turns on the molecular dynamics engine of ATDYN. For the Go-model with
SHAKE constraints, time step can be 20 fs.
In [CONSTRAINTS] section, we enable SHAKE algorithm on all bonded pairs (rigid_bond=YES).
To supress SETTLE algorithm applied for non-existent TIP3P water molecules, we have to disable it
explicitly (fast_water=NO). The tolerance for SHAKE is rather large compared to atomistic simulations (shake_tolerance=1.0e-6).
In [ENSEMBLE] section, LANGEVIN thermostat is chosen for an isothermal simulation with the friction constant of 1.0 ps-1 .
Finally, in [ENSEMBLE] section, no boundary condition is imposed in this system.
[INPUT]
topfile
parfile
psffile
pdbfile

=
=
=
=

../1_setup/GO_1pgb.top
../1_setup/GO_1pgb.param
../1_setup/go.psf
../1_setup/go.pdb

[OUTPUT]
dcdfile = run.dcd
rstfile = run.rst

#
#
#
#

topology file
parameter file
protein structure file
PDB file

# DCD trajectory file

# restart file

[ENERGY]
forcefield
electrostatic
switchdist
cutoffdist
pairlistdist
table

=
=
=
=
=
=

KBGO
CUTOFF
997.0
998.0
999.0
NO

#
#
#
#

switch distance
cutoff distance
pair-list cutoff distance
usage of lookup table

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
rstout_period
crdout_period

=
=
=
=
=
=

LEAP
100000000
0.020
10000
10000
10000

#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
timestep (ps)
energy output period
restart output period
coordinates output period

15.2. Production simulation

53

GENESIS User Guide, Release 1.0

nbupdate_period = 10000
[CONSTRAINTS]
rigid_bond

= YES

fast_water
= NO
shake_tolerance = 1.0e-6
[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 325

gamma_t

= 0.01

[BOUNDARY]
type

= NOBC

# nonbond update period

#
#
#
#

constraints all bonds

involving hydrogen
settle constraint
tolerance (Angstrom)

#
#
#
#
#
#

[NVE,NVT,NPT]
thermostat
initial and target
temperature (K)
thermostat friction (ps-1)
in [LANGEVIN]

# [PBC, NOBC]

15.3 Analysis: RMSD calculation

For the analysis, we use crd_convert program which is one of the post-processing programs in the GENESIS package. crd_convert is a utility to derive various quantities from the trajectory. The following
commands calculate the RMSDs from the trajectory of the simulation.
#
$
#
$
#
$

change to the analysis directory

cd tutorial/go/3_analysis/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform analysis with crd_convert
crd_convert run.inp | tee run.out

The control file for calculating the RMSDs is shown below. In [TRAJECTORY] section, we set the
input trajectory file as trjfile1=../2_production/run.dcd.
Note: crd_convert supports multiple input files, so we can specify another file as trjfile2=XXX.
In [SELECTION] section, we define a group (group1) of all beads in the model which are fitted for the RMSD calculation. In the [FITTING] section, the fitting method is specified. In this
case, the translations TR and rotations ROT are allowed for the fitting. Finally the RMSD output file
(rmsfile=run.rms) is set in [OUTPUT] section.
[INPUT]
psffile = ../1_setup/go.psf
reffile = ../1_setup/go.pdb

# protein structure file

# PDB file

[OUTPUT]
rmsfile = run.rms

# RMSD file

[TRAJECTORY]
trjfile1
md_step1

# trajectory file
# number of MD steps

= ../2_production/run.dcd
= 100000000

15.3. Analysis: RMSD calculation

54

GENESIS User Guide, Release 1.0

mdout_period1
ana_period1
trj_format
trj_type

=
=
=
=

10000
1
DCD
COOR

#
#
#
#

MD output period
analysis period
(PDB/DCD)
(COOR/COOR+BOX)

[SELECTION]
group1

= all

# selection group 001

[FITTING]
fitting_method
fitting_atom

= TR+ROT
= 1

# method
# atom group

[OPTION]
check_only

= NO

# (YES/NO)

After running crd_convert, we get the RMSD file (run.rms). The first column of the file is for the
time step, and the second column is for the RMSD values in unit of Angstrom. This can be plotted with
gnuplot:
$ gnuplot
gnuplot> set xlabel "step"
gnuplot> set ylabel "RMSD [angstrom]"
gnuplot> plot "run.rms" w lp

The noticeable fluctuations of RMSD values correspond to folding/unfolding events of the protein.

15.3. Analysis: RMSD calculation

55

CHAPTER

SIXTEEN

TUTORIAL 3: REMD SIMULATION OF

ALANINE DIPEPTIDE
In this chapter, a tutorial for a replica-exchange molecular dynamics (REMD) simulation with GENESIS is provided. The aim is to introduce a new GENESIS user how to perform REMD simulation
of alanine dipeptide (ALAD) in water. ALAD has often been used as a model in theoretical studies
of backbone conformational analyses. Usage of the REMD function in GENESIS is basically same
between ATDYN and SPDYN. Hereafter, we show an example for ATDTN.
The input files of this chapter are in tutorial/alad_remd/ of the GENESIS package. Before you
start the tutorial, please note the following things:
Running all of the simulations requires several days.
VMD [17] is required for building a simulation system and the visualization of results.
gnuplot [45] is required for making plots.
PATH environment variable includes a directory with GENESIS binary.
If you are a Mac OSX user, it is recommended to make an alias for VMD:
$ alias vmd="/Applications/VMD\ 1.9.app/Contents/MacOS/startup.command"

tutorial/alad_remd/1_setup directory includes topfile (topology file), parfile (parameter file),

psffile (PSF file), pdbfile (initial structure) required for the REMD simulation of ALAD.

16.1 Minimization
As the initial structures often contain non-physical steric clashes or non-equilibrium geometries, it is
recommended to relax the system before the REMD simulation by minimizing the potential energy of
the system. ATDYN and SPDYN support the steepest descent algorithm for minimization, which moves
atoms proportionally to their negative gradients of the potential energy.
The following commands perform a 1000-step minimization with ATDYN.
#
$
#
$

change to the minimization directory

cd tutorial/alad_remd/2_minimization/
set the number of OpenMP threads
export OMP_NUM_THREADS=1

56

GENESIS User Guide, Release 1.0

# perform minimization with ATDYN by using 8 MPI processes

$ mpirun -np 8 atdyn remd_min.inp | tee remd_min.out

The control file (remd_min.inp, shown below) consists of several sections, such as [INPUT], [OUTPUT], and [ENERGY], etc., where we can specify the control parameters for the simulation.
In [INPUT] section, input files for the minimization are specified. topfile (topology file), parfile (parameter file), psffile (PSF file), pdbfile (initial structure) are always required. An optional input, reffile
(reference file), is the reference structure for positional restraints (see Input and Output files for an
explanation about each input file).
In [OUTPUT] section, output files are specified. ATDYN does not create any output file unless we
explicitly specify the files. Here, rstfile (restart file) is used for the restart of the simulation (see Input
and Output files for an explanation about each output file).
In [ENERGY] section, we set the keyword related to the energy and force evaluation. Here, the particle
mesh ewald (PME) method is selected for computing electrostatic interactions, usually combined with the
periodic boundary condition (PBC) in [BOUNDARY] section. table_order=0 is specified for nonbonded interactions (by default, an interpolation scheme with the lookup table is used for the evaluations
of non-bonded interactions). When table_order=0, the forces values of the nearest grid point in the
table are used (without any interpolations), and thus large forces which may occur at the beginning of
minimization can be safely truncated. See Energy for details.
[MINIMIZE] section turns on the minimization engine of ATDYN. Here, we specify 1000 steps of the
steepest descent algorithm (SD). See Minimize for details.
In [BOUNDARY] section, the boundary condition and the simulation box size are set. Here, we use the
periodic boundary condition (PBC). See Boundary for details.
In [SELECTION] section, we define a group (group1) consisting of backbone atoms of the protein to
impose positional restraints. For the detailed expressions for selection, see Selection.
In [RESTRAINTS] section, positional restraints are specified for the group of backbone atoms defined
in the previous [SELECTION] section. The positional restraints are imposed with respect to the reference coordinates given by reffile in [INPUT] section. See Restraints for details.
[INPUT]
topfile
parfile
psffile
pdbfile
reffile

=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ala.psf
../1_setup/ala.pdb
../1_setup/ala.pdb

#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
reference for restraints

[OUTPUT]
dcdfile = ./remd_min.dcd
rstfile = ./remd_min.rst

# DCD trajectory file

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table
table_order
pme_ngrid_x
pme_ngrid_y

#
#
#
#
#
#
#
#

16.1. Minimization

=
=
=
=
=
=
=
=

PME
7.5
8.0
9.0
YES
1
64
64

[CUTOFF,PME]
switch distance
cutoff distance
pair-list distance
usage of lookup table
order of lookup table
grid size_x in [PME]
grid size_y in [PME]

57

GENESIS User Guide, Release 1.0

pme_ngrid_z

= 64

# grid size_z in [PME]

[MINIMIZE]
nsteps
eneout_period
crdout_period
rstout_period
nbupdate_period

=
=
=
=
=

1000
10
10
1000
5

#
#
#
#
#

number of steps
energy output period
coordinates output period
restart output period
nonbond update period

