On Jul 23, 2013, at 5:44 PM, Chinthaka Ratnaweera wrote:
> I mean 0.78 g/ml
>
OK, I assume you have got an adequate starting box structure to begin. I would
find a machine where you can run sander in serial mode overnight, set up a NPT
run and let it go. The system may take several hours to approach the
experimental density, but while the density is equilibrating the system is also
relaxing to the force field. I, personally, would let the computer do the work
instead of spending more effort trying to make the initial stages more efficient.
Unless of course, the objective of your study is finding a better method to
generate solvent boxes. :-)
It is true that the box might work OK if it is not completely equilibrated, but
my feeling is that the sharper the tools, the better job they do (and I treat
the box as a tool.) I remember attending a seminar in the late 80s where the
methodology of the study was MD. The speaker said that all their spare computer
cycles (this was, of course, before the PC era) were spent running Monte Carlo
classical force field simulations of their water boxes. They thought it was
that important. (They were also probably validating their force fields at the
same time, of course.)
Good luck, sounds like you are well on your way.
Bud Dodson
> On Tue, Jul 23, 2013 at 5:43 PM, Chinthaka Ratnaweera <cnr88.msstate.edu>wrote:
>
>> Thank you so much,
>> I have one more question
>> My density seems to be very low around 0.08. The density of ethanol is
>> around 078. That means do I have reduce the starting box size? Otherwise it
>> will take longer time to equilibrate isn't it?
>> Chinthaka
>>
>>
>> On Tue, Jul 23, 2013 at 4:37 PM, M. L. Dodson <mldodson.comcast.net>wrote:
>>
>>>
>>> On Jul 23, 2013, at 3:49 PM, Chinthaka Ratnaweera wrote:
>>>
>>>> Thanks
>>>> I tried to equilibrate what ever the box I have now. ( May be I can try
>>> the
>>>> truncated octahedral later.) But it won't allow me to do that and
>>> request
>>>> velocity info.
>>>> ERROR: I could not find enough velocities in
>>>> ethbox.inpcrd
>>>> Does that mean I must at least minimize few steps?
>>>>
>>>> Chinthaka
>>>>
>>>
>>> Your starting struct. file has no velocities. If it has a box that is no
>>> good. Hmmmm….. Try this: Write a simple shell script (you can just cut
>>> and paste
>>> commands you are using to run sander on multiple lines if you do not know
>>> shell
>>> scripting.) Run a single step minimization and write a restart file.
>>> Use that
>>> restart file as starting structure for a single step dynamics run and
>>> write a
>>> restart file. Copy and paste this two run sequence 15-20 times or so,
>>> sequentially, in your shell script, using the restart file for the
>>> previous run
>>> as the starting structure for the next. (You will have to come up with a
>>> file
>>> naming convention that keeps the files organized.) In my experience,
>>> this, MOST
>>> OF THE TIME, will eventually result in an output without NaN or ****
>>> entries.
>>> Use that restart file to do a regular minimization/NPT equilibration
>>> sequence. Be sure to plot the densities during the NPT equilibration to
>>> know
>>> when it has plateaued. Then I would run a fairly long final MD
>>> equilibration to
>>> be sure the box is relaxed to the force field.
>>>
>>> Good luck
>>>
>>> PS, it might be a good idea to set the temperature to 100K or so during
>>> the
>>> minimization/dynamics alternating sequence. You want enough temp to
>>> allow the
>>> atoms to move but not so much that you get vlimit violations. Remember,
>>> the
>>> only atoms that need to move to get an adequate starting structure are
>>> those on
>>> the box boundary that intrude into the next image.
>>>
>>> PPS, I think there are some tools mentioned in the crystal dynamics
>>> tutorial(s)
>>> that will allow this to work better than Packmol, but I never tried to
>>> use them.
>>> Maybe some enterprising graduate student might write single species and
>>> mixture
>>> periodic box preparation tutorials? :-)
>>>
>>>
>>>>
>>>> On Tue, Jul 23, 2013 at 3:26 PM, M. L. Dodson <mldodson.comcast.net>
>>> wrote:
>>>>
>>>>>
>>>>> On Jul 23, 2013, at 3:10 PM, Chinthaka Ratnaweera wrote:
>>>>>
>>>>>> Ok
>>>>>> I used 'setbox ETHBOX vdw {5,5,5}'
>>>>>> It make a box bigger than previous.
