Difference between revisions of "Task 8 - Jorunal (PKU)"
From Bioinformatikpedia
(Created page with "==Job submission == The following script was used to submit AGroS to the queue. For the two mutants this worked flawlessly, the wildtype required a bit more work. #!/bin/bash …") |
(→Job submission) |
||
| Line 1: | Line 1: | ||
==Job submission == |
==Job submission == |
||
The following script was used to submit AGroS to the queue. For the two mutants this worked flawlessly, the wildtype required a bit more work. |
The following script was used to submit AGroS to the queue. For the two mutants this worked flawlessly, the wildtype required a bit more work. |
||
| + | <source lang="bash"> |
||
| − | |||
#!/bin/bash |
#!/bin/bash |
||
#SBATCH -o $HOME/logs/<NAME>_MD.out |
#SBATCH -o $HOME/logs/<NAME>_MD.out |
||
| Line 18: | Line 18: | ||
cd $HOME/ |
cd $HOME/ |
||
$HOME/AGroS/AGroS $HOME/Gromacs/Models/*.pdb -dir $HOME/MD/ -threads 32 --scwrlPATH $HOME/SCWRL |
$HOME/AGroS/AGroS $HOME/Gromacs/Models/*.pdb -dir $HOME/MD/ -threads 32 --scwrlPATH $HOME/SCWRL |
||
| + | </source> |
||
| − | |||
Since the wildtype repeatedly had problems equilibrating and minimizing and broke the script several times, we took the intermediate results of AGroS after a breakdown, checked 1J8U_solv_min.pdb briefly for errors and called the next preparation steps of AGroS by hand: |
Since the wildtype repeatedly had problems equilibrating and minimizing and broke the script several times, we took the intermediate results of AGroS after a breakdown, checked 1J8U_solv_min.pdb briefly for errors and called the next preparation steps of AGroS by hand: |
||
<source lang="bash"> |
<source lang="bash"> |
||
Revision as of 12:44, 29 June 2012
Job submission
The following script was used to submit AGroS to the queue. For the two mutants this worked flawlessly, the wildtype required a bit more work. <source lang="bash">
#!/bin/bash #SBATCH -o $HOME/logs/<NAME>_MD.out #SBATCH -D $HOME/MD/ #SBATCH -J 1J8U_wildtype #SBATCH --clusters=mpp1 #SBATCH --get-user-env #SBATCH --ntasks=32 #SBATCH --mail-type=end #SBATCH --mail-user=<email> #SBATCH --export=NONE #SBATCH --time=10:00:00 source /etc/profile.d/modules.sh module load gromacs cd $HOME/ $HOME/AGroS/AGroS $HOME/Gromacs/Models/*.pdb -dir $HOME/MD/ -threads 32 --scwrlPATH $HOME/SCWRL
</source> Since the wildtype repeatedly had problems equilibrating and minimizing and broke the script several times, we took the intermediate results of AGroS after a breakdown, checked 1J8U_solv_min.pdb briefly for errors and called the next preparation steps of AGroS by hand: <source lang="bash">
repairPDB 1J8U_solv_min.pdb -jprot > 1J8U_solv_min.pdb2 genrestr -f 1J8U_solv_min_0.pdb -o posre.itp grompp -v -f minim_solv.mdp -c 1J8U_solv_min.pdb -p 1J8U_solv.top -o 1J8U_solv_min2.tpr mdrun -v -deffnm 1J8U_solv_min2.tpr -c 1J8U_solv_min2.pdb #create minim_solv2.mdp according to AGroS grompp -v -f minim_solv2.mdp -c 1J8U_solv_min2.pdb -p 1J8U_solv.top -o 1J8U_solv_min3.tpr mdrun -v -deffnm 1J8U_solv_min3.tpr -c 1J8U_solv_min3.pdb converged in 4519 steps #create nvt.mdp according to AGroS grompp -v -f nvt.mdp -c 1J8U_solv_min3.pdb -p 1J8U_solv.top -o 1J8U_nvt.tpr mdrun -v -deffnm 1J8U_nvt #create npt.mdp according to AGroS grompp -v -f npt.mdp -c 1J8U_nvt.gro -p 1J8U_solv.top -o 1J8U_npt.tpr mdrun -v -deffnm 1J8U_npt #create md.mdp according to AGroS grompp -v -f md.mdp -c 1J8U_npt.gro -p 1J8U_solv.top -o 1J8U_md.tpr
</source> The final run was submited to the queue with the following script:
<source lang="bash">
- !/bin/bash
- SBATCH -o $HOME/logs/1J8U_wildtype_MD.out
- SBATCH -D $HOME/testrun/
- SBATCH -J 1J8U_wildtype
- SBATCH --clusters=mpp1
- SBATCH --get-user-env
- SBATCH --ntasks=32
- SBATCH --mail-type=end
- SBATCH --mail-user=<email>
- SBATCH --export=NONE
- SBATCH --time=10:00:00
source /etc/profile.d/modules.sh module load gromacs export PATH="$HOME/AGroS:$PATH" export PATH="$HOME/AGroS:$PATH" cd $HOME mpiexec -n 32 mdrun_mpi -v -deffnm $HOME/MD/mdrun_1J8U/1J8U_md </source>
