Re: [gmx-users] Alchemical free energy calculations - right choice of lambda points? What else might I do wrong?

[hannes.loeffler]

Hi,

you wouid _not_ set bonded terms to zero.  That would mean that you have 
essentially a non-bonded end state allowing the atom to be "everywhere" 
(reading Stefan Boresch' papers from 1999 on this may be helpful).  The 
dihedrals are debatable (a collaborator of mine thinks that having them zero 
may help in replicate methods or so).  Our strategy with FESetup 
(http://www.hecbiosim.ac.uk/fesetup) is to copy those terms over from the state 
without the dummies.

The mass-lambdas are best kept at one end state as discussed previously to 
avoid any bad interactions with constraints (you say you have X-C-H3 to 
X-H-DU3).

Cheers,
Hannes.


From: gromacs.org_gmx-users-boun...@maillist.sys.kth.se 
[gromacs.org_gmx-users-boun...@maillist.sys.kth.se] on behalf of Julian 
Zachmann [frankjulian.zachm...@uab.cat]
Sent: 29 July 2015 14:10
To: gmx-us...@gromacs.org
Subject: [gmx-users] Alchemical free energy calculations - right choice of 
lambda points? What else might I do wrong?

Dear Gromacs-Users,

I calculate relative binding energies using free energy perturbation (FEP)
and mbar. I have the binding data for 4 receptors and two ligands which are
almost identical. The only difference is the change from a methyl group to
a hydrogen. I alchemically convert a methyl group into one hydrogen atom
with 3 dummy atoms - inside the receptor and in water. First I have several
equilibration steps; the production run takes 4ns. Unfortunately my results
don't match the experimental results and I would like to ask for input
about what I might do wrong.

First the results:

Experimental | Theoretical (kcal) mbar calculated with pymbar
Receptor 1:0.85 |  0.57
Receptor 2:0.43 | -1.94
Receptor 3:1.13 |   0.78
Receptor 4:2.41 |   0.29

I am using the following lambda points:
;   0  1  2 3
   4   5  6  7 8  9 10
   11121314151617
  18 19
fep_lambdas =  0. 0.2000 0.4000 0.5000 0.6000 0.7000 0.8000
0.9000 1. 1. 1. 1. 1. 1. 1. 1. 1.
1. 1. 1.
vdw_lambdas =  0. 0. 0. 0. 0. 0. 0.
0. 0. 0.2000 0.4000 0.6000 0.7000 0.8000 0.9000 0.9300 0.9600
0.9900 0.9950 1.

I think this might be a mistake as the graphic of free energy differences
 shows there
is a huge difference between lambda points 7-8 - when fep_lambdas turn from
0.9 to 1.0. I think of either introducing another lambda point such as
fep_lambdas = 0.95 or to split up fep_lambdas and converge bonded_lambdas,
coul_lambdas, mass_lambdas differently fast to 1.0. I think
temperature_lambdas and restraint_lambdas are not important for my system
so I my gradually convert them from 0.0 to 1.0 over all 20 lambda points. I
would be very thankful if you could advise me how it would be best to
converge the bonded-, coul-, mass_lambdas, and vdw_lambdas to 1.0.

Other guesses about what I might do wrong are in the mol.itp

 file:

I am not 100% sure, but I think it is correct to put the bond, angle, and
dihedral strengths for dummy atoms to zero.

[ bonds ]
;  aiaj funct  r  k
(...)
   1618 1  1.0920e-01  2.8225e+05  1.0920e-01  2.8225e+05
   1314 1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
   1315 1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
8 9  1  1.0910e-01  2.8342e+05  1.0910e-01  0.
810 1  1.0910e-01  2.8342e+05  1.0910e-01  0.
811 1  1.0910e-01  2.8342e+05  1.0910e-01  0.
712 1  1.0330e-01  3.0878e+05  1.0330e-01  3.0878e+05
4 5  1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
4 6  1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
(...)