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z

=
=
=
=

PBC
64.0
64.0
64.0

#
#
#
#

[PBC,NOBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

1
POSI
10.0
1

number of functions
restraint function type
force constant
restraint group

16.2 Equilibration
In this step, we equilibrate the system under the condition of 300 K and 1 atm by a 10-picosecond
molecular dynamics simulation. In the first equilibration (eq1), positional restraints are imposed on the
backbone, and at the second equilibration (eq2) all the restraints are removed to relax the whole system
for the production simulation.
The following commands perform the first 10-picosecond NPT molecular dynamics simulation under
the condition of 300 K and 1 atm with restraints (eq1).
#
$
#
$
#
$

change to the equilibration directory

cd tutorial/alad_remd/3_equilibration/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform equilibration with ATDYN by using 8 MPI processes
mpirun -np 8 atdyn remd_eq1.inp | tee remd_eq1.out

In [INPUT] section of the control file (remd_eq1.inp), rstfile is set to the restart file from the previous
minimization.
The main differences from the previous minimization step-up are [DYNAMICS], [CONSTRAINTS],
and [ENSEMBLE] sections, which contain the keywords for molecular dynamics simulations.
[DYNAMICS] section enables the molecular dynamics engine of ATDYN. Parameters related to molecular dynamics integrator are specified in this section. See Dynamics for details.
[CONSTRAINTS] section specifies constraints during a simulation. rigid_bond is a keyword for
the SHAKE algorithm. For TIP3P waters, a more fast algorithm (SETTLE) is automatically applied if
rigid_bond=YES. See Constraints for details.

16.2. Equilibration

58

GENESIS User Guide, Release 1.0

In [ENSEMBLE] section, we can choose a thermostat and a barostat from several standard options. For
typical simulations, LANGEVIN or BERENDSEN is recommended. See Ensemble for details.
In [BOUNDARY] section, we no longer need to specify the simulation box size, as the box size are read
from the restart file.
[INPUT]
topfile
parfile
psffile
pdbfile
rstfile
reffile

=
=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ala.psf
../1_setup/ala.pdb
../2_minimization/remd_min.rst
../1_setup/ala.pdb

#
#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
restart file
reference for restraints

[OUTPUT]
dcdfile = ./remd_eq1.dcd
rstfile = ./remd_eq1.rst

# DCD trajectory file

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=
=

PME
7.5
8.0
9.0
YES
1
64
64
64

#
#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list distance
usage of lookup table
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period
nbupdate_period

=
=
=
=
=
=
=

LEAP
5000
0.002
50
50
5000
5

#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period
nonbond update period

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NPT
= LANGEVIN
= 300.0

pressure

= 1.0

#
#
#
#
#

[BOUNDARY]
type

= PBC

# [PBC,NOBC]

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1

= 1
= POSI

# number of functions
# restraint function type

16.2. Equilibration

[NVE,NVT,NPT]
thermostat and barostat
initial and target
temperature (K)
target pressure (atm)

59

GENESIS User Guide, Release 1.0

constant1
select_index1

= 5.0
= 1

# force constant
# restraint group

In the second equilibration, [SELECTION] and [RESTRAINTS] sections in the control file are removed, as we perform a restraint-free simulation (eq2). Most of the control parameters are the same to
the previous equilibration (eq1) except rstfile, which is set to the restart file of the first equilibration.
[INPUT]
topfile
parfile
psffile
pdbfile
rstfile
reffile

=
=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ala.psf
../1_setup/ala.pdb
./remd_eq1.rst
../1_setup/ala.pdb

#
#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
restart file
reference for restraints

[OUTPUT]
dcdfile = ./remd_eq2.dcd
rstfile = ./remd_eq2.rst

# DCD trajectory file

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=
=

PME
7.5
8.0
9.0
YES
1
64
64
64

#
#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list distance
usage of lookup table
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period
nbupdate_period

=
=
=
=
=
=
=

LEAP
5000
0.002
50
50
5000
5

#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period
nonbond update period

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NPT
= LANGEVIN
= 300.0

pressure

= 1.0

#
#
#
#
#

[BOUNDARY]
type

= PBC

# [PBC]

[NVE,NVT,NPT]
thermostat and barostat
initial and target
temperature (K)
target pressure (atm)

Also, it is recommended to visualize the trajectory to confirm, the protein structure is not disrupted. We
can visualize the trajectory with VMD by the following command:

16.2. Equilibration

60

GENESIS User Guide, Release 1.0

$ vmd -psf ../1_setup/ala.psf -dcd remd_eq2.dcd

Note: For real applications, 10 ps may be too short for equilibration. For your research, longer equilibration than 1 ns is recommended.
After the equilibration, it is always a good practice to check the convergence of quantities related to
thermodynamic conditions (such as temperature, pressure, or volume). This can be done in a similar
way described before.

16.3 Replica temperatures

To perform REMD simulation we need to determine the number of replicas and their temperatures.
These parameters have to be chosen carefully, as their are one of the most important factors affecting the
results of REMD simulation.
For this purpose we recommend to use Temperature generator for REMD-simulations [50]. This web
server automatically generate number of replicas and their temperatures according to provided information.

16.4 Production simulation

Before performing REMD simulation, the system needs to be equilibrated at each temperature. The
following commands perform a 20-picosecond NVT molecular dynamics simulations.
# change to the production directory
$ cd tutorial/alad_remd/4_production/
# perform equilibration at each temperature with ATDYN by
submitting batch job script
$ qsub remd_run_eq.sh

The batch job script remd_run_eq.sh is at tutorial/alad_remd/4_production/ for reference.

In [OUTPUT] section of the control file, logfile gives log file for each replica and remfile gives replica
exchange parameter ID file. In this section {} returns a series of 32 output files (number of replicas).
In [REMD] section of the control file, the number of dimensions is set to dimension=1, and the period
between replica exchanges is set to exchange_period=0 for equilibration (no need to be exchanged
in this step). Type of exchanged variable in each dimension is set to type(1)=TEMPERATURE, the
number of replicas in each dimension is set to nreplica(1)=32, and the variables in each dimension are set to parameters(1)=300 301 ... 331. cyclic_params(1)=NO determines if
ATDYN performs replica exchanges cyclically or not. See REMD for details.
In [DYNAMICS] section of the control file, the total number of simulation steps is set to
nsteps=10000, and the time step is set to timestep=0.002 ps. This results into a
nstep*timestep = 20 ps simulation. Since the output period for coordinates is set to
crdout_period=50, we have the total of nsteps/crdout_period=200 snapshots in the output trajectory.

16.3. Replica temperatures

61

GENESIS User Guide, Release 1.0

[INPUT]
topfile
parfile
psffile
pdbfile
rstfile

=
=
=
=
=

#
#
#
#
#

remd_run_eq{}.log
remd_run_eq{}.dcd
remd_run_eq{}.rst
remd_run_eq{}.rem

log file of each replica

DCD trajectory file
restart file
replica exchange ID file

[OUTPUT]
logfile =
dcdfile =
rstfile =
remfile =

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ala.psf
../1_setup/ala.pdb
../3_equilibration/remd_eq2.rst

#
#
#
#

topology file
parameter file
protein structure file
PDB file
restart file

[REMD]
dimension
= 1
exchange_period = 0
type(1)
nreplica(1)
parameters(1)

= TEMPERATURE
= 32
= 300 301 302 303 304 305 306 307 308 309 310 311 312 \
313 314 315 316 317 318 319 320 321 322 323 324 325 \
326 327 328 329 330 331
cyclic_params(1)= NO
[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
7.5
8.0
9.0
1
64
64
64

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period
nbupdate_period

=
=
=
=
=
=
=

LEAP
10000
0.002
50
50
10000
5

#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period
nonbond update period

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 300.0

#
#
#
#

[BOUNDARY]
type
box_size_x
box_size_y

= PBC
= 64.0
= 64.0

# [NOBC,PBC]
# box size (x) in [PBC]
# box size (y) in [PBC]

16.4. Production simulation

[NVE,NVT,NPT]
thermostat
initial and target
temperature (K)

62

GENESIS User Guide, Release 1.0

box_size_z

= 64.0

# box size (z) in [PBC]

After the equilibrating the system, we move on to the production simulation of REMD. The following
commands perform 1-nanosecond REMD simulations in NVT ensemble:
# perform production simulation of REMD with ATDYN by submitting a
batch job script
$ qsub remd_run.sh

The batch job script remd_run.sh is at tutorial/alad_remd/4_production/ for reference.