>>>>>>
>>>>>> The box dimensions now read as
>>>>>>> setbox ETHBOX vdw {5,5,5}
>>>>>> Box dimensions: 65.435279 64.286901 64.791138
>>>>>>
>>>>>> my Packmol box should be 50*50*50. Is this size enough now?
>>>>>>
>>>>>> but this gives the same problem in minimization. Are you mean I have
>>> to
>>>>> do
>>>>>> the equilibration at constant pressure without minimization? I try
>>> that
>>>>>>
>>>>>
>>>>> I've been away from this stuff for a couple of years, so all I can
>>> comment
>>>>> is
>>>>> that the boxes set with LEaP and Packmol were not oriented the same
>>> (if I
>>>>> remember correctly.) I figured this out a couple of years ago, but the
>>>>> details
>>>>> are hazy. IIRC, the most foolproof way I found was to create (counter
>>>>> intuitively) an origin-centered sphere with the correct number of
>>>>> molecules.
>>>>>
>>>>> Then enclose the sphere with with a truncated octahedron substantially
>>>>> bigger
>>>>> than the sphere. If your packmol structure is not tightly packed, you
>>> can
>>>>> probably get away with no (or minimal) minimization. Of course the
>>> larger
>>>>> the truncated octahedron "box," the longer the constant pressure run to
>>>>> equilibrate the density.
>>>>>
>>>>> I MAY have had to make sure the trunc. oct. and the sphere were
>>>>> co-centered,
>>>>> I can't remember.
>>>>>
>>>>> After you get an equilibrated restart file, you can cut out an
>>>>> appropriately
>>>>> sized cube with ptraj (IIRC, although can't remember the exact
>>> commands.)
>>>>>
>>>>> Good luck,
>>>>> Bud Dodson
>>>>>
>>>>> PS, another way to do this is to use a PDB with only a single molecule,
>>>>> then
>>>>> "solvate" that molecule with more of the same molecule. This is
>>> probably
>>>>> less trouble than using Packmol for single species boxes. Packmol is
>>>>> needed
>>>>> for mixtures, though, I believe.
>>>>>
>>>>>> On Tue, Jul 23, 2013 at 2:52 PM, M. L. Dodson <mldodson.comcast.net>
>>>>> wrote:
>>>>>>
>>>>>>>
>>>>>>> On Jul 23, 2013, at 2:45 PM, M. L. Dodson wrote:
>>>>>>>
>>>>>>>>
>>>>>>>> On Jul 23, 2013, at 2:19 PM, Chinthaka Ratnaweera wrote:
>>>>>>>>
>>>>>>>>> I followed the procedure in tutorial 15.
>>>>>>>>>
>>>>>>>>> This is my packmol input file
>>>>>>>>>
>>>>>>>>> tolerance 2.0
>>>>>>>>> output ethbox.pdb
>>>>>>>>> filetype pdb
>>>>>>>>> structure ethanol.pdb
>>>>>>>>> number 1000 # Number of molecules
>>>>>>>>> inside cube 0. 0. 0. 50. # x, y, z coordinates of box, and length
>>> of
>>>>>>> box in
>>>>>>>>> Angstroms
>>>>>>>>> end structure
>>>>>>>>>
>>>>>>>>> Then use leap to generate initial coordinates and topologies.