[ angles ]
;  aiajak funct  theta   cth
(...)
   131618 1  1.1005e+02  3.8802e+02  1.1005e+02  3.8802e+02
   12 713 1  1.1011e+02  3.8652e+02  1.1011e+02  3.8652e+02
   10 811 1  1.1074e+02  3.2669e+02  1.1074e+02  0.
9 810 1  1.1074e+02  3.2669e+02  1.1074e+02  0.
9 811 1  1.1074e+02  3.2669e+02  1.1074e+02  0.
8 712 1  1.1011e+02  3.8652e+02  1.0811e+02  3.3907e+02
7 8 9 1  1.0791e+02  4.1020e+02  1.0791e+02  0.
7 810 1  1.0791e+02  4.1020e+02  1.0791e+02  0.
7 811 1  1.0791e+02  4.1020e+02  1.0791e+02  0.
71314 1  1.0791e+02  4.1020e+02  1.0791e+02  4.1020e+02
71315 1  1.0791e+02  4.1020e+02  1.0791e+02  4.1020e+02
(...)

[ dihedrals ]
;i  j   k  l func   C0  ...  C5
(...)
12   713   15 3 0.65270 1.95811 0.0-2.61082

[gmx-users] Study of sampling of villin headpiece

[Mario Fernández Pendás]

Dear all,

Due to the computational facilities that I can use I am running 5000 ps
simulations with the villin headpiece system and I am trying to compare the
sampling efficiency of two different methods.

Since for this short simulations, I am not able to observe the folding
phenomena, what would you suggest me to see the difference in sampling
between my two methods?

Thank you very much.

Best,
Mario
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

[gmx-users] Alchemical free energy calculations - right choice of lambda points? What else might I do wrong?

[Julian Zachmann]

Dear Gromacs-Users,

I calculate relative binding energies using free energy perturbation (FEP)
and mbar. I have the binding data for 4 receptors and two ligands which are
almost identical. The only difference is the change from a methyl group to
a hydrogen. I alchemically convert a methyl group into one hydrogen atom
with 3 dummy atoms - inside the receptor and in water. First I have several
equilibration steps; the production run takes 4ns. Unfortunately my results
don't match the experimental results and I would like to ask for input
about what I might do wrong.

First the results:

Experimental | Theoretical (kcal) mbar calculated with pymbar
Receptor 1:0.85 |  0.57
Receptor 2:0.43 | -1.94
Receptor 3:1.13 |   0.78
Receptor 4:2.41 |   0.29

I am using the following lambda points:
;   0  1  2 3
   4   5  6  7 8  9 10
   11121314151617
  18 19
fep_lambdas =  0. 0.2000 0.4000 0.5000 0.6000 0.7000 0.8000
0.9000 1. 1. 1. 1. 1. 1. 1. 1. 1.
1. 1. 1.
vdw_lambdas =  0. 0. 0. 0. 0. 0. 0.
0. 0. 0.2000 0.4000 0.6000 0.7000 0.8000 0.9000 0.9300 0.9600
0.9900 0.9950 1.

I think this might be a mistake as the graphic of free energy differences
 shows there
is a huge difference between lambda points 7-8 - when fep_lambdas turn from
0.9 to 1.0. I think of either introducing another lambda point such as
fep_lambdas = 0.95 or to split up fep_lambdas and converge bonded_lambdas,
coul_lambdas, mass_lambdas differently fast to 1.0. I think
temperature_lambdas and restraint_lambdas are not important for my system
so I my gradually convert them from 0.0 to 1.0 over all 20 lambda points. I
would be very thankful if you could advise me how it would be best to
converge the bonded-, coul-, mass_lambdas, and vdw_lambdas to 1.0.

Other guesses about what I might do wrong are in the mol.itp

 file:

I am not 100% sure, but I think it is correct to put the bond, angle, and
dihedral strengths for dummy atoms to zero.

[ bonds ]
;  aiaj funct  r  k
(...)
   1618 1  1.0920e-01  2.8225e+05  1.0920e-01  2.8225e+05
   1314 1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
   1315 1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
8 9  1  1.0910e-01  2.8342e+05  1.0910e-01  0.
810 1  1.0910e-01  2.8342e+05  1.0910e-01  0.
811 1  1.0910e-01  2.8342e+05  1.0910e-01  0.
712 1  1.0330e-01  3.0878e+05  1.0330e-01  3.0878e+05
4 5  1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
4 6  1  1.0910e-01  2.8342e+05  1.0910e-01  2.8342e+05
(...)

