Difference between revisions of "Fabry:Molecular Dynamics Simulations"

From Bioinformatikpedia
(Trajectory)
Line 131: Line 131:
 
</figure>
 
</figure>
 
</div>
 
</div>
  +
<br style="clear:both">
 
 
==== Step ====
 
==== Step ====
 
In <xr id="fig:Step"/>, it is demonstrated, how the whole solvent box is divided into smaller parts. In this case the files ''step32242b_n6.pdb'', ''step32242b_n9.pdb'', ''step32242b_n13.pdb'' and ''step32242b_n16.pdb'' are displayed.
 
In <xr id="fig:Step"/>, it is demonstrated, how the whole solvent box is divided into smaller parts. In this case the files ''step32242b_n6.pdb'', ''step32242b_n9.pdb'', ''step32242b_n13.pdb'' and ''step32242b_n16.pdb'' are displayed.
Line 137: Line 137:
 
<div style="float: left; margin-right: 20px;">
 
<div style="float: left; margin-right: 20px;">
 
<figure id="fig:Step">
 
<figure id="fig:Step">
[[File:FABRY_Traj_wt.gif|thumb|left|400px|<caption>Trajectory of the first 17 states of the wildtype - test call on partition mpp1_inter</caption>]]
+
[[File:FABRY_steps.gif|thumb|left|400px|<caption>Trajectory of the first 17 states of the wildtype - test call on partition mpp1_inter</caption>]]
 
</figure>
 
</figure>
 
</div>
 
</div>

Revision as of 16:00, 2 July 2012

Fabry Disease » Molecular Dynamics Simulations


The following analyses were performed on the basis of the α-Galactosidase A sequence. Please consult the journal for the commands used to generate the results.

Preparation

<figtable id="tab:MutSCWRL"> Mutagenesis of the 2 selected SNPs performed with SCWRL in Task 8. This was done on the basis of a backbone dependant library. The wildtype amino acid is shown in green, the mutated one in red. The whole structure of the wildtype is colored light blue, the mutant light red. Hydrogen bonds of the mutant to the surrounding are depicted in red, those of the not mutated site in green.

R118H
R356W

</figtable>

We chose to use the SNPs R118H and R356W. For the first one we want to examine if it is really not disease causing, although all results indicate otherwise so far. The second one has a strong effect on the protein. For both, we used the structure obtained by SCWRL in Task 8

Steps

The following steps have been taken from the AGroS source code:

  1. Run RepairPDB and findBreaks, additional scripts to adapt PDB files to MD reality.
  2. Run STRWater, an additional script that saves information for Structural Water Molecules in PDB file.
  3. Run SCWRL, a program employed to correctly assign side chain information to PDB structures.
  4. Run minimization in vacuum.
  5. Create water box, includes solvent (water with 0.1 [NaCl]) and neutralizes protein intrinsic charge.
  6. Create restrain files for each chain individually.
  7. Run solvent minimization with fixed Protein (backbone + sidechains).
  8. Run minimization with fixed Backbone.
  9. Run short production NVT MD.
  10. Run short production NPT MD.
  11. Run Production MD.

Intermediate PDB files

The intermediate pdb files listed in <xr id="tab:files"/> have been created during the test run of the wildtype structure file.

<figtable id="tab:files"> The following files were obtained from a short test run with the wild type structure 3S5Y. The chain itself is shown in pink, everything else in blue.