Differences from the previous equilibration step are [INPUT], [REMD] and [DYNAMICS] sections.
In [INPUT] section of the control file (remd_run.inp), rstfile is set to the restart from the equilibration. In this section {} also returns a series of 32 input files.
In [REMD] section of the control file, the period between replica exchanges is set to
exchange_period=1000. See REMD for details. You can break a long line into multiline using
backslash.
In [DYNAMICS] section of the control file, the total number of simulation steps is set
tonsteps=1500000, and the time step is set to timestep=0.002 ps. This results into
a nstep*timestep = 3 ns simulation. Since the output period for coordinates is set to
crdout_period=1000, we have the total of nsteps/crdout_period=1500 snapshots in the
output trajectory.
[INPUT]
topfile
parfile
psffile
pdbfile
rstfile

=
=
=
=
=

[OUTPUT]
logfile =
dcdfile =
rstfile =
remfile =

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ala.psf
../1_setup/ala.pdb
./remd_run_eq{}.rst

#
#
#
#
#

remd_run{}.log
remd_run{}.dcd
remd_run{}.rst
remd_run{}.rem

log file of each replica

DCD trajectory file
restart file
replica exchange ID file

[REMD]
dimension
exchange_period
type(1)
nreplica(1)
parameters(1)

=
=
=
=
=

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x

=
=
=
=
=
=

#
#
#
#

topology file
parameter file
protein structure file
PDB file
restart file

1
1000
TEMPERATURE
32
300 301 302 303 304 305 306 307 308 309 310 311 312 \
313 314 315 316 317 318 319 320 321 322 323 324 325 \
326 327 328 329 330 331
cyclic_params(1)= NO

PME
7.5
8.0
9.0
1
64

16.4. Production simulation

#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list distance
order of lookup table
grid size_x in [PME]

63

GENESIS User Guide, Release 1.0

pme_ngrid_y
pme_ngrid_z

= 64
= 64

# grid size_y in [PME]

# grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
nbupdate_period
rstout_period

=
=
=
=
=
=
=

#
#
#
#
#
#
#

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 300.0

#
#
#
#

[NVE,NVT,NPT]
thermostat
initial and target
temperature (K)

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z

=
=
=
=

#
#
#
#

[NOBC,PBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

LEAP
1500000
0.002
1000
1000
5
1500000

PBC
64.0
64.0
64.0

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
nonbond update period
restart output period

16.5 Analysis of REMD simulation

In this step, we examine the performance of REMD simulation by analyzing the standard output file,
logfile (.log) and dcdfile (.dcd). We calculate (1) a time series of replica exchange, (2) a time series of
temperature exchange of three arbitrary chosen replicas, and (3) a time series of total potential energy
of three arbitrary chosen replicas.
First of all, we use remd_convert program which is one of the post-processing programs in the GENESIS
package. remd_convert is an utility to extract various quantities from a REMD trajectory:
#
$
#
$

change to the analysis directory

cd tutorial/alad_remd/5_analysis/
process the REMD trajectory with remd_convert
remd_convert Inp_remd_conv | tee Inp_remd_conv.log

The control file for remd_convert (Inp_remd_conv) are shown below. In [INPUT] section, we specify
the output files (dcdfile and remfile) from the REMD simulation. In [SELECTION] section, we
define a group (group1) of the heavy atoms of the ALAD which are fitted. In the [FITTING] section,
the fitting method is set. In this case, the translations TR and rotations ROT are allowed for the fitting.
In the [OPTION] section, the type of conversion is set to convert_type=PARAMETER, and the
indices of the type to be converted are set to convert_ids=1 17 32. Period of the dcdfile
trajectory and the output format are set to dcd_md_period=1000 and trjout_format=DCD,
respectively. The selection of output atoms is set to trjout_atom=1, and a periodic boundary correction is set to pbc_correct=MOLECULE.
16.5. Analysis of REMD simulation

64

GENESIS User Guide, Release 1.0

[INPUT]
psffile
reffile
dcdfile
remfile

=
=
=
=

../1_setup/ala.psf
../1_setup/ala.pdb
../4_production/remd_run{}.dcd
../4_production/remd_run{}.rem

[OUTPUT]
pdbfile = ./remd_run_param.pdb
trjfile = ./remd_run_param{}.trj

# protein structure file

# PDB file

# PDB file
# trajectory file

[SELECTION]
group1
mole_name1

= molname:alad and heavy

# selection group 1
= alad PROT:1:ALAD PROT:1:ALAD

[FITTING]
fitting_method
fitting_atom
zrot_ngrid
zrot_grid_size

=
=
=
=

TR+ROT
1
10
1.0

#
#
#
#

[OPTION]
check_only
convert_type
convert_ids
dcd_md_period
trjout_format
trjout_type
trjout_atom
pbc_correct

=
=
=
=
=
=
=
=

NO
PARAMETER
1 17 32
1000
DCD
COOR
1
MOLECULE

# (YES/NO)
# (REPLICA/PARAMETER)
#
#
#
#
#

method
atom group
number of z-rot grids
z-rot grid size

input dcdfile MD period

(PDB/DCD)
(COOR/COOR+BOX)
selection group output
(NO/MOLECULE)

To examine (1) the time series of replica exchange, we use the standard output file of the REMD simulation to extract time series of ReplicaID:
$

grep ReplicaID : ./remd_run.out |

awk / ReplicaID :/{print $3, $19, $34}> ./remd_run.replica
$ gnuplot
gnuplot> set xlabel "Time (ns)"
gnuplot> set ylabel "Replica"
gnuplot> plot "remd_run.replica" u ($0*0.002):1 w p pt 7 ps 0.5 lt 1
title "300K"
gnuplot> plot "remd_run.replica" u ($0*0.002):2 w p pt 7 ps 0.5 lt 2
title "316K"
gnuplot> plot "remd_run.replica" u ($0*0.002):3 w p pt 7 ps 0.5 lt 3
title "331K"

16.5. Analysis of REMD simulation

65

GENESIS User Guide, Release 1.0

16.5. Analysis of REMD simulation

66

GENESIS User Guide, Release 1.0

We can observed random walks in replica space at parameterID=1 (300 K), 17 (316 K) and 32 (331
K) colored by red, green and blue, respectively.
To examine (2) the time series of temperature exchange of three arbitrary chosen replicas (replica id = 1,
17 and 32), we use the standard output file of the REMD simulation to extract time series of Parameter:
$ grep Parameter : ./remd_run.param |
awk / Parameter :/{print $3, $19, $34}>
$ gnuplot
gnuplot> set xlabel "Time (ns)"
gnuplot> set ylabel "Temperature [K]"
gnuplot> plot "remd_run.param" u ($0*0.002):1
title "replicaID=1",\
"remd_run.param" u ($0*0.002):2
title "replicaID=17",\
"remd_run.param" u ($0*0.002):3
title "replicaID=32"

16.5. Analysis of REMD simulation

./remd_run.param

w l lw 2 lt 1
w l lw 2 lt 2
w l lw 2 lt 3

67

GENESIS User Guide, Release 1.0

To examine (3) the time series of total potential energy of three arbitrary chosen replicas (replica id = 1,
17 and 32), we use the logfile of the REMD simulation to extract time series of POTENTIAL_ENE:
$ grep INFO: remd_run01.log | tail -n +2 | awk {print
./remd_run01.ene
$ grep INFO: remd_run17.log | tail -n +2 | awk {print
./remd_run17.ene
$ grep INFO: remd_run32.log | tail -n +2 | awk {print
./remd_run32.ene
$ gnuplot
gnuplot> set xlabel "Time (ns)"
gnuplot> set ylabel "Potential energy (kcal/mol)"
gnuplot> plot "remd_run01.ene" u ($0*0.002):1 w l lw 1.5
title "replicaID=1",\
"remd_run17.ene" u ($0*0.002):1 w l lw 1.5
title "replicaID=17",\
"remd_run32.ene" u ($0*0.002):1 w l lw 1.5
title "replicaID=32"

16.5. Analysis of REMD simulation

$5}>
$5}>
$5}>

lt 1
lt 1
lt 1

68

GENESIS User Guide, Release 1.0

We can observe random walks in temperature and potential energy spaces at replicaID=1, 17 and
32, colored by red, green and blue, respectively. These results indicate that the REMD simulation
performed satisfactorily.
Finally we plot PMF (Potential of Mean Force) surface versus two torsion angles: CLP-NL-CA-CRP
() and NL-CA-CRP-NR ().
#
$
#
$

generate ctrl file for trj_analysis to calculate torsion angles

trj_analysis -h ctrl > Inp_remd_torsion
calculate torsion angles with trj_analysis
trj_analysis Inp_remd_torsion | tee Inp_remd_torsion.log

The control file for trj_analysis (Inp_remd_torsion) are shown below. In [OPTION] section, we specify
four atom names which are consisting torsion angle. Each atom name is defined using the segment name
and the residue name.
[INPUT]
psffile = ../1_setup/alad.psf
reffile = ../1_setup/alad.pdb

# protein structure file

# PDB file

[OUTPUT]
torfile = ./remd_run_param01.tor

# torsion file

[TRAJECTORY]
trjfile1
md_step1
mdout_period1
ana_period1
repeat1

=
=
=
=
=

trj_format
trj_type
trj_natom

= DCD
= COOR
= 10

./remd_run_param01.trj
1500000
1000
1
1

#
#
#
#

trajectory file
number of MD steps
MD output period
analysis period

#
#
#
#

(PDB/DCD)
(COOR/COOR+BOX)
number of atoms
in trajectory file

[OPTION]