>>>>>>>>> source leaprc.gaff
>>>>>>>>> ETH = loadmol2 ethanol.mol2
>>>>>>>>> ETHBOX = loadpdb ethbox.pdb
>>>>>>>>> setbox ETHBOX centers
>>>>>>>>> saveAmberParm ETHBOX ethbox.prmtop ethbox.inpcrd
>>>>>>>>>
>>>>>>>>> But when running minimization under NVT it gave me this error
>>>>>>>>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>>>>>>>>>
>>>>>>>>> my minimization input is
>>>>>>>>> test run 1
>>>>>>>>> &cntrl
>>>>>>>>> imin = 1, ntpr = 100, ntwx=100, maxcyc = 10000, ntb=1
>>>>>>>>> &end
>>>>>>>>>
>>>>>>>>> This starts the first step of minimization
>>>>>>>>> TESTING RELATIVE ERROR over r ranging from 0.0 to cutoff
>>>>>>>>> | CHECK switch(x): max rel err = 0.2738E-14 at 2.422500
>>>>>>>>> | CHECK d/dx switch(x): max rel err = 0.8314E-11 at 2.736960
>>>>>>>>> ---------------------------------------------------
>>>>>>>>> | Local SIZE OF NONBOND LIST = 1264139
>>>>>>>>> | TOTAL SIZE OF NONBOND LIST = 1264139
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> NSTEP ENERGY RMS GMAX NAME
>>> NUMBER
>>>>>>>>> 1 2.1423E+09 4.7538E+08 4.7855E+10 H4
>>> 2698
>>>>>>>>>
>>>>>>>>> BOND = 547891.8779 ANGLE = 201752.1037 DIHED =
>>>>>>>>> 2455.8141
>>>>>>>>> VDWAALS = ************* EEL = 12317.8836 HBOND =
>>>>>>>>> 0.0000
>>>>>>>>> 1-4 VDW = 72044523.0714 1-4 EEL = -22796.8084 RESTRAINT =
>>>>>>>>> 0.0000
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Is this because of my box size?
>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>> Probably not. Boxes created with packmol are not equilibrated with
>>>>>>> periodic boundaries.
>>>>>>>> Packmol just puts packs spheres (connected spheres in the case of
>>>>>>> ethanol.) At the
>>>>>>>> box boundaries, parts of molecules extend beyond the box. When the
>>>>> boxes
>>>>>>> are tiled,
>>>>>>>> some of the atoms overlap with the next image so you get VDW
>>> clashes.
>>>>>>> You need a box
>>>>>>>> bigger than the packmol box when you tile the boxes, then
>>> equilibrate
>>>>> at
>>>>>>> constant
>>>>>>>> pressure to shrink the box, equilibrate the density, and arrange the
>>>>> box
>>>>>>> edges to be
>>>>>>>> consistent with periodic boundaries.
>>>>>>>>
>>>>>>>
>>>>>>> Let me make this less ambiguous.
>>>>>>>
>>>>>>> Your Packmol box is not the problem, the box assigned by LEaP is too
>>>>> small
>>>>>>> if it is
>>>>>>> the same as the Packmol box. Boxes created with packmol are not
>>>>>>> equilibrated with
>>>>>>> periodic boundaries. Packmol just puts packs spheres (connected
>>> spheres
>>>>> in
>>>>>>> the case
>>>>>>> of ethanol.) At the Packmol box boundaries, parts of molecules extend
>>>>>>> beyond the box.
>>>>>>> When the boxes are tiled with LEaP, some of the atoms overlap with
>>> the
>>>>>>> next image so
>>>>>>> you get VDW clashes. You need a LEaP box bigger than the packmol box
>>>>> when
>>>>>>> you tile
>>>>>>> the boxes, then equilibrate at constant pressure to shrink the PBC
>>> box,
>>>>>>> equilibrate
>>>>>>> the density, and arrange the PBC box edges to be consistent with
>>>>> periodic
>>>>>>> boundaries.
>>>>>>>
>>>>>>>> Good luck,
>>>>>>>> Bud Dodson
>>>>>>>>
>>>>>>>>> On Tue, Jul 23, 2013 at 10:02 AM, David A. Case <
>>>>>>> case.biomaps.rutgers.edu>wrote:
>>>>>>>>>
>>>>>>>>>> On Mon, Jul 22, 2013, Chinthaka Ratnaweera wrote:
>>>>>>>>>>>
>>>>>>>>>>> Once the saveoff command make a box 'unit' ( I guess I have to
>>> save
>>>>>>> it
>>>>>>>>>> as
>>>>>>>>>>> a lib file), inorder to use that unit to solvate the protein, do
>>> I
>>>>>>> have
>>>>>>>>>> to
>>>>>>>>>>> use loadoff command to open it and give this unit name as second
>>>>>>> argument
>>>>>>>>>>> in solvatebox command?
>>>>>>>>>>
>>>>>>>>>> Yes.