[ angles ]
;  aiajak funct  theta   cth
(...)
   131618 1  1.1005e+02  3.8802e+02  1.1005e+02  3.8802e+02
   12 713 1  1.1011e+02  3.8652e+02  1.1011e+02  3.8652e+02
   10 811 1  1.1074e+02  3.2669e+02  1.1074e+02  0.
9 810 1  1.1074e+02  3.2669e+02  1.1074e+02  0.
9 811 1  1.1074e+02  3.2669e+02  1.1074e+02  0.
8 712 1  1.1011e+02  3.8652e+02  1.0811e+02  3.3907e+02
7 8 9 1  1.0791e+02  4.1020e+02  1.0791e+02  0.
7 810 1  1.0791e+02  4.1020e+02  1.0791e+02  0.
7 811 1  1.0791e+02  4.1020e+02  1.0791e+02  0.
71314 1  1.0791e+02  4.1020e+02  1.0791e+02  4.1020e+02
71315 1  1.0791e+02  4.1020e+02  1.0791e+02  4.1020e+02
(...)

[ dihedrals ]
;i  j   k  l func   C0  ...  C5
(...)
12   713   15 3 0.65270 1.95811 0.0-2.61082
0.0 0.0   0.65270 1.95811 0.0-2.61082
  0.0 0.0
12   713   16 3 0.65270 1.95811 0.0-2.61082
0.0 0.0   0.65270 1.95811 0.0-2.61082
  0.0 0.0
11   8712 3 0.65270 1.95811 0.0-2.61082
0.0 0.0   0.0 0.0 0.0 0.0
  0.0 0.0
11   8713 3 0.65270 1.95811 0.0-2.61082
0.0 0.0   0.0 0.0 0.0 0.0
  0.0 0.0
10   8712 3 0.65270 1.95811 0.0-2.61082
0.0 0.0   0.0 0.0 0.0 0.0
  0.0 0.0
10   8713 3 0.65270 1.95811 0.0-2.61082
0.0 0.0   0.0 0.0 0.0 0.0
  0.0 0.0
98712 3 0.65270 1.95811 0.0-2.6108

Re: [gmx-users] Building a Larger Liquid Phase - Truncate 'Vacuum'

[Elton Carvalho]

On Thu, Jun 25, 2015 at 7:39 PM, Justin Lemkul  wrote:
>
> On 6/25/15 2:46 PM, John Degenstein wrote:
>>
>> I suppose in retrospect that I could simply use a single processor which I
>> think would reduce the number of DD cells and thus prevent this error.
>>
>
> Or just use OpenMP parallelization.  Any time you drastically change the
> size of the system, this will be the case.
>

Is DD treated differently if you run MPI or OpenMP parallelization? If
so, how? I don't recall reading that in the docs and I always assumed
that DD worked distributing one domain per thread (be it OpenMP or
MPI).

Thanks.
-- 
Elton Carvalho
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

[gmx-users] Jumping to other side of box

[Marzieh Saeedi Masineh]

Dear Users, 

I have done a simulation of DMPC lipid bilayer which has 10 drug molecules in 
the water phase for 50 ns. In the first 10 ns of simulation, one of the drug 
molecules diffuses from water into the other side of lipid bilayer and remains 
there. I applied (trjconv –pbc mol), (trjconv –pbc res), ( trjconv –pbc atom) , 
and (trjconv –center this molecule) to turn back this molecule to the first 
side of lipid bilayer. After that I analyzed the distance of this molecule from 
the center of mass of lipid bilayer. It was attached. As you can see, the 
molecule jumps to other side of bilayer and it doesn’t correct though –pbc 
command. What should I do to fix it? 

Thanks in advance for any help, 

Marzieh, 


-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

[gmx-users] a few detailed questions about the topology file format

[Eric Smoll]

Hello Gromacs users,


As far as I understand, values listed in the top file override any
information pulled from share/gromacs/top. However, ff an atom type (e.g.,
"opls_100") serves as a references to information contained in the
share/gromacs/top directory, is it necessary to include a value for the
charge in a itp/top file? It seems that the corresponding entry in
ffnonbonded should supply this information. Going further, what is the
minimal amount of information necessary in the atoms directive?

The gromacs manual includes an example topology for urea. The bonds
directive only lists pairs of atom indices. When the function type is not
listed, does it default to "1"?