16.5. Analysis of REMD simulation

69

GENESIS User Guide, Release 1.0

check_only
torsion1
torsion2

= NO
= PROT:1:ALAD:CLP
PROT:1:ALAD:CA
= PROT:1:ALAD:NL
PROT:1:ALAD:CRP

16.5. Analysis of REMD simulation

# (YES/NO)
PROT:1:ALAD:NL \
PROT:1:ALAD:CRP
PROT:1:ALAD:CA \
PROT:1:ALAD:NR

70

CHAPTER

SEVENTEEN

TUTORIAL 4: REUS SIMULATION OF

ALANINE DIPEPTIDE
In this chapter, a tutorial for a replica-exchange umbrella sampling molecular dynamics (REUS) simulation with GENESIS is provided. The aim is to introduce a new GENESIS user how to perform REUS
simulation of alanine dipeptide (ALAD) in water. We recommend, the new user should try Tutorial
3: REMD Simulation of Alanine Dipeptide before, as the procedures of performing the simulation are
simplier.
The input files of this chapter are in tutorial/alad_reus/ of the GENESIS package. Before you
start the tutorial, please note the following things:
Running all of the simulations requires several days.
VMD [17] is required for building a simulation system and the visualization of results.
gnuplot [45] is required for making plots.
PATH environment variable includes a directory with GENESIS binary.
If you are a Mac OSX user, it is recommended to make an alias for VMD:
$ alias vmd="/Applications/VMD\ 1.9.app/Contents/MacOS/startup.command"

tutorial/alad_reus/1_setup directory includes topfile (topology file), parfile (parameter file),

psffile (PSF file), pdbfile (initial structure) required for the REUS simulation of ALAD.
We skip Minimization and Equilibration sections required to perform before the REUS simulation, as
these procedures are already described in Tutorial 3: REMD Simulation of Alanine Dipeptide. Please
proceed to the next section after completing minimization and equilibration of your system as in Tutorial
3: REMD Simulation of Alanine Dipeptide.

17.1 Replica temperatures and umbrella potentials

To perform REUS simulation we need to set the type of exchanged variables. In this case we use
temperature and restraint potentials of dihedral angle.
1. To choose replica temperatures, we recommend to use Temperature generator for REMDsimulations [50]. This web server automatically generate number of replicas and their temperatures according to provided information. In this case, we use four temperatures of 300K, 301K,
302K and 303K.

71

GENESIS User Guide, Release 1.0

2. To choose umbrella potentials, a user needs to find appropriate reaction coordinates describing the
conformational changes of the system. ALAD is a common model for backbone conformational
analyses, so we use dihedral angles as the reaction coordinates.

17.2 Production simulation

Before performing REUS simulation, the system needs to be equilibrated at each temperature and each
dihedral restraint. The following commands perform s 20-picosecond NVT molecular dynamics simulations.
# change to the production directory
$ cd tutorial/alad_reus/4_production/
# perform equilibration at each temperature and each dihedral restraint
with ATDYN by submitting batch job script
$ qsub reus_run_eq.sh

The batch job script reus_run_eq.sh is at tutorial/alad_reus/4_production/ for reference.

In [OUTPUT] section of the control file, logfile gives log file for each replica and remfile gives replica
exchange parameter ID file. In this section {} returns a series of 16 output files (product of number of
replicas in each dimension).
In [REMD] section of the control file, the number of dimensions is set to dimension=2, and the period
between replica exchanges is set to exchange_period=0 for equilibration (no need to be exchanged
in this step). Type of exchanged variable in each dimension is set to type(1)=TEMPERATURE and
type(2)=RESTRAINT , the number of replicas in each dimension is set to nreplica(1)=4 and
nreplica(2)=4. The variables in each dimension are set to parameters(1)=300 301 302
303 and parameters(2)= (1.0 144.0) (1.1 146.0) (1.2 148.0) (1.3 150.0).
For type=RESTRAINT, force constants and reference values should be set in the parentheses.
cyclic_params(1)=NO determines whether ATDYN performs replica exchanges cyclically. See
REMD for details.
In [DYNAMICS] section of the control file, the total number of simulation steps is set to
nsteps=10000, and the time step is set to timestep=0.002 ps. This results into a
nstep*timestep = 20 ps simulation. Since the output period for coordinates is set to
crdout_period=50, we have the total of nsteps/crdout_period=200 snapshots in the output trajectory.
In [SELECTION] section of the control file, the groups of selected atom(s) used in [REMD] and
[RESTRAINTS] sections are set. The available expressions in group keyword are listed in Selection.
In this case, atom names NL, CA, CRP and NR of ala.pdb are selected for each group.
In [RESTRAINTS] section of the control file, there are keywords for external harmonic restraint functions. The restraint functions are applied to the selected atoms in the [SELECTION] section. In this
case, a restraint potential with constant1=1.0 and reference1=144.0 is applied to dihedral
angle NL-CA-CRP-NR of ALAD.
[INPUT]
topfile
parfile
psffile
pdbfile
rstfile

=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf # topology file

../1_setup/par_all27_prot_lipid.prm # parameter file
../1_setup/ala.psf
# protein structure file
../1_setup/ala.pdb
# PDB file
../3_equilibration/reus_eq2.rst
# restart file

17.2. Production simulation

72

GENESIS User Guide, Release 1.0

reffile = ../1_setup/ala.pdb
[OUTPUT]
logfile =
dcdfile =
rstfile =
remfile =

reus_run_eq{}.log
reus_run_eq{}.dcd
reus_run_eq{}.rst
reus_run_eq{}.rem

# reference for restraints

#
#
#
#

log file of each replica

DCD trajectory file
restart file
replica exchange ID file

[REMD]
dimension
= 2
exchange_period = 0
type(1)
=
nreplica(1)
=
parameters(1)
=
cyclic_params(1)=

TEMPERATURE
4
300 301 302 303
NO

type(2)
=
nreplica(2)
=
parameters(2)
=
cyclic_params(2)=
rest_function(2)=

RESTRAINT
4
(1.0 144.0) (1.1 146.0) (1.2 148.0) (1.3 150.0)
NO
1

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
7.5
8.0
9.0
1
64
64
64

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
nbupdate_period
rstout_period

=
=
=
=
=
=
=

LEAP
10000
0.002
50
50
5
10000

#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
nonbond update period
restart output period

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 300.0

#
#
#
#

[NVE,NVT,NPT]
thermostat
initial and target
temperature (K)

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z

=
=
=
=

#
#
#
#

[NOBC,PBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

PBC
64.0
64.0
64.0

17.2. Production simulation

73

GENESIS User Guide, Release 1.0

[SELECTION]
group1
group2
group3
group4

=
=
=
=

an:NL
an:CA
an:CRP
an:NR

#
#
#
#

restraint
restraint
restraint
restraint

[RESTRAINTS]
nfunctions
function1
constant1
reference1
select_index1

=
=
=
=
=

1
DIHED
1.0
144.0
1 2 3 4

#
#
#
#
#

number of functions
restraint function type
force constant
reference
restraint groups

group
group
group
group

1
2
3
4

After equilibrating the system, we move on to the production simulation of REUS. The following commands perform 6-nanosecond REUS simulations in NVT ensemble:
# perform production simulation of REUS with ATDYN by submitting a
batch job script
$ qsub reus_run.sh

The batch job script reus_run.sh is at tutorial/alad_reus/4_production/ for reference.

Differences from the previous equilibration step are [INPUT], [REUS] and [DYNAMICS] sections.
In [INPUT] section of the control file (reus_run.inp), rstfile is set to the restart file from the equilibration. In this section {16} also returns a series of 16 input files.
In [REUS] section of the control file, the period between replica exchanges is set to
exchange_period=1000.
In [DYNAMICS] section of the control file, the total number of simulation step is set to
nsteps=3000000, and the time step is set to timestep=0.002 ps. This results into
a nstep*timestep = 6 ns simulation. Since the output period for coordinates is set to
crdout_period=1000, we have the total of nsteps/crdout_period=3000 snapshots in the
output trajectory.
[INPUT]
topfile
parfile
psffile
rstfile
pdbfile

=
=
=
=
=

[OUTPUT]
logfile =
dcdfile =
rstfile =
remfile =

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/alad.psf
./reus_run_eq{}.rst
../1_setup/alad.pdb

#
#
#
#
#

reus_run{}.log
reus_run{}.dcd
reus_run{}.rst
reus_run{}.rem

log file of each replica

DCD trajectory file
restart file
replica exchange ID file

[REMD]
dimension
exchange_period
type(1)
nreplica(1)
parameters(1)

=
=
=
=
=

#
#
#
#

topology file
parameter file
protein structure file
restart file
PDB file

2
1000
TEMPERATURE
4
300 301 302 303

17.2. Production simulation

74

GENESIS User Guide, Release 1.0

cyclic_params(1)=
type(2)
=
nreplica(2)
=
parameters(2)
=
cyclic_params(2)=
rest_function(2)=

NO
RESTRAINT
4
(1.0 144.0) (1.1 146.0) (1.2 148.0) (1.3 150.0)
NO
1

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
7.5
8.0
9.0
1
64
64
64

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period
nbupdate_period

=
=
=
=
=
=
=

LEAP
3000000
0.002
1000
1000
3000000
5

#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period
nonbond update period

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 300.0

#
#
#
#

[NVE,NVT,NPT]
thermostat
initial and target
temperature (K)

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z

=
=
=
=

PBC
64.0
64.0
64.0

#
#
#
#

[NOBC,PBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

[SELECTION]
group1
group2
group3
group4

=
=
=
=

an:NL
an:CA
an:CRP
an:NR

#
#
#
#

restraint
restraint
restraint
restraint

[RESTRAINTS]
nfunctions
function1
constant1
reference1
select_index1

=
=
=
=
=

1
DIHED
1.0
144.0
1 2 3 4

#
#
#
#
#

number of functions
restraint function type
force constant
reference
restraint groups

