[gmx-users] Difference between pbc=full and pbc=xyz

[WU Yanbin]

Hi, Everyone,
  I'm simulating a infinite polymer and so I set "pbc=full". After I shift
some atoms by box size, the energy minimization just broke down. The error
message is just like the following:
  Fatal error: ci = -2147483648 should be in 0 .. 1457.
  Everything is OK before the shift.
  Then where might be the problem?
  And still I donot know the details of "pbc=full". What's the difference
between "pbc=full" and "pbc=xyz"? As I guess the problem might be from here.
  Thanks in advance.
Yours Sincerely,
 WU Yanbin
___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

[gmx-users] construct lipid bilayer from a single lipid

[Chris Neale]

Yesterday I tried to construct a lipid bilayer through a single lipid. The 
process is like this:
1. editconf -f lipid.pdb -princ -c -box 0.8 0.8 3.0 -o lipid.gro
2. genconf -f lipid.gro -nbox 4 4 1 -o box.gro
3. modified vanderwals radii in the vanradii.dat in the top folder ,changed the 
C value to 0.3
4. editconf -f box.gro -box 3.2 3.2 4.5 -o box1.gro
5. genbox -cp box1.gro -cs -o mix.gro
after these steps ,I got a lipid monolayer and water monolayer got together 
,without water in the lipid hydrophobic tails.
However when I ran energy minimization steps, it converged to the precision of computer after only 16 steps and still have 
very high energy. How can I make the starting configuration energy lower?

By the way ,I have tried both steep and cg method, neither worked well.Any 
suggestion will be appreciated.



1. There is a delay between your posting and the appearance of your question on 
the mailing list.
Don't post multiple times to the mailing list. It will only reduce your chances 
of getting a reply.

2. Don't post separately to a particular user. It will only reduce your chances 
of getting a reply.

3. Search the archives. Had you done that, you would already have your 
answer.That answer is:
a) use double precision for energy minimization
b) start md. The acid test of a minimization is your ability to run md without 
crashing.

Look at your strating structure using VMD. in the middle vertically is a row of 
4 tabs. Use the one on the right and
turn on the periodicity. Does it look as you expected?

You mention that you got a lipid monolayer but you said that you wanted a 
bilayer. This could be the reason you got
high energy.

Chris.


___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]

Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [Bulk] 回复:Re: [Bulk] [gmx-users] a qu estion about potential energy

[Yang Ye]

hi,

my suggestion is still the previous one, get g_energy to plot a Energy
vs. time for you so you will know where this average value comes from.