Best,
Eric
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

[gmx-users] Settle vs. 3 normal constraints

[Andreas Mecklenfeld]

Dear GROMACS users,

I want to calculate the solvation free energy of a rigid water molecule, 
e.g. TIP4P-Ew.
To do this, I define one solute molecule which I refer to as 
couple-moltype in the .mdp-file (SOL).
The solvens molecules are exactly the same, just named differently for 
distinction (SOLB).


Since I can't simulate two molecule types with settle, I want to use 
constraints. My .itp-file looks like this:




;
; tip4pew Water model
;
[ moleculetype ]
; molnamenrexcl
SOL2

[ atoms ]
; id  at type res nr  res name  at name  cg nr  chargemass
  1   OW_tip4pew  1   SOL   OW   1   0 16.0
  2   HW_tip4pew  1   SOL   HW1  1   0.52422 1.00800
  3   HW_tip4pew  1   SOL   HW2  1   0.52422 1.00800
  4   MW1   SOL   MW   1  -1.04844 0.0



[ constraints ]
1210.09572; func. 1: chemical bond (O-H)
1310.09572; func. 1: chemical bond (O-H)
2320.15139; func. 2: no chemical bond (H-H)



[ virtual_sites3 ]
; Vsite  from  funct   a   b
4   1   2   3   1   0.106676721 0.106676721


[ exclusions ]
1234
2134
3124
4123


; The position of the virtual site is computed as follows:
;
;  O
;
;  V
;
;HH
;
; Ewald tip4p:
; const = distance (OV) / [ cos (angle(VOH)) * distance (OH) ]
;  0.0125 nm/ [ cos (52.26 deg)* 0.09572 nm]
;then a = b = 0.5 * const = 0.106676721
;
; Vsite pos x4 = x1 + a*(x2-x1) + b*(x3-x1)



The temperature derivates by 20 K using the sd-intergrator as thermostat.
By using the settle algorithm for only one molecule (same .mdp and .gro
files as for the constrained), I gain results as expected.

I'm sorry if this is a trivial question.

Thanks for your support!
Andreas


--
M. Sc. Andreas Mecklenfeld
Stipendiat

Technische Universität Braunschweig
Institut für Thermodynamik
Hans-Sommer-Straße 5
38106 Braunschweig
Deutschland / Germany

Tel: +49 (0)531 391-2634
Fax: +49 (0)531 391-7814

http://www.ift-bs.de

--
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] None constraints and l-bfgs.

[Dawid das]

2015-07-29 12:36 GMT+01:00 Justin Lemkul :

> To turn that off you would need "define = -DFLEXIBLE" but you should
> subsequently re-minimize with constraints to avoid weird geometries.



I forgot about it. Thank you!

Dawid Grabarek
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] None constraints and l-bfgs.

[Justin Lemkul]

On 7/29/15 7:30 AM, Dawid das wrote:

Dear Gromacs Experts,

I am trying to run l-bfgs minimization with following options (apart from
others):

integrator   = l-bfgs ; (other: cg, l-bfgs)
constraints  = none ; no constraints

but I still get a message that l-bfgs does not work with constraints so I
should either change to cg or steep (which work) or remove constraints.
Aren't my constraints removed with the keyword above? I also use restraints
but this is a different case, isn't it?



It has nothing to do with restraints.  If you have water molecules, they are 
constrained with SETTLE by default.  To turn that off you would need "define = 
-DFLEXIBLE" but you should subsequently re-minimize with constraints to avoid 
weird geometries.


-Justin

--
==

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==
--
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] Fwd: Question about editconf -noc flag

[Justin Lemkul]

On 7/29/15 2:32 AM, James Lord wrote:

Hi all,
I have made an oil slab (energy minimized, equilibrated nvt and npt), some
of the oil molecules are at the other end (periodicity).
https://drive.google.com/open?id=0B0YMTXH1gmQsYmdvQ3huYWtDaGs

then I decided to increase the box size in z direction a bit to make enough
room for protein, using

editconf -f OCT_npt.gro -oOCT_newbox.gro -box  6.47271 6.47271 16 -noc

https://drive.google.com/open?id=0B0YMTXH1gmQsT1M3dzBNNTJrT28

I used the last flag to put the oil at the ends of box but it is not doing
for one end? any thoughts?