17.2. Production simulation

group
group
group
group

1
2
3
4

75

GENESIS User Guide, Release 1.0

17.3 Analysis of REUS simulation

In this step, we examine the performance of REUS simulation by analyzing the standard output file,
logfile (.log) and dcdfile (.dcd). We calculate (1) a time series of replica exchange, (2) a time series of
parameter exchange of three arbitrary chosen replicas, and (3) a time series of total potential energy of
three arbitrary chosen replicas.
First, we use remd_convert program which is one of the post-processing programs in the GENESIS
package. remd_convert is an utility to extract various quantities from a REUS trajectory:
#
$
#
$

change to the analysis directory

cd tutorial/alad_reus/5_analysis/
process of the REUS trajectory with remd_convert
remd_convert Inp_reus_conv | tee Inp_reus_conv.log

The control file for remd_convert (Inp_reus_conv) is shown below. In [INPUT] section, we specify
the output files (dcdfile and remfile) from the REUS simulation. In [SELECTION] section, we
define a group (group1) of the heavy atoms of ALAD which are fitted. In the [FITTING] section, the
fitting method is set. In this case, the translations TR and rotations ROT are allowed for the fitting.
In the [OPTION] section, the type of conversion is set to convert_type=PARAMETER, and the indices of the type to be converted are set to convert_ids=1 9 16. Period of the dcdfile trajectory
and the output format are set to dcd_md_period=1000 and trjout_format=DCD, respectively.
The selection of output atoms is set to trjout_atom=1, and a periodic boundary correction is set to
pbc_correct=MOLECULE.
[INPUT]
psffile
reffile
dcdfile
remfile

=
=
=
=

../1_setup/ala.psf
../1_setup/ala.pdb
../4_production/reus_run{}.dcd
../4_production/reus_run{}.rem

[OUTPUT]
pdbfile = ./reus_run_param.pdb
trjfile = ./reus_run_param{}.trj

# protein structure file

# PDB file

# PDB file
# trajectory file

[SELECTION]
group1
mole_name1

= molname:alad and heavy

# selection group 1
= alad PROT:1:ALAD PROT:1:ALAD

[FITTING]
fitting_method
fitting_atom
zrot_ngrid
zrot_grid_size

=
=
=
=

TR+ROT
1
10
1.0

#
#
#
#

[OPTION]
check_only
convert_type
convert_ids
dcd_md_period
trjout_format
trjout_type
trjout_atom
pbc_correct

=
=
=
=
=
=
=
=

NO
PARAMETER
1 9 16
1000
DCD
COOR
1
MOLECULE

# (YES/NO)
# (REPLICA/PARAMETER)

17.3. Analysis of REUS simulation

#
#
#
#
#

method
atom group
number of z-rot grids
z-rot grid size

input dcdfile MD period

(PDB/DCD)
(COOR/COOR+BOX)
selection group output
(NO/MOLECULE)

76

GENESIS User Guide, Release 1.0

To examine (1) the time series of replica exchange, we use the standard output file of the REUS simulation to extract time series of ReplicaID:
$

grep ReplicaID : ./reus_run.out |

awk / ReplicaID :/{print $3, $11, $18}> ./reus_run.replica
$ gnuplot
gnuplot> set xlabel "Time (ns)"
gnuplot> set ylabel "Replica"
gnuplot> plot "reus_run.replica" u ($0*0.002):1 w p pt 7 ps 0.5 lt 1
title "ParameterID=1"
gnuplot> plot "reus_run.replica" u ($0*0.002):2 w p pt 7 ps 0.5 lt 2
title "ParameterID=9"
gnuplot> plot "reus_run.replica" u ($0*0.002):3 w p pt 7 ps 0.5 lt 3
title "ParameterID=16"

17.3. Analysis of REUS simulation

77

GENESIS User Guide, Release 1.0

We can observed random walks in replica space at parameterID=1 (300 K, (1.0 144.0)),
parameterID=9 (302 K, (1.0 144.0)) and parameterID=16 (303 K, (1.3 150.0)) colored by red,
green and blue, respectively.
To examine (2) the time series of parameter exchange of three arbitrary chosen replicas (replica id = 1,
9 and 16), we use the standard output file of the REUS simulation to extract time series of ParameterID:
$ grep ParameterID : ./remd_run.param |
awk / ParameterID :/{print $2, $10, $17}> ./remd_run.param
$ gnuplot
gnuplot> set xlabel "Time (ns)"
gnuplot> set ylabel "Temperature [K]"
gnuplot> plot "reus_run.param" u ($0*0.002):1 w l lw 2 lt 1
title "replicaID=1",\
"reus_run.param" u ($0*0.002):2 w l lw 2 lt 2
title "replicaID=9",\
"reus_run.param" u ($0*0.002):3 w l lw 2 lt 3
title "replicaID=16"

17.3. Analysis of REUS simulation

78

GENESIS User Guide, Release 1.0

To examine (3) the time series of total potential energy of three arbitrary chosen replicas (replica id = 1,
9 and 16), we use the enefile of the REUS simulation to extract time series of POTENTIAL_ENE:
$ grep INFO: reus_run01.log | tail -n +2 | awk {print $5}>
./reus_run01.ene
$ grep INFO: reus_run09.log | tail -n +2 | awk {print $5}>
./reus_run09.ene
$ grep INFO: reus_run16.log | tail -n +2 | awk {print $5}>
./reus_run16.ene
$ gnuplot
gnuplot> set xlabel "Time (ns)"
gnuplot> set ylabel "Potential energy (kcal/mol)"
gnuplot> plot "reus_run01.ene" u (\$0*0.002):1 w l lw 1.5 lt 1
title "replicaID=1",\
"reus_run09.ene" u (\$0*0.002):1 w l lw 1.5 lt 1
title "replicaID=9",\
"reus_run16.ene" u (\$0*0.002):1 w l lw 1.5 lt 1
title "replicaID=16"

17.3. Analysis of REUS simulation

79

GENESIS User Guide, Release 1.0

We can observe random walks in temperature and potential energy spaces at replicaID=1, 9 and 16,
colored by red, green and blue, respectively. These results indicate that the REUS simulation performed
satisfactorily.
To plot PMF (Potential of Mean Force) surface versus two torsion angles: CLP-NL-CA-CRP () and
NL-CA-CRP-NR () at 300K with restraint bias removed, WHAM (Weighted Histogram Analysis
Method) could be applied to the REUS trajectory. However, WHAM is beyond the scope of this tutorial.

17.3. Analysis of REUS simulation

80

CHAPTER

EIGHTEEN

TUTORIAL 5: PARALLEL
INPUT/OUTPUT SCHEME
If a simulation system is huge, and a larger number of processors is availalbe, it takes a significant
amount of time to gather data for restart and trajectory file updates. So, it is preferred, each process writes
local information separately, rather than a single does for all. In GENESIS, the parallel input/output
(I/O) scheme is available to deal effectively with larger restart or trajectory files.
Note: The parallel I/O scheme are available in SPDYN only.
In this tutorial, we explain an example of the parallel I/O with BPTI protein used in Tutorial 1: Building
and Simulating BPTI in Water. The procedures of the simulation are identical, except that we are dealing
with the parallel I/O using the same number of input files and the number of MPI processes.
The input files of this tutorial are at tutorial/bpti_parallel_io/ of the GENESIS package. In
all cases, we use 8 MPI processes.

18.1 Minimization with restraints on the protein

Minimization is done to remove unphysical steric clashes or inappropriate geometries. Thus, it is recommended to do a minimiation before doing a molecular dynamics simulation. To make use of the parallel
i/o, the first step to do before minimization is to make the parallel restart files, each of which will be
assigned to each MPI processor for minimization run.

18.1.1 Making restart for minimization

prst_setup is used to make parallel restart files. First, lets check the control file (run.inp) in the
BPTI tutorial.
[INPUT]
topfile
parfile
psffile
pdbfile
reffile

=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ionize.psf
../1_setup/ionize.pdb
../1_setup/ionize.pdb

[OUTPUT]
dcdfile = run.dcd

#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
reference for restraints

# DCD trajectory file

81

GENESIS User Guide, Release 1.0

rstfile = run.rst

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=
=

PME
10.0
12.0
13.5
YES
0
72
80
72

#
#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
usage of lookup table
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[MINIMIZE]
nsteps
eneout_period
crdout_period
rstout_period

=
=
=
=

1000
100
100
1000

#
#
#
#

number of steps
energy output period
coordinates output period
restart output period

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z

=
=
=
=

PBC
70.8250
83.2579
69.0930

#
#
#
#

[PBC,NOBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

1
POSI
10.0
1

number of functions
restraint function type
force constant
restraint group

The control file is very similar to this except a few changes. First, in [OUTPUT] section, instead
of output restart file, parallel restart file for minimization should be specified. Second, you do not
need [MINIMIZE] section ( you can just leave the same information written in the control file of the
minimization). Third, you should specify the number of domains in each direction or the total number
of domains (which is identical to the number of MPI processes).
Below we write the input (setup.inp) to generate multiple restart files for minimization. It is difficult
to distinguish the necessary keywords from unnecessary ones. So, we wrote all the keywords identically
to control file for minimization. It is necessary to change in [OUTPUT] and [BOUNDARY] sections
only. In [BOUNDARY] section, you need to set the number of sub-domains in each direction using
domain_x, domain_y, and domain_z, or total number of sub-domains using domain_xyz. In
this example, we set the number of sub-domains in each domain:
[INPUT]
topfile
parfile
psffile
pdbfile
reffile

