Difference between revisions of "Team:Aalto-Helsinki/Model Setup"
| Line 107: | Line 107: | ||
algorithm. this energy minimization was followed by 0.5 ns NVT equilibration with a time step of | algorithm. this energy minimization was followed by 0.5 ns NVT equilibration with a time step of | ||
2 fs. During NVT equilibration, the temperature was controlled by the stochastic velocity rescaling | 2 fs. During NVT equilibration, the temperature was controlled by the stochastic velocity rescaling | ||
| − | thermostat developed by Bussi et al. [25] with | + | thermostat developed by Bussi et al. [25] with τT = 0:1 ps. The NVT equilibration was followed by |
NPT equilibration for 1 ns with a time step of 2 fs. During the NPT equilibration, the pressure of | NPT equilibration for 1 ns with a time step of 2 fs. During the NPT equilibration, the pressure of | ||
| − | the system was set to 1.0 bar using the isotropic Parrinello-Rahman [26] pressure control with | + | the system was set to 1.0 bar using the isotropic Parrinello-Rahman [26] pressure control with τp = 2:0 |
ps. The temperature was controlled by the stochastic velocity rescaling thermostat by Bussi et al. [25] | ps. The temperature was controlled by the stochastic velocity rescaling thermostat by Bussi et al. [25] | ||
| − | with | + | with τT = 0:1 ps. During both NVT and NPT equilibration runs, position restrains were applied to the |
protein structure. | protein structure. | ||
</p> | </p> | ||
| Line 118: | Line 118: | ||
<p id="paragraph"> | <p id="paragraph"> | ||
For the production run, the pressure of the system was set to 1:0 bar using the isotropic Parrinello- | For the production run, the pressure of the system was set to 1:0 bar using the isotropic Parrinello- | ||
| − | Rahman [26] pressure control with | + | Rahman [26] pressure control with τp = 12 ps. The temperature was controlled by the stochastic |
| − | velocity rescaling thermostat by Bussi et al. [25] with | + | velocity rescaling thermostat by Bussi et al. [25] with τT = 0:1 ps. A time-step of 20 fs was used. For |
all equilibration and production runs, electrostatic interactions were calculated using the Particle-Mesh | all equilibration and production runs, electrostatic interactions were calculated using the Particle-Mesh | ||
Ewald method[27] and periodic boundary conditions were applied. The simulation time of the systems | Ewald method[27] and periodic boundary conditions were applied. The simulation time of the systems | ||
Revision as of 15:42, 28 October 2017
Computational methods and simulation set-up
Molecular modeling methods ultimately aim for direct comparison with experimental measurements. As such, a good model of molecular interaction is essential. Quantum chemistry based Ab initio molecular dynamics methods aim to reduce the amount of fitting and guesswork required for accurate modeling of molecular interactions. However, such approaches are generally limited to small systems and short timescales due to the added computational demand[15]. In classical molecular dynamics, molecules are described using stick-and-ball models: spherical atoms are connected by springs that represent bonds. As such, internal molecular forces can be described by simple mathematical models. For example, Hooke's law can be used to describe bonded interactions, while non-bonded interactions can be described by Lennard-Jones potential.[13, 18]
Molecular dynamics is based on numerical, step-by-step, evaluation of Newton's equations of motion.
Due to the many-body nature of the problem, Newton's equations of motion are discretized and solved
numerically. MD trajectories, that describe the time evolution of the system, consist of both position
and velocity vectors of the particles in the system. The position and velocity vectors are reevaluated
according to finite time interval by using numerical integrators. The position vectors define the geometric
configuration of the system while the velocity vectors define the kinetic energy and temperature of the
system.[13, 18, 15]
We investigated the behavior of DCD-1L in aqueous solvents at varying salt concentrations and tem- peratures using molecular dynamics. the simulations make use of the GROMOS forcefield parameter set 53a6[17]. GROMOS53A6 is considered a united-tom forcefield, which maps a carbon and its associated apolar hydrogens as one interaction center. The parametrization of the forcefield is based on accurate reproduction of free enthalpies of hydration and apolar solvation of a wide range of compounds. The relative free enthalpy of solvation between apolar and polar environments is an important property in many biological phenomena, including protein folding and membrane formation.
