restraint with colvars mocule

From: Ling Wu (emilia.l.wu_at_gmail.com)
Date: Sun Sep 11 2011 - 16:25:39 CDT

Hi All,

I have a question regarding the colvars function of NAMD.

I am trying to run a simulation of a pure POPC membrane system. I want to enforce a positional restrain on the head group and the tail group of POPC to constrain them at a certain area along Z-axis for the initial equilibration. After I apply this restrain on the system, I checked if the center of mass of the head group or tail group is in the area I want (head: +19 and -19; tail: between +5 and -5), but what I found is the membrane atoms are all shift up after 25ps MD, even though the center of the entire system (membrane+water+ion) is 0, 0, 0, but the center of the membrane shift to like 0, 0, 2. when I looked at the structure in VMD, it is obvious that there are thinner layer of water on the top of the membrane, and a much thicker layer of water on the bottom. I also found the bigger force constant I used, the more the center will shift up. if I remove the colvar part from my input file, then there is no prolem.

here is my ColvarsConfig,

##Lipid Head###
 
Colvarstrajfrequency 100
Colvarsrestartfrequency 100
 
colvar {
   name head_POPC
   distanceZ {
      ref {
         dummyAtom ( 0.000, 0.000, 0.000 )
      }
       main {
             atomsFile head_POPC_main.ref # head atoms #
             atomsCol B
             atomsColValue 1.0
         }
   }
}
harmonic {
  colvars head_POPC
  centers 19 # or wherever the head group should be (15, 18, 17...) #
  forceConstant 5
}
 
 
###lipid tail###
 
colvar {
    name tail
    width 0.1
    lowerboundary -10.0
    upperboundary 10.0
    lowerWall -5.0
    upperWall 5.0
    lowerWallConstant 5
    upperWallConstant 5
    distanceZ {
        main {
           atomsFile tail_main.ref
             atomsCol B
             atomsColValue 1.0
         }
        ref {
           dummyAtom ( 0.000, 0.000, 0.000 )
         }
          axis (0.0, 0.0, 1.0)
    }
 }

Can anyone give me some suggestions here, I am really confused. Thanks a lot!

Emilia

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:57:44 CST