Regards,
Yang Ye

On 8/12/2007 10:34 AM, [EMAIL PROTECTED] wrote:
>
> Dear Yany Ye
>
> Thank you for your help. I am indeed a new man as you guess. I am very
> very sorry about what I did and will take care to never make such stupid
> mistakes.
>
> The simulation I posted is a eq run, and I extracted the
> protein potential energy following what you taught me before(just want
> to check
> the protein). What puzzled me is that the potential energy in every
> frame is almost
> lower than -9(you can see it in protein.xvg in attachment),but the
> average protein
> potential energy in log is just -188 and the RMS-fluctuations is
> 4146(see below).
> How did gromacs get the avergae value and rms value like these?
> According to my
> understanding the average energy should also be about -9. Does my
> md run bad?
> Could you give me a explanation or give me any suggestions? Thanks!
>
> Best regards
>
> Paul Tao
>
>
> Ps: the part of log
>
> <== ### ==>
> < A V E R A G E S >
> <== ### ==>
>
> Energies (kJ/mol)
> G96Angle Proper Dih. Improper Dih. LJ-14 Coulomb-14
> 4.47120e+01 2.05332e+01 1.46051e+01 8.86024e+00 2.87573e+02
> LJ (SR) LJ (LR) Coulomb (SR) Coul. recip. Potential
> -9.50660e+01 -3.65730e+00 -1.01766e+02 -3.63918e+02 -1.88123e+02
> Kinetic En. Total Energy Temperature Pressure (bar)
> 6.82162e+01 -1.19907e+02 6.09296e-01 -1.52389e-01
>
> Box-X Box-Y Box-Z Volume Density (SI)
> 3.03565e-02 3.03565e-02 3.03565e-02 6.82831e+00 1.49606e-01
> pV
> -3.09662e+01
>
> Total Virial (kJ/mol)
> 3.78382e+01 7.27218e-01 -1.68112e-01
> 7.27214e-01 3.77270e+01 -8.52153e-01
> -1.68124e-01 -8.52145e-01 3.91003e+01
>
> Pressure (bar)
> -1.49730e-01 -9.56125e-03 3.1e-03
> -9.56121e-03 -1.44637e-01 5.64105e-03
> 3.10011e-03 5.64098e-03 -1.62801e-01
>
> Total Dipole (Debye)
> -2.13479e+01 -2.18829e+01 -2.22107e+01
>
> <== ### ==>
> < R M S - F L U C T U A T I O N S >
> <== ### ==>
>
> Energies (kJ/mol)
> G96Angle Proper Dih. Improper Dih. LJ-14 Coulomb-14
> 9.89738e+02 4.53448e+02 3.23204e+02 1.94891e+02 6.34017e+03
> LJ (SR) LJ (LR) Coulomb (SR) Coul. recip. Potential
> 2.09723e+03 8.06436e+01 2.24368e+03 8.02594e+03 4.14647e+03
> Kinetic En. Total Energy Temperature Pressure (bar)
> 1.50381e+03 2.64284e+03 1.34318e+01 3.54340e+00
>
> Box-X Box-Y Box-Z Volume Density (SI)
> 6.69374e-01 6.69374e-01 6.69374e-01 1.50584e+02 3.29814e+00
> pV
> 7.19966e+02
>
> Total Virial (kJ/mol)
> 8.49070e+02 8.20897e+01 7.54720e+01
> 8.20897e+01 8.50494e+02 9.04527e+01
> 7.54718e+01 9.04529e+01 8.81896e+02
>
> Pressure (bar)
> 3.57605e+00 8.22075e-01 7.49848e-01
> 8.22075e-01 3.55242e+00 8.78451e-01
> 7.49846e-01 8.78454e-01 3.95141e+00
>
> Total Dipole (Debye)
> 4.70706e+02 4.82556e+02 4.89913e+02
>
> - 原文 -
>
> *From:* Yang Ye 
> *To:* Discussion list for GROMACS users
> 
> *Subject:* Re: [Bulk] [gmx-users] a question about potential energy
> *Sent:* Sat Aug 11 11:41:35 CST 2007
>
>
>
>
> Hi,
>
> You shall help yourself to find the answer by plotting your energy
> with
> g_energy.
>
> Have your equilibrated your system well? Which stage of simulation are
> you running with, em, eq or production run? Having large
> fluctuation is
> sometimes normal and sometimes bad.
>
> Also, please attach necessary and concise information to the mails. We
> may open a table-like file (e.g. xvg) for energy but not a log
> file just
> for energies. I understand that you could be new to gromacs, just take
> note of this.
>
> Regards,
> Yang Ye
>
> On 8/10/2007 3:54 PM, [EMAIL PROTECTED]
>  wrote:
> >
> > Hi, all
> >
> > I have found a problem that in my Md run the potential energy in
> every
> > frame was about -9e+04 while the average potential energy of all was
> > -1.88e+02. What is wrong with it? Why are they so different? Can
> > anybody tell me the reason? Thanks!
> >
> > pS: the attachement is the md.log. You can clearly see that the
> > average energy is so different with the energy in each frame in it.
> >
> > Paul
> >
> >
> >
> >
> 
> > 《天龙八部》续作之“仙人指路”8月15日线上热播！
> >
> 
>  
> >
> >
> > *用搜狗拼音写邮件，体验更流畅的中文输入>>
> > 

[gmx-users] Fatal error: more tahn

[Mu Yuguang \(Dr\)]

Dear all,
When I use distance constrains(NMR), there happens such errors:

Fatal error:
More than 8 graph edges per atom (atom 1)
I have more than 8 constrains on atom 1.

What's reason?

Best regards
Yuguang
 
Dr. Yuguang Mu
Assistant Professor
School of Biological Sciences
Nanyang Technological University
60 Nanyang Drive  Singapore 637551 Tel: +65-63162885 Fax: +65-67913856
http://genome.sbs.ntu.edu.sg/Staff/YGMu/index.php

___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

Re: [gmx-users] Suggestion needed for new workstation

[Martin Höfling]

Am Freitag, 10. August 2007 schrieb Erik Lindahl:

> >> RAM size shouldn't be an issue, except if you're simulating huge
> >> systems.
> >
> > RAM is essential for some analysis tools and if you want to do ab
> > initio
> > calculations.

Ok, i was mainly concerned about doing longer simulations, you're right with 
analysis and probably too with ab initio ;-)

> And remember that there are 8 cores in this machine, so with 8
> processes running in parallel I don't think 8Gb is excessively much.

This probably strongly depends on what you're doing. We have 4 way nodes with 
8G. Our simulations seldom use more than a few percent, so thinking about it 
befor can save you some money and ram can also be upgraded easily.

> However, if you priced it out at apple.com you should be aware that
> Apple charges a _lot_ for memory. FB-DIMMs are expensive, but you can
> often find mac-certified memory for half the price elsewhere (e.g.
> Kingston).

Personally, I never have seen an advantage in "original" ram, no matter if it 
is lenovo or apple or ... whatever. We upgraded our G5 workstation 
with "cheap" ram and many others are also running with it. Reliability might 
make a difference for critical infrastructure but "original" ram can fail 
too.

> > Today: Dual Intel Woodcrest with 3GHz and Infiniband connection, i
> > don`t know

How pricy is infiniband interconnect per node? Is it in range for low end 
cluster systems already? Some years ago. The only "cheap" interconnect 
compared to rest of the hardware was Gigabit, which is probably the worst 
choice if you're doing inter-node calculations.