You have broken molecules.  Make them whole with trjconv before resizing the 
box.  Note that with -noc, your slab will always be on one "end" of the box, but 
the distinction between "ends" is irrelevant with periodicity.


-Justin

--
==

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==
--
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] Atom type SDMSO not found

[Justin Lemkul]

On 7/29/15 12:48 AM, su wrote:

Hello everyone
I am doing protein-ligand simulation according to Justin's tutorial. After 
running gmx grompp command, i encountered following error:
Fatal error:
Atomtype SDMSO not found.
Is this the force field matching problem? because i am not able to find out any 
such atom types in any of the generated files.
Please guide me where the problem is. All other commands ran properly.



Some GROMOS force fields use SDMSO, others use SDmso, so it's probably just a 
capitalization issue depending on which parameter set you're using.


-Justin

--
==

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==
--
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] Atom type SDMSO not found

[Justin Lemkul]

On 7/29/15 12:59 AM, Sun Iba wrote:

Here is the generated topology of ligand with PRODRG server, i believe
something is wrong with atom typs here;

  PRODRG COORDS
29
 1UNK  CLAW 1   1.328  -0.638  -0.151
 1UNK  CAM  2   1.384  -0.765  -0.049
 1UNK  SAU  3   1.520  -0.861  -0.089
 1UNK  CAG  4   1.503  -0.954   0.053
 1UNK  CLAV 5   1.606  -1.085   0.095
 1UNK  CAN  6   1.396  -0.913   0.128
 1UNK  HAR  7   1.368  -0.958   0.223
 1UNK  CAO  8   1.327  -0.803   0.069
 1UNK  SAT  9   1.181  -0.728   0.145
 1UNK  OAR 10   1.144  -0.805   0.261
 1UNK  OAS 11   1.215  -0.589   0.171
 1UNK  NAQ 12   1.058  -0.732   0.029
 1UNK  HAS 13   1.052  -0.655  -0.035
 1UNK  CAC 14   0.963  -0.835   0.018
 1UNK  CAD 15   0.837  -0.802  -0.035
 1UNK  HAD 16   0.817  -0.699  -0.065
 1UNK  CAE 17   0.739  -0.899  -0.051
 1UNK  HAE 18   0.642  -0.870  -0.093
 1UNK  CAB 19   0.987  -0.969   0.051
 1UNK  HAB 20   1.085  -0.998   0.090
 1UNK  CAA 21   0.887  -1.065   0.036
 1UNK  HAA 22   0.908  -1.168   0.066
 1UNK  CAF 23   0.759  -1.034  -0.015
 1UNK  NAP 24   0.657  -1.133  -0.031
 1UNK  CAH 25   0.519  -1.090  -0.064
 1UNK  CAI 26   0.421  -1.196  -0.011
 1UNK  CAJ 27   0.447  -1.336  -0.069
 1UNK  CAK 28   0.595  -1.359  -0.104
 1UNK  CAL 29   0.690  -1.276  -0.016
1.33540   1.33540   1.33540

The atom types look strange to me. Please help


Probably because those aren't atom types (they are names) and this isn't a 
topology (it's a coordinate file).


-Justin



On Wed, Jul 29, 2015 at 10:18 AM, su  wrote:


Hello everyone
I am doing protein-ligand simulation according to Justin's tutorial. After
running gmx grompp command, i encountered following error:
Fatal error:
Atomtype SDMSO not found.
Is this the force field matching problem? because i am not able to find
out any such atom types in any of the generated files.
Please guide me where the problem is. All other commands ran properly.

Suniba
Sent from my iPhone


--
==

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalem...@outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==
--
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

[gmx-users] None constraints and l-bfgs.

[Dawid das]

Dear Gromacs Experts,

I am trying to run l-bfgs minimization with following options (apart from
others):

integrator   = l-bfgs ; (other: cg, l-bfgs)
constraints  = none ; no constraints

but I still get a message that l-bfgs does not work with constraints so I
should either change to cg or steep (which work) or remove constraints.
Aren't my constraints removed with the keyword above? I also use restraints
but this is a different case, isn't it?

