Re: Equilibration of the membrane-protein system done in CHARMM-GUI

From: Sunhwan Jo (sunhwan_at_uchicago.edu)
Date: Wed Nov 06 2013 - 09:29:54 CST

I don't think different barostat can prevent diffusion. What you need is cancellation of translational momentum only for your protein along XY dimension, but I do not know if there is any option for that in NAMD.

by the way what exactly role of the restrains in the equilibration steps (some of them are defined via ColVars as I noticed)?

3 sets of restraints were used.

1. protein backbone/sidechain restraint
2. planar restraint that holds lipid head/tail groups in plane
3. dihedral restraints that retain chirality of lipid head group and double bonds

These are used and gradually removed during STEP6. For more information, please read our manuscripts or tutorial.

Best,
Sunhwan

On Nov 6, 2013, at 9:02 AM, James Starlight <jmsstarlight_at_gmail.com<mailto:jmsstarlight_at_gmail.com>> wrote:

Sunhwan.

the RMSD have gradually increased during x-y diffusion and then stabilized when protein have reached position. I suppose that Its more appropriate to measure msd => Diffusion coefficent (not RMSD) from such trajectory but I cant find this possibility in namd. By the way what barostat options could prevent such diffusion ? ( I've tried to increase constant for coupling to P_bath but there were no any changing).

by the way what exactly role of the restrains in the equilibration steps (some of them are defined via ColVars as I noticed)?

James

2013/11/6 Sunhwan Jo <sunhwan_at_uchicago.edu<mailto:sunhwan_at_uchicago.edu>>
James,

STEP7 inputs provided by CHARMM-GUI comes with no restraints. Under such condition, I believe diffusion along membrane is normal. You should be able to recenter the protein from the trajectory if needed.

Your comment about increased RMSD is interesting, though. Are you calculating RMSD after reorientation?

Best,
Sunhwan

On Nov 6, 2013, at 12:51 AM, James Starlight <jmsstarlight_at_gmail.com<mailto:jmsstarlight_at_gmail.com>> wrote:

> Dear all,
>
>
> As I've mentioned I had problems with the simulation of the protein-membrane complex made in Charm-gui. Briefly I had no problems during all equilibration phases but on the 7.1 ptoduction run step l've observed the diffusion of the protein as the whole (!!!) in the x-y plane of the membrane (analysis of the RMSD provides me hight increase in RMSD (up to 10A) during first 2ns when such 2D diffusion have been detected). I'm not sure if this simulation was OK because I've never seen such motion in X-Y plane (previously making long simulation in Gromacs with Langevins dynamics and Parinello's barostat). As I've mentioned such motion is observed during 7.1 step ( here any restraints are removed from the conf file like
> # planar restraint
> colvars on
> exec sed -e "s/Constant \$fc/Constant 0/g" membrane_lipid_restraint.namd.col > restraints/$outputname.col
> colvarsConfig restraints/$outputname.col
>
> # dihedral restraint
> extraBonds yes
> exec sed -e "s/\$FC/0/g" restraints/dihe.txt > restraints/$outputname.dihe
> extraBondsFile restraints/$outputname.dihe
>
> Does I need more prolonged equilibration in case where I simulate protein inserted in the membrane ( in comparison to the pure bilayer) or may be some additional restrains should be included in the 7.1 production run as well ?
>
> James

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:23:57 CST