Hi Jun,
Charmm uses a switching function for the VDW while AMBER uses a long range
continuum correction - which don't match obviously. Neither is more wrong
than the other - they are both wrong. ;-)
For sanity turn off any switching function in charmm and in amber set
vdwmeth=0 in the ewald namelist and you should be able to make the VDW
terms match. I don't recommend running with this though - better just to
use the default of (vdwmeth=1) when running the charmm forcefield in AMBER.
Note you should be able to get EELEC to match as well - you just have to
monkey with the ewald coefficient and FFT settings to make sure they are
exactly the same in both codes.
Note AMBER and Charmm do shake in different ways as well so for a perfect
comparison you should turn off shake in both runs.
See:
$AMBERHOME/test/chamber/md_engine/dhfr_cmap_pbc
All the best
Ross
On 8/30/13 2:55 PM, "Jun Feng" <jufeng.mail.wvu.edu> wrote:
>Dear CHAMBER developers:
>I tried to compare energy in AMBER and CHARMM under PBC. My workflow is:
>1. created psf and pdb in CHARMM with c36 FF
>2. use CHAMBER to generate prmtop and inpcrd
>3. run one step energy calculation in CHARMM and pmemd under PBC using
>PME,
>compare energy. cut=9 in AMBER, cton=ctoff=9, cutoff=12 in CHARMM
>Result:
>I tried several systems and VDW energy is significantly differently under
>PBC in all cases. But VDW energy is exactly the same in vacumm without
>PBC(in another thread).
>I provided one output of energy calculation under PBC:
>#CHARMM
>ENER ENR: Eval# ENERgy Delta-E GRMS
>ENER INTERN: BONDs ANGLes UREY-b DIHEdrals
> IMPRopers
>ENER CROSS: CMAPs PMF1D PMF2D PRIMO
>ENER EXTERN: VDWaals ELEC HBONds ASP
>USER
>ENER IMAGES: IMNBvdw IMELec IMHBnd RXNField
> EXTElec
>ENER EWALD: EWKSum EWSElf EWEXcl EWQCor
>EWUTil
> ---------- --------- --------- --------- ---------
> ---------
>ENER> 0-182266.18440 27243.20758 14.18262
>ENER INTERN> 5091.22704 17744.99438 5862.57551 18429.68264
> 312.64340
>ENER CROSS> -45.81685 0.00000 0.00000 0.00000
>ENER EXTERN> 10368.05537-183453.70289 0.00000 0.00000
> 0.00000
>ENER IMAGES> -809.55028 -11082.53004 0.00000 0.00000
> 0.00000
>ENER EWALD> 3219.0147-1409623.2881 1361720.5107 0.0000
>0.0000
>
>#AMBER
>NSTEP = 0 TIME(PS) = 0.000 TEMP(K) = 0.00 PRESS =
>-1222.4
> Etot = -187850.8243 EKtot = 0.0000 EPtot =
>-187850.8243
> BOND = 5091.2270 ANGLE = 17744.9944 DIHED =
>18431.1978
> UB = 5862.5755 IMP = 312.6434 CMAP =
>-45.8167
> 1-4 NB = 3842.7942 1-4 EEL = 28245.3711 VDWAALS =
>129.0351
> EELEC = -267464.8461 EHBOND = 0.0000 RESTRAINT =
>0.0000
> EKCMT = 0.0000 VIRIAL = 24509.4032 VOLUME =
> 928651.5199
> Density =
>1.0114
>
> CHARMM AMBER
>VDW total 9558.55 3971.8293
>ELEC total -239219 -239219.475
>Please also note the difference in dihedrals
>
>I understand that VDW energy is sensitive to cutoff, but can anybody
>explain the HUGE and significant difference in VDW energy? I am really
>puzzled.
>
>Jun Feng
>_______________________________________________
>AMBER mailing list
>AMBER.ambermd.org
>http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 30 2013 - 15:30:03 PDT
