hbond and nonbond-spec in dynamics

Posted by: pmj

hbond and nonbond-spec in dynamics - 03/06/19 10:13 AM

Dear all,

can someone tell me what is the usual input part for the hbond and nonbond specifications?

DYNAMICS LEAP VERLET RESTART(*) NSTEP 20000 TIMESTEP 0.001(+) -
IPRFRQ 1000 IHTFRQ 1000 IEQFRQ 5000 NTRFRQ 5000 -
IUNREA 30 IUNWRI 31 IUNCRD 50 IUNVEL -1 KUNIT 70 -
NPRINT 100 NSAVC 100 NSAVV 0 INBFRQ 25 -
hbond-spec nonbond-spec -
FIRSTT 100.0 FINALT 300.0 TEMINC 100.0 -
IASORS 1 IASVEL 1 ISCVEL 0 ICHECW 0 TWINDH 10.0 TWINDL -10.0

Thank you
Paula
Posted by: rmv

Re: hbond and nonbond-spec in dynamics - 03/06/19 12:47 PM

The HBOND term in the potential has been deprecated for a couple decades, and any hbond-spec keywords should not be used.

The default nbond-spec for the C36 protein force field (par_all36_prot.prm) is

NONBONDED nbxmod 5 atom cdiel fshift vatom vdistance vfswitch -
cutnb 14.0 ctofnb 12.0 ctonnb 10.0 eps 1.0 e14fac 1.0 wmin 1.5


In general, these defaults should be used for most solvated simulations, with the notable exception that the electrostatic part (FSHIFT) should be overridden by PME keywords. Rather than clutter up the DYNA command, I generally precede that with an UPDATE command, such as

update cutim 14. imgfrq -1 cutnb 14. inbfrq -1 -
bycb atom vatom vfswitch ctofnb 12. ctonnb 10. -
ewald pme spline order 6 fftx @FX ffty @FX fftz @FX kappa 0.34


(The variable @FX is an integer close to the edge length, required for the PME grid.)

I have posted some more complete examples in the Script Archive forum.
Posted by: pmj

Re: hbond and nonbond-spec in dynamics - 03/07/19 04:38 AM

Dear prof. Venable,

thank you for you answer. Can you please suggest what nonbonded specifications one should use for normal dynamics and then for SGLD in case implicit solvent is used with approximately 120 000 atoms?

Would

NBOND atom switch cdie vdw vswitch cutnb 20.0 ctofnb 16.0 ctonnb 15.0

be appropriate?

Thank you
Paula
Posted by: rmv

Re: hbond and nonbond-spec in dynamics - 03/07/19 10:49 AM

The documentation for the implicit models typically has recommended nonbond settings, which should probably be used.

I don't use implicit models (or SGLD) much; perhaps someone who uses these methods might respond and offer some suggestions.
Posted by: pmj

Re: hbond and nonbond-spec in dynamics - 03/08/19 08:49 AM

Dear pof. Venable,

thank you for all your suggestions and help. I will try to find more information in the documentation and online.

Have a great day!

Paula