=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ionize.psf
../1_setup/ionize.pdb
../1_setup/ionize.pdb

[OUTPUT]
rstfile = setup().rst

18.1. Minimization with restraints on the protein

#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
reference for restraints

# parallel restart file

82

GENESIS User Guide, Release 1.0

cachepath = ./cache
[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
10.0
12.0
13.5
0
72
80
72

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[MINIMIZE]
nsteps
eneout_period
crdout_period
rstout_period

=
=
=
=

1000
100
100
1000

#
#
#
#

number of steps
energy output period
coordinates output period
restart output period

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z
domain_x
domain_y
domain_z

=
=
=
=
=
=
=

PBC
70.8250
83.2579
69.0930
2
2
2

#
#
#
#

[PBC,NOBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

1
POSI
10.0
1

number of functions
restraint function type
force constant
restraint group

Another noticeable change is cachepath in [OUTPUT] section. It is a path to a directory where

intermediate files are written before generating final restart files. If you do not specify cachepath, all
the intermediate files will be saved in memory only, and it takes more time.
Note: rst_setup is OpenMP parallized only (no MPI parallization).
The following commands makes multiply restart files for minimization, using 2 OpenMP threads:
#
$
#
$
#
$

change to the minimization directory

cd tutorial/bpti_parallel_io/2_minimization/
set the number of OpenMP threads
export OMP_NUM_THREADS=2
make multiple restart files
rst_setup setup.inp | tee setup.out

After executing these commands, you generate 8 restart files named setup0.rst ~ setup7.rst.
These files will be used instead of PDB and PSF files for minimization.

18.1. Minimization with restraints on the protein

83

GENESIS User Guide, Release 1.0

18.1.2 Minimization with restraint

After making restart files for minimization, you can minimize with 8 MPI processes, as the number
of restart files is 8. Each MPI process read one corresponding restart file (i.e., MPI rank 0 reads
setup0.rst).
The control input of minimization with the parallel I/O is very similar to the one without it:
[INPUT]
topfile = ../1_setup/top_all27_prot_lipid.rtf # topology file
parfile = ../1_setup/par_all27_prot_lipid.prm # parameter file
rstfile = setup().rst
[OUTPUT]
dcdfile = run().dcd
rstfile = run().rst

# DCD trajectory file

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
10.0
12.0
13.5
0
72
80
72

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[MINIMIZE]
nsteps
eneout_period
crdout_period
rstout_period

=
=
=
=

1000
100
100
1000

#
#
#
#

number of steps
energy output period
coordinates output period
restart output period

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z
domain_x
domain_y
domain_z

=
=
=
=
=
=
=

PBC
70.8250
83.2579
69.0930
2
2
2

#
#
#
#

[PBC,NOBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

1
POSI
10.0
1

number of functions
restraint function type
force constant
restraint group

The main difference between this and control input of minimization without the parallel I/O is at the
keywords of input and output files. If there are multiple files corresponding to each MPI process, ()
should be added as above.
You can perform minimization with the following command:

18.1. Minimization with restraints on the protein

84

GENESIS User Guide, Release 1.0

$ mpirun -np 8 spdyn run.inp | tee run.out

The outputs of the parallel I/O is identical to the ones without it.

18.2 Heat-up with restraints on the protein

The purpose of this procedure is explained in Tutorial 1: Building and Simulating BPTI in Water. Here,
we just explain how to use the parallel I/O.
For the task, first you need to change to the heat-up directory:
# change to the heat-up directory
$ cd tutorial/bpti_parallel_io/3_heating/

18.2.1 Restart files for heat-up simulation

As mentioned in the previous section, we may think that same control input by adding () is enough to
set up the parallel I/O simulation.
The control input file, corresponding to the one in Tutorial 1: Building and Simulating BPTI in Water,
can be written as follow by just by replacing input and output file names:
[INPUT]
topfile = ../1_setup/top_all27_prot_lipid.rtf # topology file
parfile = ../1_setup/par_all27_prot_lipid.prm # parameter file
rstfile = ../2_minimization/run().rst
# restart file
[OUTPUT]
dcdfile = run().dcd
rstfile = run().rst

# DCD trajectory file

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
10.0
12.0
13.5
1
72
80
72

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period
annealing
anneal_period
dtemperature

=
=
=
=
=
=
=
=
=

LEAP
5000
0.002
50
50
5000
YES
50
3

#
#
#
#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period
simulated annealing
annealing period
temperature change at
annealing (K)

18.2. Heat-up with restraints on the protein

85

GENESIS User Guide, Release 1.0

[CONSTRAINTS]
rigid_bond

= YES

# constraints of all bonds

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 0.1

# [NVE,NVT,NPT]
# thermostat
# initial temperature (K)

[BOUNDARY]
type
domain_x
domain_y
domain_z

=
=
=
=

# [PBC,NOBC]

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

PBC
2
2
2

1
POSI
10.0
1

number of functions
restraint function type
force constant
restraint group

However, if you use this control input, you get the following error message:
Pio_Check_Compatible>
Pio_Check_Compatible>
Pio_Check_Compatible>
Pio_Check_Compatible>

ERROR : [Constraint]rigid-bond: must be NO

ERROR : [Ensemble]ensemble: must be NVE
WARNING:[Boundary]box size: t=0 : 70.12 82.56 68.39
ERROR rank_no =
0

This is due to [CONSTRAINTS] section, which does not exist in the previous control input file. Unlike
the restart files without the parallel I/O, the potential function information is included in the restart files
(thats why PSF files is not necessary for the parallel I/O). So any change of potential functions requires
a consistent restart files to be regenerated.
The procedure of regenerating restart files is very similar to the case of minimization. The control input
to generate restart files is as follow:
[INPUT]
topfile
parfile
psffile
pdbfile
reffile
rstfile

=
=
=
=
=
=

../1_setup/top_all27_prot_lipid.rtf
../1_setup/par_all27_prot_lipid.prm
../1_setup/ionize.psf
../1_setup/ionize.pdb
../1_setup/ionize.pdb
../2_minimization/run().rst

[OUTPUT]
rstfile = setup().rst
cachepath = ./cache
[OUTPUT]
dcdfile = run().dcd
rstfile = run().rst

#
#
#
#
#
#

topology file
parameter file
protein structure file
PDB file
reference for restraints
restart file

# parallel restart file

# DCD trajectory file

# restart file

[ENERGY]

18.2. Heat-up with restraints on the protein

86

GENESIS User Guide, Release 1.0

electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
10.0
12.0
13.5
1
72
80
72

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
[CONSTRAINTS]
rigid_bond

= YES

# constraints of all bonds

[ENSEMBLE]
ensemble
tpcontrol

= NVT
= LANGEVIN

# [NVE,NVT,NPT]
# thermostat

[BOUNDARY]
type
box_size_x
box_size_y
box_size_z
domain_x
domain_y
domain_z

=
=
=
=
=
=
=

#
#
#
#

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

PBC
70.8250
83.2579
69.0930
2
2
2

1
POSI
10.0
1

[PBC,NOBC]
box size (x) in [PBC]
box size (y) in [PBC]
box size (z) in [PBC]

number of functions
restraint function type
force constant
restraint group

In this case, the restart files generated after minimization are in [INPUT] section. Here, we generated
restart files without setting the number of sub-domains in each direction, but setting the total number of
processes only with the keyword domain_xyz.
After running prst_setup setup.inp | tee setup.out, the restart files for the heat-up simulation are regenerated. Except [INPUT], [OUTPUT], and [BOUNDARY] sections.

18.2.2 Making restart files for heat-up simulation

Similarly to minimization, setup0.rst ~ setup7.rst files are used as restart files for the simulation. The control file for the simulations is:
[INPUT]
topfile = ../1_setup/top_all27_prot_lipid.rtf # topology file
parfile = ../1_setup/par_all27_prot_lipid.prm # parameter file
rstfile = setup().rst
# restart file
[OUTPUT]
dcdfile = run().dcd

18.2. Heat-up with restraints on the protein

# DCD trajectory file

87

GENESIS User Guide, Release 1.0

rstfile = run().rst

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
table_order
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=
=

PME
10.0
12.0
13.5
1
72
80
72

#
#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
order of lookup table
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period
annealing
anneal_period
dtemperature

=
=
=
=
=
=
=
=
=

LEAP
5000
0.002
50
50
5000
YES
50
3

#
#
#
#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period
simulated annealing
annealing period
temperature change at

[CONSTRAINTS]
rigid_bond

= YES

# constraints of all bonds

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NVT
= LANGEVIN
= 0.1

# [NVE,NVT,NPT]
# thermostat
# initial temperature (K)

[BOUNDARY]
type
domain_x
domain_y
domain_z

=
=
=
=

# [PBC,NOBC]

[SELECTION]
group1

= backbone

# index of restraint group 1

[RESTRAINTS]
nfunctions
function1
constant1
select_index1

=
=
=
=

#
#
#
#

PBC
2
2
2

1
POSI
10.0
1

number of functions
restraint function type
force constant
restraint group

By running mpirun -np 8 spdyn run.inp | tee run.out, the procedure of Heat-up with
restraints on the protein could be performed. Unlike the control in the previous section of minimization,
some informations in the [BOUNDARY] section are missing. This is due to the fact the information of
the [BOUNDARY] section is already saved in the parallel restart files, setup0.rst~setup7.rst.