> > how clovertown architecture will scale with more than 2 nodes
> > (haven`t tested
> > it yet or seen any benchmark results)
> >
> > Later this year i would think about Penryn architecture from Intel.
> > Opteron
> > CPUs are nice, but if you are using mainly gromacs, buy Intel.
>
> I think the new Barcelona Opterons will be very interesting even if
> they are about the same speed as the Intel CPUs.
> First, they should be very competitively prices, and second the
> memory is standard cheap registered DDR2 (ECC), which matters a lot
> when you are buying ~8Gb per dual quad-core machine.

Not to forget the other hardware. Chipsets and so on are pricier on intel arch 
too.

Best and thanks for the comments
Martin
___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

RE: [gmx-users] a basic question

[Yulia Einav]

I had exactly this problem. The reason was I've edited mdp file at
Windows. Sounds silly, I know, but just do your editing for mdp at
Linux, it solved the problem for me.

 

Yulia Einav Ph.D. 

 

 



From: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] On Behalf Of Q733
Sent: Saturday, August 11, 2007 5:36 AM
To: gmx-users@gromacs.org
Subject: [gmx-users] a basic question

 

Hello everyone, I met with a problem today. I tried to simulate dppc
lipid in water.However, when I want to minimize energy ,the grompp
showed the error message like this: "number of coordinates in coordinate
file (box.gro 8253 ) doesn't match  topology ( topol.top , 0 ).

I have used gromacs for sometime and checked the topology and itp file
carefully , but can't find out the problem.can anyone help me ?

here is the mdp file

title= 
cpp  = /lib/cpp -traditional-cpp
include  = none
define  =
constraints= none ;h-bonds
lincs_iter   = 4

 

; RUN CONTROL PARAMETERS = 
integrator   = steep

 

tinit= 0
dt   = 0.001 ; ps
nsteps   = 6000

 

nstxtcout= 100 ; pos to xtc file

 

nstlist = 5
ns_type = grid
rlist  = 1.0
coulombtype= PME
rcoulomb = 1.0
vdw-type = Cut-off
rvdw  = 1.0
fourierspacing   = 0.095 ; default to 0.12
fourier_nx= 0
fourier_ny= 0
fourier_nz= 0
pme_order= 4
ewald_rtol= 1e-5
optimize_fft   = yes

 

; ENERGY MINIMIZATION OPTIONS = 
emtol  = 100.0 ; default to 100
emstep = 0.02  ; default to 0.01

here is the topology file,and I have added the ffgmx forcefield into
"lipid1.itp"

 

#include "lipid1.itp"
#include "dppc.itp"

 

[ moleculetype ]
; molname nrexcl
SOL  2

 

[ atoms ]
;   nr   type  resnr residue  atom   cgnr charge   mass 
 1OW1 SOL OW  1  -0.82   15.99940
 2HW 1 SOLHW1  1   0.411.00800
 3HW 1 SOLHW2 1   0.411.00800

 


[ settles ]
; OW funct doh dhh
  1   1   0.1 0.16330

 

[ exclusions ]
1 2 3
2 1 3
3 1 2

 

[ system ]
; name
DSPC in water

 

[ molecules ]
; name number
 DSP  64
 SOL  1599

 
___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php

[gmx-users] construct lipid bilayer from single lipid (fwd)

[David van der Spoel]

-- Forwarded message --
Date: Sat, 11 Aug 2007 20:12:28 +0800
From: Q733 <[EMAIL PROTECTED]>
To: David van der Spoel <[EMAIL PROTECTED]>
Subject: construct lipid bilayer from single lipid

Dear Professor Spoel:

I have read your steps on how to construct a lipid bilayer from a single lipid 
in the mailing list and I want to repeat it .But I met with some problem,the 
process is like this:
1. editconf -f lipid.pdb -princ -c -box 0.8 0.8 3.0  -o lipid.gro
2. genconf -f lipid.gro -nbox 4 4 1 -o box.gro
3. modified vanderwals radii in the vanradii.dat in the top folder ,changed the 
C value to 0.3
4. editconf -f box.gro -box 3.2 3.2 4.5 -o box1.gro
5. genbox -cp box1.gro -cs -o mix.gro

after these steps ,I got a lipid monolayer and water monolayer getting together 
,without water in the lipid hydrophobic tails.However when I ran energy 
minimization steps, it converged to the precision of computer after only 16 
steps and still have very high energy. How can I make the starting 
configuration energy lower?
By the way ,I have tried both steep and cg method, neither worked well. I tried 
to post this message in the mailing list ,but it didn't appear on the net, I 
just didn't know why.
Will you kindly give me some suggestion?
Thanks very much

Regards

 Shanshan Qin

___
gmx-users mailing listgmx-users@gromacs.org
http://www.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at http://www.gromacs.org/search before posting!
Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to [EMAIL PROTECTED]
Can't post? Read http://www.gromacs.org/mailing_lists/users.php