Best wishes,

Dawid Grabarek
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] on MD simulation of ATP hydrolysis driven protein conformational change

[Ana Sofia Fernandes Oliveira]

Dear Brett:

The answer to your questions is yes. In our lab, we use MD simulations to
study  and identify the ATP-hydrolysis induced conformational changes in
several ABC transporters (both exporters and importers). You can check our
papers if you are interested:


   - "Insights into the Molecular Mechanism of an ABC Transporter:
   Conformational Changes in the NBD Dimer of MJ0796" (
   http://pubs.acs.org/doi/full/10.1021/jp905735y)



   - "Structural consequences of ATP hydrolysis on the ABC transporter NBD
   dimer: Molecular dynamics studies of HlyB" (
   http://onlinelibrary.wiley.com/doi/10.1002/pro.650/abstract)



   - "Conformational changes induced by ATP-hydrolysis in an ABC
   transporter: A molecular dynamics study of the Sav1866 exporter" (
   http://onlinelibrary.wiley.com/doi/10.1002/prot.23023/abstract)



   - "Inter-domain Communication Mechanisms in an ABC Importer: A Molecular
   Dynamics Study of the MalFGK2E Complex" (
   http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1002128
   )


Cheers,
Ana Oliveira

On Tue, Jul 28, 2015 at 1:46 PM, Brett  wrote:

> Dear All,
>
>
> There are some biological processes involving ATP hydrolysis driven
> protein conformational change. By MD, can we simulate the protein global
> conformational change driven by ATP hydrolysis? Or by MD can we investigate
> how the energy stored in ATP converts to protein conformational change?
>
>
> Brett
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-requ...@gromacs.org.
>



-- 
Ana Sofia Fernandes Oliveira
PhD Student
Protein Modeling Group
Instituto de Tecnologia Química e Biológica
Avenida da Republica
Estação Agronómica Nacional
2780-157 Oeiras
Portugal
Phone: (+351) 21 446 91 00
Fax: (+351) 21 441 12 77
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

[gmx-users] request

[fatemeh]

can you please email me and tell where i can see response to my question?
thank you in advance

-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] performance of e5 2630 CPU with gtx-titan GPU

[Netaly Khazanov]

 thanks a lot for your answer.
I  will definitely take a look on this study.
Regards,
Netaly

On Wed, Jul 29, 2015 at 11:22 AM, Kutzner, Carsten  wrote:

> Hi Netaly,
>
> in this study http://arxiv.org/abs/1507.00898 are GROMACS performance
> evaluations for many CPU/GPU combinations. Although your combination
> is not among them, you could try to estimate its performance from
> similar setups.
>
> There is for example an E5-1620 CPU with a TITAN GPU. Although the E5-1620
> has only 4 cores instead of the 6 of the E5-2630 (factor 1.5 difference),
> it is also clocked higher (3.6 GHz as compared to 2.3 GHz, about the same
> factor).
>
> Although there are other differences between the two CPUs, for an estimate
> it should be OK.
>
> Best,
>   Carsten
>
>
> > On 28 Jul 2015, at 20:31, Netaly Khazanov  wrote:
> >
> > Hello All,
> > Does anybody know what is the performance of this combination of CPU and
> > GPU?
> >
> > Thanks in advance.
> >
> >
> >
> > --
> > Netaly
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-requ...@gromacs.org.
>
> --
> Dr. Carsten Kutzner
> Max Planck Institute for Biophysical Chemistry
> Theoretical and Computational Biophysics
> Am Fassberg 11, 37077 Goettingen, Germany
> Tel. +49-551-2012313, Fax: +49-551-2012302
> http://www.mpibpc.mpg.de/grubmueller/kutzner
> http://www.mpibpc.mpg.de/grubmueller/sppexa
>
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-requ...@gromacs.org.
>



-- 
Netaly
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] performance of e5 2630 CPU with gtx-titan GPU

[Kutzner, Carsten]

Hi Netaly,

in this study http://arxiv.org/abs/1507.00898 are GROMACS performance
evaluations for many CPU/GPU combinations. Although your combination
is not among them, you could try to estimate its performance from
similar setups.

There is for example an E5-1620 CPU with a TITAN GPU. Although the E5-1620
has only 4 cores instead of the 6 of the E5-2630 (factor 1.5 difference), 
it is also clocked higher (3.6 GHz as compared to 2.3 GHz, about the same
factor).

Although there are other differences between the two CPUs, for an estimate
it should be OK. 