file name properties picture
3S5Y_chainA_repaired.pdb Repaired pdb file (repairPDB applied) of structure 3S5Y as we used it in Task 7
FABRY original.png
3S5Y_chainA_repaired_repair.pdb
3S5Y_chainA_repaired_repair_0.pdb
Repaired pdb file for structure 3S5Y, for this step the script repairPDB is used to remove potentially included DNA and to change the indices in the pdb file. This is done also for each chain separately, but since we decided to simulate only one chain, there is only one additional file "_0"
FABRY repair.png
3S5Y_chainA_repaired_br.pdb
3S5Y_chainA_repaired_br_0.pdb
Structure 3S5Y, removed ligands and water. Now only the protein itself is included.
The applied script finds potential breaks in the structure and divides it into parts. Since our used structure does not have any breaks, there is only one structure and both files are equal.
FABRY br.png
3S5Y_chainA_repaired_dna.pdb Contains potential included water less than 15 angstrom away from the structure, that is removed from original structure file in the repair step. --
3S5Y_chainA_repaired_dna.pdb Contains potential included DNA, that is removed from original structure file in the repair step. --
3S5Y_chainA_repaired_sc.pdb SCWRL applied to the structure.
Although we used the SCWRL output file from last week's task there were some changes in the secondary structure. One example can be seen in the lower picture. The original structure is shown in pink, the SCWRL changed one in green.
FABRY sc.png
Although we used the SCWRL output file from last week's task there were some changes in the secondary structure. One example can be seen on the right. The original structure is shown in pink, the SCWRL changed one in green.
FABRY CompareOriSc.png
3S5Y_chainA_repaired_nh.pdb No hydrogens; extracts the protein itself from the SCWRL structure.
FABRY nh.png
Summary of all pdb files Again, all chain structures, except for one are shown in pink. The green structure is the SCWRL modified one, which differs in some structural aspects from the others. Hetero atoms are shown in blue, those that differ are shown in red. Here we can see, that the repair step deleted the ligand (2R,3S,4R,5S)-2-(hydroxymethyl)piperidine- 3,4,5-triol.
FABRY allIntermedPDB.png
file name properties picture
3S5Y_chainA_repaired_solv_tmp.pdb
3S5Y_chainA_repaired_solv.pdb
Solvent (blue), Natrium (red) and Calcium (green) are added to the protein (pink) in the temp file. In the solvent file, potentially overlapping solvent molecules are deleted. Since both files contain the same number of atoms, we assume nothing has been changed and there were no overlaps.
FABRY solvTmp.png
3S5Y_chainA_repaired_solv_tmp.pdb
3S5Y_chainA_repaired_solv_0.pdb
The protein only.
FABRY solv0.png
3S5Y_chainA_repaired_solv_min.pdb First minimization run in which the protein's backbone and sidechains are fixed and only the solvent can be altered.
FABRY solvMin.png
3S5Y_chainA_repaired_solv_min_0.pdb Protein chain only for the minimization step
FABRY solvMin0.png
3S5Y_chainA_repaired_solv_min2.pdb Second minimization run in which only the protein's backbone is fixed and the solvent and sidechains can be altered.
FABRY solvMin2.png
3S5Y_chainA_repaired_solv_min3.pdb Third minimization run in which again only the protein's backbone is fixed and the solvent and sidechains can be altered. Here a different integrator is used.
FABRY solvMin3.png
3S5Y_chainA_repaired_solv_min.pdb
3S5Y_chainA_repaired_solv_min2.pdb
3S5Y_chainA_repaired_solv_min3.pdb
Comparison of the three minimization steps. It can clearly be observed, that the backbone is fixed throughout the three steps and the surrounding solvent and ions are changed.
FABRY solv.gif

</figtable>


Trajectory

The test run already delivered 17 states, which can be converted to a trajectory file. This file is graphically shown in <xr id="fig:Trajectory"/>

<figure id="fig:Trajectory">

Trajectory of the first 17 states of the wildtype - test call on partition mpp1_inter

</figure>


Step

In <xr id="fig:Step"/>, it is demonstrated, how the whole solvent box is divided into smaller parts. In this case the files step32242b_n6.pdb, step32242b_n9.pdb, step32242b_n13.pdb and step32242b_n16.pdb are displayed.

<figure id="fig:Step">

Trajectory of the first 17 states of the wildtype - test call on partition mpp1_inter

</figure>