See https://www.r-ccs.riken.jp/labs/cbrt/installation/ for installation requirements
Clone the NMMD repository to /path/to/NMMD. The repository contains a modified version of GENESIS 1.4.0 to perform NMMD and a NMA program called ElNemo.
To build GENESIS :
cd /path/to/NMMD
./configure
make install
To build ElNemo :
cd /path/to/NMMD/ElNemo
make
In the INP file :
/path/to/NMMD/bin/atdyn INP
The INP file is used as in GENESIS (see how to define the INP file in https://www.r-ccs.riken.jp/labs/cbrt/usage/). To use NMMD, you must chose the integrator=NMMD in the DYNAMICS section and add a section NMMD to the INP file as follow :
...
[DYNAMICS]
integrator = NMMD
...
[NMMD]
nm_number = 10
nm_mass = 10.0
nm_limit = 1000.0
elnemo_cutoff = 8.0
elnemo_rtb_block = 10
elnemo_path = /path/to/NMMD/ElNemo
nm_prefix = /prefix/for/NMMD
nm_number: The number of normal modes to use after skipping the 6 first modes (example : 10 will use modes from 7-16), default 10nm_mass: Mass value of Normal modes for NMMD, default : 10.0nm_limit: Threshold of normal mode amplitude above/below which the normal modes are updated, default : 1000.0elnemo_cutoff: Cutoff distance for elastic network model (Angstrom) default : 8.0elnemo_rtb_block: Number of residue per RTB block in the NMA computation : 10elnemo_path: Path to ElNemo installation, default : Nonenm_prefix: Path to a temporary directory to store NMMD files during the simulation, default : None
- NMMD is available only for LANGEVIN temperature control in the NVT enemble
- SHAKE/RATTLE algorithm have to be turn off
[EXPERIMENTS]
emfit = YES # YES/NO
emfit_type = IMAGE # VOLUME/IMAGE
emfit_target = path/to/image/file.spi # 2D SPIDER file
emfit_sigma = 2.0 # sigma of 2D gaussian
emfit_tolerance = 0.01 # Gaussian truncation threshold
emfit_period = 1 # not used
emfit_roll_angle = 0.0 # Euler roll angle in degrees
emfit_tilt_angle = 0.0 # Euler tilt angle in degrees
emfit_yaw_angle = 0.0 # Euler yaw angle in degrees
emfit_shift_x = 0.0 # Shift in x direction in pixels
emfit_shift_y = 0.0 # Shift in y direction in pixels
emfit_pixel_size = 1.0 # Size of a pixel in Angstrom