Re: Patch-size Error (not like the other ones)

From: Jeffrey Potoff (jpotoffx_at_gmail.com)
Date: Fri Jul 24 2015 - 13:38:52 CDT

What is happening with your box dimensions (in the XSC file)? If you
print on a relatively small interval, to you see them decreasing, or
remaining constant? What does your simulation look like when you
visualize a pre-patch grid error set of coordinates in VMD? My gut
feeling is that you probably don't have enough water in the system to
get to a stable equilibrium density, and are likely forming "bubbles" as
the water condenses. Your water density is low. It should be closer to
0.99 g/cm^3, not 0.89 g/cm^3.

-- 
----
Jeffrey Potoff
Professor & Director of Early Engineering Programs
Department of Chemical Engineering and Materials Science
Wayne State University			
http://potoff1.eng.wayne.edu
http://gomc.eng.wayne.edu
On 07/24/2015 02:14 PM, Merrill, Dante wrote:
>
> Hello,
>
> I am well aware of the multitude of patch-size errors on the mailing 
> list and I have read through the majority of them and have yet to find 
> a good answer that fits my issue. Please just read the next paragraph 
> and then judge whether this is like the other issues posted to the 
> mailing list.
>
> I am attempting to correctly equilibrate and produce a relatively 
> small protein (~50,000 atoms) in a water box (currently ~15,000) for a 
> total of 100 ns using 2 fs timesteps using NPT (config file for 
> production run 1 below). I have added margin, turned Flexiblecell on 
> and off (and its two counterparts), restarted, elongated minimization 
> and equilibration, checked the water density (its at 0.89 g/cm^3), 
> resolvated, moved the origin to 0,0,0 , and messed with pairlistdist 
> and switchdist settings.
>
> I can get through my 10,000 step minimization, and 4 200,000 step 
> equilibrations just fine. However I cannot for the life of me get my 
> production run to go past 6 million steps without having to restart. 
> Which would be annoying but do able if not for the fact that after 
> every restart my patch size decreases so by restart 4 I am getting 
> about 200,000 steps per run.
> I understand that you receive a lot of emails about this issue and I 
> am sorry to have to add to that however I feel I have exhausted my 
> resources in trying to figure this out myself and I humbly ask for 
> your help with this issue. Any guidance as to how to fix this or even 
> just elongate my runs would be greatly appreciated.
>
>
> Thank you for your time and efforts.
>
>
> Best,
>
> Dante
>
>
>
> timestep             2.0
> numsteps             50000000 # 100000000fs = 100000ps = 100ns
>
> useConstantRatio     yes
> useflexiblecell      yes
> useConstantArea       no
>
> #structure ../scr/syt1_pops_prot_final.pdb.xplor.psf
> structure            ../scr/final2_slp4.psf
> paraTypeCharmm       on
>
>
> # INPUT
> #coordinates ../scr/syt1_pops_prot_final.pdb.solv.pdb
> #bincoordinates          ../scr/equilibration/equil5.coor
> #binvelocities           ./scr/equilibration/equil5.vel
> coordinates          ../scr/equilibration/equil4.coor
> velocities           ../scr/equilibration/equil4.vel
> extendedsystem       ../scr/equilibration/equil4.xsc
>
> # OUTPUT
> outputenergies       1000
> outputtiming         1000
> outputpressure       1000
> binaryoutput         no
>
> outputname           ../scr/production/prod1
> restartname          ../scr/production/prod1
> restartfreq          10000
> binaryrestart        yes
>
> XSTFreq              1000
>
> dcdfile              ../dcd/prod1.dcd
> dcdfreq              1000
>
> # CUT-OFFs
> vdwForceSwitching    on
> splitpatch           hydrogen
> hgroupcutoff         3.0
> stepspercycle        10
> switching            on
> switchdist           11.0
> cutoff               12.0
> pairlistdist         20.0
> pairlistsPerCycle    2
> # H-MATRIX
>
> # CONSTANT-T
> langevin             on
> #langevintemp         310.0
> langevintemp         298.0
> langevindamping      1.0
>
> # CONSTANT-P
> langevinpiston       on
>
>
> langevinpistontarget 1
> langevinpistonperiod 200
> langevinpistondecay  50
> #langevinpistontemp   310
> langevinpistontemp   298
>
> strainrate           0.  0.  0.
> usegrouppressure     yes
>
> # PME
> PME                  yes
> PMETolerance         10e-5
> PMEInterpOrder       4
> PMEGridSpacing        1.0
>
> # MULTIPLE TIME-STEP
> fullelectfrequency   2
> nonbondedfreq        1
>
> # SHAKE/RATTLE
> rigidbonds           all
> rigidtolerance       0.00001
> rigiditerations      400
>
> # 1-4's
> exclude              scaled1-4
> 1-4scaling           1.0
>
>
> #parameters           ../lib/par_all36_na.prm
> #parameters           ../lib/par_all36_lipid.prm
> parameters           ../lib/par_all36_prot.prm
> parameters           ../lib/toppar_water_ions.str
>
>
>

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:22:00 CST