Molecular Dynamcis analysis

From Bioinformatikpedia
Revision as of 11:50, 16 September 2011 by Landerer (talk | contribs) (Root mean square fluctuations)

by Robert Greil and Cedric Landerer

Wildtype

First of all, we checked the resulting file with gmxcheck.

  • gmxcheck -f ref_md.tpr.xtc

Result

Reading frame       0 time    0.000   
# Atoms  68601
Precision 0.001 (nm)
Last frame       2000 time 10000.000   

Item        #frames Timestep (ps)
Step          2001    5
Time          2001    5
Lambda           0
Coords        2001    5
Velocities       0
Forces           0
Box           2001    5

The Simulation toked 6h33:50 ans the simulation speed was 36.564 ns/day. So, to reach 1 second of simulation, we had to wait around 75061 years. But this is a bit to long for this Project, so we just used the results we got. The potential energy was fluctuating about -9.185e+05 kJ/mol with a range of about 0.15e+04 kJ/mol. These information are given in the different log-files provided by the simulation.


Figure 1: Motion of the wild-type protein, Cartoon representation
Figure 2: Motion of the wild-type protein

To create the images, we saved each frame in PyMol<ref>The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, LLC</ref> as an image, which we converted into gif format by for file in *.png; do convert "$file" "$(basename $file .png).gif"; done. Than we were able to use gifsicle to create an animated gif with the command gifsicle *.gif -loop.

As one can see in Figure 1 and Figure 2, we have mostly a motion in space. The motion within the protein, like it is shown by the normal mode analysis is not identifiable. There is no movement of the beta-sheet or helical region against each other.

Quality assurance

For the quality assurance, we checked the convergence of different relevant parameters like the temperature, pressure or density. If they reach a thermal equilibrium, we can be sure our simulation has not to be extended any further.

Convergence of energy terms

To create the necessary files to check the convergence of energy terms we used the command:

  • g_energy -f ref_md.tpr.edr -o xvg/xxxx.xvg.

Temperature

Figure 3: Temperature over time of the molecular dynamics simulation of the wild-type protein
Energy                      Average   Err.Est.       RMSD  Tot-Drift 
-------------------------------------------------------------------------------
Temperature                 297.942     0.0055    1.11471  0.0143589  (K)

Figure 3 shows the temperature over the simulation time. The energy fluctuates between 294K (20.85°C) and 302K (28.85°C) around a mean of about 298K (24.85°C). So, the protein is simulated in a temperature range at which a hypothermia would be the consequence. As the temperature stays the whole simulation time in the same range, we can rate this as a stable region, but the fluctuation is about 8K. So the region is quite large and as there is no trend observable, so we can not see any convergence.

Pressure

Figure 4: Pressure over time of the molecular dynamics simulation of the wild-type protein
Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Pressure                   0.995939      0.014     83.256  0.0113633  (bar)

Figure 4 shows the pressure during the simulation. The pressure is not cnovereged as it ranges from -300 bar to 300 bar, which corresponds to about 600 atmospheres. The normal pressure in a human cell is about 0.37 bar<ref>D. A. T. Dick and Leah M. Lowenstein Osmotic Equilibria in Human Erythrocytes Studied by Immersion Refractometry</ref>, so the pressure during the simulation is not in the physiological range.

Energy

Figure 5: Energy over time of the molecular dynamics simulation of the wild-type protein
Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Potential                   -919346         21    889.917   -54.8929  (kJ/mol)
Kinetic En.                  164147          3    614.133    7.91086  (kJ/mol)
Total Energy                -755199         19    1096.12   -46.9806  (kJ/mol)

Figure 5 shows the energy terms of the simulation. Included are the potential, the kinetic and the total energy. The total energy is calculated by substracting the potential energy from the kinetic energy. All energy terms reach the end of the simulation with about the same value as at the begining of the simulation. So, the fluctuation is just in a small range.

Volume

Figure 6: Volume over time of the molecular dynamics simulation of the wild-type protein
Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Volume                      694.068      0.012   0.512013 0.00522608  (nm^3)

The simulation volume shown in Figure 6 moves around 694 nm^3. A convergence is not obersvable as the fluctuation is about 4 nm^3 during the whole simulation. But as the range of fluctuation is very small, we would classify this as a stable volume.

Density

Figure 7: Densety over time of the molecular dynamics simulation of the wild-type protein
Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Density                     999.461      0.018   0.737307 -0.00750088  (kg/m^3)

The density of the simulation stays all the time in the same range (Figure 7), with just a small divergence. Therefor we can assume a stable density for the simulation.

Box

Figure 8: Box size in X,Y and Z direction over time of the molecular dynamics simulation of the wild-type protein. The X and Y values are equal over time.
Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Box-X                       9.93815    5.8e-05 0.00244379 2.49479e-05  (nm)
Box-Y                       9.93815    5.8e-05 0.00244379 2.49479e-05  (nm)
Box-Z                       7.02734    4.1e-05 0.00172803 1.76268e-05  (nm)

The size of the box containing the simulation volume is very constant (Figure 8).

Interaction Energy: Coulomb

Figure 9: Coulomb energy of the simulation over time.
Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
Coul-SR:Protein-non-Protein   -21213.5         39    373.942    197.347  (kJ/mol)
Coul-14:Protein-non-Protein          0          0          0          0  (kJ/mol)

The coulomb energy is not converged at all like it is shown in Figure 9. So the high fluctuation means that we have a high fluctuation interaction with the surrounding solvet. That could mean, that the physiological environment is not matched at all. This could also be the reason for the strong movement in space. So the electrostatic interferences pushes the protein in one direction. As Coul-14:Protein-non-Protein is just 0, we did not include this term to the figure.

Interaction Energy: Van der Waals

Figure 10: Van der Waals energy of the simulation over time.
Energy                      Average   Err.Est.       RMSD  Tot-Drift
-------------------------------------------------------------------------------
LJ-SR:Protein-non-Protein    -2237.6         18    140.778    93.3922  (kJ/mol)
LJ-14:Protein-non-Protein          0          0          0          0  (kJ/mol)

The Van der Waals energy describes the close range interaction of non covalent bound atoms. Figure 10 shows the interaction energy of the protein with the solvent. Here we see a smaller fluctuation compared to the coulomb energy, but also here, a convergence is not obervable. As LJ-14:Protein-non-Protein is just 0, we did not include this term to the figure.

Minimum distances between periodic images

Root mean square fluctuations

Convergence of RMSD

Convergence of radius of gyration

Mutation 8a [C282Y]

still waiting to finish // LRZ won't do anything because of LRZ cluster update

Mutation 8b [C282S]

still waiting to finish // LRZ won't do anything because of LRZ cluster update

References

<references />