>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Does size of the solvate box matters when setup the initial box
>>>>> using
>>>>>>>>>>> packmol?
>>>>>>>>>>
>>>>>>>>>> It's not a big deal which size you choose. Most of the "standard"
>>>>>>> boxes
>>>>>>>>>> are really smaller than they should be (some were made back in the
>>>>>>> 1980s
>>>>>>>>>> when
>>>>>>>>>> computers were much slower.) The solvateBox and solvateOct
>>> routines
>>>>>>> will
>>>>>>>>>> replicate and truncate boxes as needed. Something on the order of
>>>>> 1000
>>>>>>>>>> ethanol molecules should be fine.
>>>>>>>>>>
>>>>>>>>>> ...good luck....dac
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> _______________________________________________
>>>>>>>>>> AMBER mailing list
>>>>>>>>>> AMBER.ambermd.org
>>>>>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> --
>>>>>>>>> Chinthaka Nadun Ratnaweera
>>>>>>>>> Hand Lab Rm 1126
>>>>>>>>> Mississippi State University
>>>>>>>>> 310 Presidents Circle
>>>>>>>>> Starkville, MS 39762
>>>>>>>>> _______________________________________________
>>>>>>>>> AMBER mailing list
>>>>>>>>> AMBER.ambermd.org
>>>>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>>>>
>>>>>>>> --
>>>>>>>> M. L. Dodson
>>>>>>>> Business email: activesitedynamics-at-gmail-dot-com
>>>>>>>> Personal email: mldodson-at-comcast-dot-net
>>>>>>>> Gmail: mlesterdodson-at-gmail-dot-com
>>>>>>>> Phone: eight_three_two-five_63-386_one
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> AMBER mailing list
>>>>>>>> AMBER.ambermd.org
>>>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>>>
>>>>>>> --
>>>>>>> M. L. Dodson
>>>>>>> Business email: activesitedynamics-at-gmail-dot-com
>>>>>>> Personal email: mldodson-at-comcast-dot-net
>>>>>>> Gmail: mlesterdodson-at-gmail-dot-com
>>>>>>> Phone: eight_three_two-five_63-386_one
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> AMBER mailing list
>>>>>>> AMBER.ambermd.org
>>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>> --
>>>>>> Chinthaka Nadun Ratnaweera
>>>>>> Hand Lab Rm 1126
>>>>>> Mississippi State University
>>>>>> 310 Presidents Circle
>>>>>> Starkville, MS 39762
>>>>>> _______________________________________________
>>>>>> AMBER mailing list
>>>>>> AMBER.ambermd.org
>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>
>>>>> --
>>>>> M. L. Dodson
>>>>> Business email: activesitedynamics-at-gmail-dot-com
>>>>> Personal email: mldodson-at-comcast-dot-net
>>>>> Gmail: mlesterdodson-at-gmail-dot-com
>>>>> Phone: eight_three_two-five_63-386_one
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> AMBER mailing list
>>>>> AMBER.ambermd.org
>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> Chinthaka Nadun Ratnaweera
>>>> Hand Lab Rm 1126
>>>> Mississippi State University
>>>> 310 Presidents Circle
>>>> Starkville, MS 39762
>>>> _______________________________________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>> --
>>> M. L. Dodson
>>> Business email: activesitedynamics-at-gmail-dot-com
>>> Personal email: mldodson-at-comcast-dot-net
>>> Gmail: mlesterdodson-at-gmail-dot-com
>>> Phone: eight_three_two-five_63-386_one
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>
>>
>>
>> --
>> Chinthaka Nadun Ratnaweera
>> Hand Lab Rm 1126
>> Mississippi State University
>> 310 Presidents Circle
>> Starkville, MS 39762
>>
>
>
>
> --
> Chinthaka Nadun Ratnaweera
> Hand Lab Rm 1126
> Mississippi State University
> 310 Presidents Circle
> Starkville, MS 39762
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
--
M. L. Dodson
Business email: activesitedynamics-at-gmail-dot-com
Personal email: mldodson-at-comcast-dot-net
Gmail: mlesterdodson-at-gmail-dot-com
Phone: eight_three_two-five_63-386_one
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Jul 23 2013 - 17:00:02 PDT
