Hi, Urszula. 

I used the following scripts to run NAMD. startloop36.tmp is a template file that namd_submit_cluster will use to general a series of input file like loop_1.inp to run simulations. namd_submit_cluster is the job_submission file. Try to customize loop_1.inp to your system and run it interactively with your NAMD (for me, it is "namd2 loop_1.inp") to see if it works. Let me know if you have further questions:) 

Best,

Yanyan

 thank you I will try it

Dear Yanyan, 

How did you calculate the cellBasisVector 1, 2 and 3 and PMEGridSize ... ? 

cellBasisVector is actually the size of the system.  The xsc file in namd documents the size of the cell if you have run some equilibration or so.  Or, you could load the structure in VMD, and use VMD to calculate it (select all, and get the minimum and maximum along x, y, z and do the subtraction), or any other software you used to visualize the structure.  

PMEGridSize should be integer of times of 2, 3, 5, and it should be also larger than the size you just got or calculated. For example, if your length along x is around 110, then you could take 120 for PMEGridSizeX.  

Thank you for your help. Do you know hot to prepare the file for the backbone harmonic constrainst? I would like to contraint CA,N,C,O atoms. I prepared one that contained ony those atoms but I got an error message: 

Infiniband support is active.

Namd ibverbs will be loaded by 'apps/namd/2.9'.

'compilers/gcc/4.8.0' load complete.

'tools/openmpi/1.6.5-gnu-4.8.0-ib' load complete.

'apps/namd/2.9' load complete.

'tools/scratch/1.0' load complete.

Charmrun> started all node programs in 1.267 seconds.

------------- Processor 0 Exiting: Called CmiAbort ------------

Reason: FATAL ERROR: Number of atoms in constraint constant PDB doesn't

match coordinate PDB

Fatal error on PE 0> FATAL ERROR: Number of atoms in constraint constant

PDB doesn't match coordinate PDB

mpiexec: killing job...

I thought that in the constriant.pdb file I should have atoms that I would like to constraint only ...