18.3 Equilibration
To do equilibration, the following command is necessary:

18.3. Equilibration

88

GENESIS User Guide, Release 1.0

# change to the equilibration directory

$ cd tutorial/bpti_parallel_io/4_equilibration/

In this step, there is a change in [ENSEMBE] section (from NVT to NPT). Thus, we need to regenerate
RST files accordingly to the change:
[INPUT]
topfile = ../1_setup/top_all27_prot_lipid.rtf # topology file
parfile = ../1_setup/par_all27_prot_lipid.prm # parameter file
rstfile = ../3_heating/run().rst
# restart file
[OUTPUT]
rstfile = setup().rst

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=

PME
10.0
12.0
13.5
72
80
72

#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period

=
=
=
=
=
=

LEAP
5000
0.002
50
50
5000

#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period

[CONSTRAINTS]
rigid_bond

= YES

# constraints of all bonds

[ENSEMBLE]
ensemble
tpcontrol

= NPT
= LANGEVIN

# [NVE,NVT,NPT]
# thermostat and barostat

[BOUNDARY]
type
domain_x
domain_y
domain_z

=
=
=
=

# [PBC,NOBC]

PBC
2
2
2

From this input with prst_setup, multiple restart files are generated. After running prst_setup
setup.inp | tee setup.out, the parallel I/O files are regenerated and these are used for the
equilibration. In this case, we set the number of sub-domains to 2 in each dimension.
The control of the simulation run is as follow:
[INPUT]
topfile = ../1_setup/top_all27_prot_lipid.rtf # topology file
parfile = ../1_setup/par_all27_prot_lipid.prm # parameter file
rstfile = ./setup().rst
# restart file
[OUTPUT]

18.3. Equilibration

89

GENESIS User Guide, Release 1.0

dcdfile = run().dcd
rstfile = run().rst

# DCD trajectory file

# restart file

[ENERGY]
electrostatic
switchdist
cutoffdist
pairlistdist
pme_ngrid_x
pme_ngrid_y
pme_ngrid_z

=
=
=
=
=
=
=

PME
10.0
12.0
13.5
72
80
72

#
#
#
#
#
#
#

[CUTOFF,PME]
switch distance
cutoff distance
pair-list cutoff distance
grid size_x in [PME]
grid size_y in [PME]
grid size_z in [PME]

[DYNAMICS]
integrator
nsteps
timestep
eneout_period
crdout_period
rstout_period

=
=
=
=
=
=

LEAP
5000
0.002
50
50
5000

#
#
#
#
#
#

[LEAP,VVER]
number of MD steps
time step (ps)
energy output period
coordinates output period
restart output period

[CONSTRAINTS]
rigid_bond

= YES

# constraints all bonds

# involving hydrogen

[ENSEMBLE]
ensemble
tpcontrol
temperature

= NPT
= LANGEVIN
= 300.0

pressure

= 1.0

#
#
#
#
#

[BOUNDARY]
type

= PBC

# [PBC,NOBC]

[NVE,NVT,NPT]
barostat and thermostat
initial and target
temperature (K)
target pressure (atm)

As in the previous heat-up simulation, we skipped the simulation box size information in [BOUNDARY]
section, because the informations is already in the restart files.

18.4 Production simulation

The control input of a production simulation is identical to that of equilibration, except the number of
simulation steps and output periods in [DYNAMICS] section. In this case, we dont have to regenerate
restart files with prst_setup.
The control input is almost identical to the case without the parallel I/O, except [INPUT] and [OUTPUT] sections. As it is already mentioned in the previous sections, () should be appended to file names
for the parallel I/O. Here, control input of [INPUT] and [OUTPUT] sections only is given:
[INPUT]
topfile = ../1_setup/top_all27_prot_lipid.rtf # topology file
parfile = ../1_setup/par_all27_prot_lipid.prm # parameter file
rstfile = ../4_equilibration/run().rst
# restart file
[OUTPUT]
dcdfile = run().dcd

18.4. Production simulation

# DCD trajectory file

90

GENESIS User Guide, Release 1.0

rstfile = run().rst

# restart file

The commands for the production simulation are:

#
$
#
$
#
$

change to the production directory

cd tutorial/bpti_paralell_io/5_production/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform production with SPDYN by using 8 MPI processes
mpirun -np 8 spdyn run.inp | tee run.out

18.5 Analysis: RMSD calculation

This is a procedure to compute RMSD with respect to the crystal structure to explore stability.
For analysis with the parallel I/O scheme we use pcrd_conver in stead of crd_conver. The command
to calculate RMSDs from the production trajectory is:
#
$
#
$
#
$

change to the analysis directory

cd tutorial/bpti_parallel_io/6_analysis/
set the number of OpenMP threads
export OMP_NUM_THREADS=1
perform analysis with pcrd_convert
pcrd_convert run.inp | tee run.out

As in the case of production simulation, the control file of the parallel I/O is almost identical to the one
without it, except [TRAJECTORY] section. First, we add () to trjfile1 due to multiple trajectory
files. Second, COOR+BOX is not necessary because the box size information is already written to each
trajectory file. Below there is the control file for RMSD calculation.
[INPUT]
psffile = ../1_setup/ionize.psf
reffile = ../1_setup/ionize.pdb

# protein structure file

# PDB file

[OUTPUT]
rmsfile = run.rms

# RMSD file

[TRAJECTORY]
trjfile1
md_step1
mdout_period1
ana_period1

=
=
=
=

[SELECTION]
group1

= backbone and an:CA # selection group 001

[FITTING]
fitting_method
fitting_atom

= TR+ROT
= 001

# method
# atom group

[OPTION]
check_only

= NO

# (YES/NO)

../5_production/run().dcd
500000
500
500

18.5. Analysis: RMSD calculation

#
#
#
#

trajectory file
number of MD steps
MD output period
analysis period

91

GENESIS User Guide, Release 1.0

Finally, with this we obtain the same RMSD values as in Tutorial 1: Building and Simulating BPTI in
Water.

18.5. Analysis: RMSD calculation

92

BIBLIOGRAPHY

[1] D. A. Case, T. A. Darden, T. E. Cheatham, III, C. L. Simmerling, J. Wang, R. E. Duke, R. Luo,

R. C. Walker, W. Zhang, K. M. Merz, B. Roberts, S. Hayik, A. Roitberg, G. Seabra, J. Swails, A. W.
Goetz, I. Kolossvry, K. F. Wong, F. Paesani, J. Vanicek, R. M. Wolf, J. Liu, X. Wu, S. R. Brozell,
T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye, J. Wang, M.-J. Hsieh, G. Cui, D. R. Roe, D. H. Mathews,
M. G. Seetin, R. Salomon-Ferrer, C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko, and
P. A. Kollman. AMBER 12. University of California, San Francisco, 2012.
[2] B. R. Brooks, C. L. Brooks, A. D. Mackerell, L. Nilsson, R. J. Petrella, B. Roux, Y. Won, G. Archontis, C. Bartels, S. Boresch, A. Caflisch, L. Caves, Q. Cui, A. R. Dinner, M. Feig, S. Fischer,
J. Gao, M. Hodoscek, W. Im, K. Kuczera, T. Lazaridis, J. Ma, V. Ovchinnikov, E. Paci, R. W. Pastor, C. B. Post, J. Z. Pu, M. Schaefer, B. Tidor, R. M. Venable, H. L. Woodcock, X. Wu, W. Yang,
D. M. York, and M. Karplus. CHARMM: The biomolecular simulation program. J. Comput. Chem.,
30(10):15451614, 2009. URL: http://dx.doi.org/10.1002/jcc.212871 , doi:10.1002/jcc.212872 .
[3] S. Pronk, S. Pll, R. Schulz, P. Larsson, P. Bjelkmar, R. Apostolov, M. R. Shirts, J. C. Smith, P. M.
Kasson, D. van der Spoel, B. Hess, and E. Lindahl. GROMACS 4.5: a high-throughput and highly
parallel open source molecular simulation toolkit. Bioinformatics, 29(7):845854, 2013.
[4] K. J. Bowers, E. Chow, H. Xu, R. O. Dror, M. P. Eastwood, B. A. Gregersen, J. L. Klepeis, I. Kolossvary, M. A. Moraes, F. D. Sacerdoti, J. K. Salmon, Y. Shan, and D. E. Shaw. Scalable algorithms
for molecular dynamics simulations on commodity clusters. In SC 2006 Conference, Proceedings
of the ACM/IEEE, 1117. IEEE, 2006.
[5] J. C. Phillips, R. Braun, W. Wang, J. Gumbart, E. Tajkhorshid, E. Villa, C. Chipot, R. D.
Skeel, L. Kal, and K. Schulten. Scalable molecular dynamics with NAMD. J. Comput. Chem.,
26(16):17811802, 2005. URL: http://dx.doi.org/10.1002/jcc.202893 , doi:10.1002/jcc.202894 .
[6] D. A. Case, T. E. Cheatham, T. Darden, H. Gohlke, R. Luo, K. M. Merz, A. Onufriev, C. Simmerling, B. Wang, and R. J. Woods. The Amber biomolecular simulation programs. J. Comput. Chem.,
26(16):16681688, 2005.
[7] A. D. MacKerell, D. Bashford, M. Bellott, R. L. Dunbrack, J. D. Evanseck, M. J. Field, S. Fischer,
J. Gao, H. Guo, S. Ha, D. Joseph-McCarthy, L. Kuchnir, K. Kuczera, F. T. K. Lau, C. Mattos,
S. Michnick, T. Ngo, D. T. Nguyen, B. Prodhom, W. E. Reiher, B. Roux, M. Schlenkrich, J. C.
Smith, R. Stote, J. Straub, M. Watanabe, J. Wiorkiewicz-Kuczera, D. Yin, and M. Karplus. All-atom
empirical potential for molecular modeling and dynamics studies of proteins. J. Phys. Chem. B,
102(18):35863616, 1998.
1