Best,
  Carsten


> On 28 Jul 2015, at 20:31, Netaly Khazanov  wrote:
> 
> Hello All,
> Does anybody know what is the performance of this combination of CPU and
> GPU?
> 
> Thanks in advance.
> 
> 
> 
> -- 
> Netaly
> -- 
> Gromacs Users mailing list
> 
> * Please search the archive at 
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!
> 
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> 
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
> mail to gmx-users-requ...@gromacs.org.

--
Dr. Carsten Kutzner
Max Planck Institute for Biophysical Chemistry
Theoretical and Computational Biophysics
Am Fassberg 11, 37077 Goettingen, Germany
Tel. +49-551-2012313, Fax: +49-551-2012302
http://www.mpibpc.mpg.de/grubmueller/kutzner
http://www.mpibpc.mpg.de/grubmueller/sppexa

-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.

Re: [gmx-users] Measuring Bilayer thickness with gmx distance in CGMD simulations

[Peter Kroon]

Hi Carlos,

this *should* work. In our lab we usually look at density profiles along
the z-axis; that way you get a sense of distribution. Additionally,
depending on the size of your bilayer, you will need to correct for
undulations (non-trivial).

Peter Kroon

On 28/07/15 20:54, Carlos Navarro Retamal wrote:
> Hi Teemu,
> Thanks for the reply.
> Unfortunately, i cannot upgrade to 5.0.6 right now ( i have to ask to the 
> admin of the cluster first, and so on and so far).
> I do can downgrade to 4.6.3 and use g_dist instead.
> I have read some papers where people claimed to have been used g_dist to 
> measure the thickness of a bilayer, but I wasn’t capable to do that.
> I created 3 different groups:
> All PO4 beads of the bilayer: G1
> PO4 beads of the top leaflet of the bilayer: G2
> PO4 beads of the bottom leaflet of the bilayer: G3
> but considering that g_dist used the COM of the groups when i used g_dist as 
> following:
> g_dist -f POPC-300K-2us.xtc -s POPC-300K-2us.tpr -n index.ndx -o PO4-distance
>
> If i select group 1 and 2 as group G1 i got a distance of 0 (which i think it 
> make sense, considering that g_dist use the COM of a group)
> but when i select group 1: G2 and group 2: G3 (PO4 beads of the top  and 
> bottom leaflet respectively) i got really nice values:
> ~3.9 at 300K
> ~3.8 at 328K
> with an standard deviation of ~0.002 each (considering the Z component, of 
> course)
> Are this results ok? Can i use this approximation  (measuring the distance 
> between the top and bottom PO4 beads of a bilayer) to get the membrane 
> thickness of a bilayer?
> Best,
> Carlos
> --
> Carlos Navarro Retamal
> Bioinformatics Engineering
> Ph. D (c) Applied Sciences.
> Center of Bioinformatics and Molecular Simulations. CBSM
> University of Talca
> Av. Lircay S/N, Talca, Chile
> T: (+56) 712201 798
> E: carlos.navarr...@gmail.com or cnava...@utalca.cl
>
>
>
> On July 28, 2015 at 3:17:47 PM, Teemu Murtola 
> (teemu.murt...@gmail.com) wrote:
>
> Please upgrade to 5.0.6. There was a bug with gmx distance -oxyz that
> caused the output file to not appear (or to overwrite some unintended
> files), which was fixed for the latest version.
>
> On Tue, Jul 28, 2015, 17:34 Carlos Navarro Retamal 
> wrote:
>
>> Hi Justin,
>> thanks for your reply.
>> Is there a way, to analyse each component (x,y,z) of the distance without
>> the flag -oxyz?
>> For some reason, I’m unable to get this data with gmx distance.
>> In fact, when i ran
>> gmx distance -f trajout329K.xtc -s POPC-329K-2us.tpr -n index.ndx -oxyz
>> the command line doesn’t generate anything (any *xvg file).
>> I don’t know what I’m doing wrong
>> Best,
>> Carlos
>> --
>> Carlos Navarro Retamal
>> Bioinformatics Engineering
>> Ph. D (c) Applied Sciences.
>> Center of Bioinformatics and Molecular Simulations. CBSM
>> University of Talca
>> Av. Lircay S/N, Talca, Chile
>> T: (+56) 712201 798
>> E: carlos.navarr...@gmail.com or cnava...@utalca.cl
>>
>>
>>
>> On July 28, 2015 at 9:34:32 AM, Justin Lemkul (jalem...@vt.edu> jalem...@vt.edu>) wrote:
>>
>> July 28, 2015 at 6:09:51 AM GMT-3 ro Retamal wrote:
>>> Dear gmx users,
>>> I’m studying the implication of higher temperatures in the membrane
>> thickness (d) of several bilayer during CGMD simulations.
>>> In order to measure d in a pure POPC membrane i create an index group
>> containing only PO4 beads.
>>> Before everything, i remove PBC conditions in the simulations as
>> following:
>>> trjconv -f POPC-300K-2us.xtc -s POPC-300K-2us.tpr -pbc nojump -o
>> trajout300K.xtc
>>> trjconv -f POPC-329K-2us.xtc -s POPC-329K-2us.tpr -pbc nojump -o
>> trajout329K.xtc
>>> Then, I used gmx distance:
>>> gmx distance -f trajout300K.xtc -s POPC-329K-2us.tpr -n index.ndx -oav
>> avg-300K
>>> gmx distance -f trajout329K.xtc -s POPC-329K-2us.tpr -n index.ndx -oav
>> avg-329K
>>> Finally, using g_analyse i analyse both *xvg files and i got:
>>>
>>> 300K:
>>> Average distance: 4.090 nm
>>> Standard deviation: 1.451 nm
>>>
>>> 329K:
>>> Average distance: 4.134 nm
>>> Standard deviation: 1.515 nm
>>>
>> Those are pretty massive standard deviations; I wouldn't expect a membrane
>> to
>> fluctuate quite that much. Are you averaging only the z-component of the
>> distance?
>>
>>> Is it normal, that I’m getting a higher thickness at higher temperature?
>> Or are my results are ‘ok', considering the values of both standard
>> deviation?
>>
>> Your results are indistinguishable, given the huge standard deviations.
>>
>> -Justin
>>
>>> In any case; I think this is strange, because using a different
>> approximation/methods (GridMAT-MD) i got at 300K, d= 3.996 and at 329K d=
>> 3.869 ( more ‘normal’ results for this type of lipid at this temperatures).
>>> Am i missing something with my approximation that is 'disrupting' my
>> results?
>>> I’ll prefer to use gmx distance to analyse d in my simulations,
>> considering the extensive amount of time that will take me if i use
>> GridMAT-MD to analyse my 1us 