System set-up
We probed the behavior of a single DCD-1L solvated in water with added 2 Na+ counter-ions to neutralize the net negative charge of the peptide, as well as added salt (NaCl), with the salt concentrations varying from 20 mM to 500 mM. Additionally we probed the temperature dependency of DCD-1L structure by simulating a single peptide in water with 2 Na+ counter-ions at temperatures of 290 K, 300 K, 310 K and 320 K. All the simulation trajectories were calculated using the Gromacs v.4.6.7 simulation package[19, 20].
As starting structure for our simulations, chose the helical crystal structure of dermcidin derived by Song et al.[6] available in the RCSB protein Data Bank (entry 2YMK). Due to the presence of missing atoms in the original model by Song et al., we constructed a a homology model based on the DCD-1L sequence used in our laboratory constructs using the SWISS-MODEL automated protein structure homology-modeling server[21, 22, 23, 24]. The resulting structure was then mapped to the GROMOS53A6 forcefield using the Gromacs 4.6.7 tool pdb2gmx.
In all simulations, the dimensions of the simulation box were cubic with a side length of 11.2 nm. The peptide was inserted and centered into the simulation box and solvated in water. For all simulations, the simple point-charge (SPC) water model was used. After solvation, two of the water molecules were replaced by 2 Na+ ions to act as counter-ions and effectively neutralize the charge on the simulated system. Additionally, for simulations involving added salt, an adequate number of water molecules were replaced by an equal number of Na+ and Cl ions to mimic the desired salt concentration of either 20 mM, 50 mM, 70 mM, 100 mM, 120 mM, 150 mM, 200 mM, 300 mM or 500 mM. Visualizations of the simulated system set-up configurations are presented in figure X.
Prior to the production run, each simulation system was energy minimized using the steepest descent algorithm. this energy minimization was followed by 0.5 ns NVT equilibration with a time step of 2 fs. During NVT equilibration, the temperature was controlled by the stochastic velocity rescaling thermostat developed by Bussi et al. [25] with τT = 0:1 ps. The NVT equilibration was followed by NPT equilibration for 1 ns with a time step of 2 fs. During the NPT equilibration, the pressure of the system was set to 1.0 bar using the isotropic Parrinello-Rahman [26] pressure control with τp = 2:0 ps. The temperature was controlled by the stochastic velocity rescaling thermostat by Bussi et al. [25] with τT = 0:1 ps. During both NVT and NPT equilibration runs, position restrains were applied to the protein structure.
For the production run, the pressure of the system was set to 1:0 bar using the isotropic Parrinello- Rahman [26] pressure control with τp = 12 ps. The temperature was controlled by the stochastic velocity rescaling thermostat by Bussi et al. [25] with τT = 0:1 ps. A time-step of 20 fs was used. For all equilibration and production runs, electrostatic interactions were calculated using the Particle-Mesh Ewald method[27] and periodic boundary conditions were applied. The simulation time of the systems varied from 50 ns to 100 ns, depending on the system.
Analysis of the simulation trajectories was carried out using Gromacs 4.6.7 built-in tools and visual- ization of trajectories was accomplished using VMD[28]. Assignment of secondary structure elements of the peptide was done using a Gromacs interface for DSSP[29, 30].
References
[1] Writers, YEAR. Name of article / book. Publication. Accessible at: [url here].
[2] Writers, YEAR. Name of article / book. Publication. Accessible at: [url here].
[3] Writers, YEAR. Name of article / book. Publication. Accessible at: [url here].
[4] Writers, YEAR. Name of article / book. Publication. Accessible at: [url here].
[5] Writers, YEAR. Name of article / book. Publication. Accessible at: [url here].