http://dx.doi.org/10.1002/jcc.21287
http://dx.doi.org/10.1002/jcc.21287
3
http://dx.doi.org/10.1002/jcc.20289
4
http://dx.doi.org/10.1002/jcc.20289
2

93

GENESIS User Guide, Release 1.0

[8] A. D. MacKerell, M. Feig, and C. L. Brooks. Improved treatment of the protein backbone in empirical force fields. J. Am. Chem. Soc., 126(3):698699, 2004.
[9] W. L. Jorgensen, D. S. Maxwell, and J. Tirado-Rives. Development and Testing of the OPLS AllAtom Force Field on Conformational Energetics and Properties of Organic Liquids. J. Am. Chem.
Soc., 118(45):1122511236, 1996.
[10] C. Oostenbrink, A. Villa, A. E. Mark, and W. F. Van Gunsteren. A biomolecular force field based
on the free enthalpy of hydration and solvation: The GROMOS force-field parameter sets 53A5 and
53A6. J. Comput. Chem., 25(13):16561676, 2004.
[11] J. Jung, T. Mori, and Y. Sugita. Efficient lookup table using a linear function of inverse distance
squared. J. Comput. Chem., 34(28):24122420, 2013.
[12] J. Jung, T. Mori, and Y. Sugita. Midpoint cell method for hybrid (MPI+OpenMP) parallelization
of molecular dynamics simulations. J. Comput. Chem., 35(14):10641072, 2014.
[13] Open MPI. http://www.open-mpi.org/.
[14] CHARMM. http://www.charmm.org/.
[15] psfgen. http://www.ks.uiuc.edu/Research/vmd/plugins/psfgen/ug.pdf.
[16] NAMD. http://www.ks.uiuc.edu/Research/namd/.
[17] VMD. http://www.ks.uiuc.edu/Research/vmd/.
[18] Y. Sugita and Y. Okamoto. Replica-exchange molecular dynamics method for protein folding.
Chem. Phys. Lett., 314(12):141151, 1999.
[19] Y. Sugita, A. Kitao, and Y. Okamoto. Multidimensional replica-exchange method for free-energy
calculations. J. Chem. Phys., 113(15):60426051, 2000.
[20] W. L. Jorgensen, J. Chandrasekhar, J. D. Madura, R. W. Impey, and M. L. Klein. Comparison of
Simple Potential Functions for Simulating Liquid Water. J. Chem. Phys., 79(2):926935, 1983.
[21] N. Foloppe and A. D. Mackerell. All-atom empirical force field for nucleic acids: I. parameter
optimization based on small molecule and condensed phase macromolecular target data. J. Comput.
Chem., 21(2):86104, 2000.
[22] A. D. Mackerell and N. K. Banavali. All-atom empirical force field for nucleic acids: II. application
to molecular dynamics simulations of dna and rna in solution. J. Comput. Chem., 21(2):105120,
2000.
[23] S. D. Feller, D. X. Yin, R. W. Pastor, and A. D. Mackerell. Molecular dynamics simulation of unsaturated lipid bilayers at low hydration: parameterization and comparison with diffraction studies.
Biophys. J., 73(5):22692279, 1997.
[24] J. B. Klauda, R. M. Venable, J. A. Freites, J. W. OConnor, D. J. Tobias, C. Mondragon-Ramirez,
I. Vorobyov, and R. W. Pastor. Update of the charmm all-atom additive force field for lipids: validation on six lipid types. J. Phys. Chem. B, 114(23):78307843, 2010.
[25] R. B. Best, X. Zhu, J. Shim, P. E. M. Lopes, J. Mittal, M. Feig, and A. D. MacKerell. Optimization
of the additive CHARMM all-atom protein force field targeting improved sampling of the backbone
\phi , \psi and side-Chain \chi 1 and \chi 2 dihedral angles. J. Chem. Theo. Comput., 8(9):32573273,
2012.
[26] J. Huang and A. D. MacKerell. CHARMM36 all-atom additive protein force field: Validation
based on comparison to NMR data. J. Comput. Chem., 34(25):21352145, 2013.

Bibliography

94

GENESIS User Guide, Release 1.0

[27] J. Karanicolas and C. L. Brooks, III. The origins of asymmetry in the folding transition states of
protein L and protein G. Protein Sci., 11(10):23512361, 2002.
[28] J. Karanicolas and C. L. Brooks III. Improved Go-like models demonstrate the robustness of protein folding mechanisms towards non-native interactions. J. Mol. Biol., 334(2):309325, 2003.
[29] N. Go. Theoretical studies of protein folding. Annu. Rev. Biophys. Bioeng., 12:183210, 1983.
[30] P. J. Steinbach and B. R. Brooks. New Spherical-Cutoff Methods for Long-Range Forces in Macromolecular Simulation. J. Comput. Chem., 15(7):667683, 1994.
[31] L. Verlet. Computer Experiments on Classical Fluids .I. Thermodynamical Properties of LennardJones Molecules. Phys. Rev., 159(1):98103, 1967.
[32] T. Darden, D. York, and L. Pedersen. Particle mesh Ewald: An Nlog(N) method for Ewald sums
in large systems. J. Chem. Phys., 98(12):1008910092, 1993.
[33] U. Essmann, L. Perera, M. L. Berkowitz, T. Darden, H. Lee, and L. G. Pedersen. A smooth particle
mesh Ewald method. J. Chem. Phys., 103(19):85778593, 1995.
[34] D. Takahashi. FFTE: A Fast Fourier Transform Package. http://www.ffte.jp/.
[35] FFTW. http://www.fftw.org/.
[36] L. Nilsson. Efficient Table Lookup Without Inverse Square Roots for Calculation of Pair Wise
Atomic Interactions in Classical Simulations. J. Comput. Chem., 30(9):14901498, 2009.
[37] J. P. Ryckaert, G. Ciccotti, and H. J. C. Berendsen. Numerical-Integration of Cartesian Equations
of Motion of a System with Constraints - Molecular-Dynamics of N-Alkanes. J. Comput. Chem.,
23(3):327341, 1977.
[38] H. C. Andersen. Rattle - a Velocity Version of the Shake Algorithm for Molecular-Dynamics Calculations. J. Comput. Chem., 52(1):2434, 1983.
[39] S. Miyamoto and P. A. Kollman. Settle - an Analytical Version of the Shake and Rattle Algorithm
for Rigid Water Models. J. Comput. Chem., 13(8):952962, 1992.
[40] S. A. Adelman and J. D. Doll. Generalized Langevin Equation Approach for Atom-Solid-Surface
Scattering - General Formulation for Classical Scattering Off Harmonic Solids. J. Chem. Phys.,
64(6):23752388, 1976.
[41] D. Quigley and M. I. J. Probert. Langevin dynamics in constant pressure extended systems. J.
Chem. Phys., 120(24):1143211441, 2004.
[42] A. Mitsutake, Y. Sugita, and Y. Okamoto. Generalized-ensemble algorithms for molecular simulations of biopolymers. Biopolymers, 60(2):96123, 2001.
[43] J. A. McCammon, B. R. Gelin, and M. Karplus. Dynamics of folded proteins.. Nature,
267(5612):585590, 1977.
[44] D. E. Shaw, P. Maragakis, K. Lindorff-Larsen, S. Piana, R. O. Dror, M. P. Eastwood, J. A. Bank,
J. M. Jumper, J. K. Salmon, Y. Shan, and W. Wriggers. Atomic-level characterization of the structural dynamics of proteins.. Science, 330(6002):341346, 2010.
[45] gnuplot. http://www.gnuplot.info/.
[46] RCSB Protein Data Bank. http://www.rcsb.org/.
[47] solvate plugin. http://www.ks.uiuc.edu/Research/vmd/plugins/solvate/.
[48] autoionize plugin. http://www.ks.uiuc.edu/Research/vmd/plugins/autoionize/.

Bibliography

95

GENESIS User Guide, Release 1.0

[49] MMTSB web service. http://mmtsb.org/webservices/gomodel.html.

[50] Temperature generator for REMD-simulations. http://folding.bmc.uu.se/remd/.

Bibliography

96