Re: [gmx-users] Issues using tabulated potentials for coarse-grained simulation

[Tamisra Pal]

Many thanks Brian !

On Tue, Jul 28, 2015 at 12:50 AM, Brian Yoo  wrote:

> Hi Tamrisa,
>
> Sorry for the slow response. Was a bit busy this weekend. I'm guessing you
> are using Bhargava and Klein's model? I made a mistake in setting the
> number of exclusions to 3 rather than 2 for their model. After I made that
> change, things ended up working out. If you are using the SDK water model,
> I should mention I also had some issues of the water freezing depending on
> how I set my initial conditions. Annealing the system resolved that issue
> (~400 K for a few ns).
>
> Hope this helps!
>
> Regards,
> Brian
>
> On Saturday, July 25, 2015, Tamisra Pal  wrote:
>
> > Hi Brian :
> >
> > I am also having a similar kind of issue in running a simulation of CG
> > model
> > Ionic liquid molecules having tabulated 9-6 LJ potential . Surprisingly
> , I
> > find that the density steadily increasing even after 30 ns , and the box
> > length consequently decreasing . Though the temperature , pressure are
> OK .
> > In my case I have used tc-grps = system .
> >
> > Can you tell me how to resolve this issue ? Are there any other
> techniques
> >  without using separate temperature couplings for group molecules?
> >
> > Thanks
> >
> > Tamisra
> >
> > --
> >
> >
> >
> >
> > --
> > Tamisra Pal
> >
> > Post Doctoral Research Fellow
> > Technische Universität Darmstadt
> > Institut für Festkörperphysik
> > Hochschulstraße 6
> > 64289 Darmstadt , Germany
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> > posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-requ...@gromacs.org .
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-requ...@gromacs.org.




-- 
Tamisra Pal
-- 
Gromacs Users mailing list

* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting!

* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a 
mail to gmx-users-requ...@gromacs.org.