[gmx-users] problem with g_density -center

[Christopher Neale]

The solution that I have posted with modified trjconv -center gives the correct 
values. However, you should always
check such things because at the end of the day it is your name on the paper so 
you are the one responsible.

If you want to reproduce a previous result, just contact them to find out 
exactly what methods they used
and use those same methods.

Chris.

 -- original message --

Ali Alizadeh ali.alizadehmojarad at gmail.com 
Fri Oct 19 20:22:54 CEST 2012
Previous message: [gmx-users] problem with g_density -center
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Dear Chris

1- Can you get correctly density profile by own method?

I will use your method and i hope to get  correctly symmetric density
profile for my symmetric system.

2- One question, if i find out that the paper (has published 2009) has
used which of old version of gromacs,

Can i hope to reproduce its results?

-- 
Sincerely
--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] A favor question: experience running Gromacs in the cloud

[Joris Poort]

At Rescale (www.rescale.com) we have cloud-based options for 10Gb/s
I/O to run mpi versions of Gromacs.  The performance for these has
been very good and can scale to hundreds of simulation dynos
(computing nodes).

Regardless of which service you use, I would highly recommend trying
some of the cloud options.  Cloud-based computing is a very cost
effective way to get your results fast without any upfront investment
and you only pay for what you use.

Cheers,
Joris

Joris Poort
Rescale
jo...@rescale.com

On Fri, Oct 19, 2012 at 9:31 AM, Szilárd Páll  wrote:
> Hi,
>
> When it comes to hardware, pretty much the only thing that matters is
> processors (CPUs and maybe GPUs) , memory will always be more than enough
> -- unless you plan to run massive analysis in the cloud. In general, if
> your simulation system runs fast enough on a single compute node (1-2
> CPUs), the could could very well be worth it. When it comes to running over
> a network, I'm not sure what the cloud-based options are, but I think it is
> not not worth bothering with anything slower than 10G Ethernet or
> Infiniband.
>
> If you are not after a few long trajectories, but you're doing
> some ensemble-simulations and need lots of independent trajectories, the
> could will be quite well suited, especially if you can manage the
> simulations in an elegant manner, e.g. see http://copernicus-computing.org.
>
> I have no first hand concrete experience, but I know that some GROMACS
> developers have been testing Amazon EC2 instances with pretty positive
> results. You can get really low prices and good performance with the
> "High-CPU Spot Instances".
>
> Just to make it clear, I do not mean to advertise Amazon EC2, just that
> I've only heard about first-hand experiences on this resource. There are
> quite a few other cloud providers, so you might want compare and try before
> buying!
>
> Cheers,
> --
> Szilárd
>
>
> On Fri, Oct 19, 2012 at 1:59 AM, Andrew DeYoung 
> wrote:
>
>> Hi, Gromacs users,
>>
>> If you have time, I am wondering if you have any advice, as a favor.
>>
>> I'm a graduate student in computational chemistry in the U.S.  My
>> department
>> has an EXCELLENT "in-house" cluster, on which I routinely run Gromacs jobs
>> on 8 to 24 cores.  However, it occasionally happens that I run into a
>> "crunch time" need for some more computational resources.  For example, I
>> recently found a mistake (an incorrect parameter) that I made in an .itp
>> file that I wrote.  This means that I need to rerun a whole slew of
>> simulations as quickly as possible so that I can continue on with the
>> project.
>>
>> I am interested in purchasing temporary computational resources for using
>> Gromacs.  Do any of you have experience running Gromacs "in the cloud" --
>> on
>> a service such as Amazon EC2 (http://aws.amazon.com/ec2/)?  For my quick
>> need for computational resources, some sort of cloud solution looks
>> interesting to me.  On the other hand, I could apply for time on XSEDE
>> (https://www.xsede.org/), but that would probably be a lengthier process,
>> and probably more suitable for a longterm solution.
>>
>> The problem with the cloud idea is that I am not an expert on computers and
>> networking, so I am worried that actually successfully using cluster time
>> might be difficult (whereas I have heard that, in contrast, Gromacs is
>> sometimes more or less pre-installed on university/national lab-based
>> supercomputer resources available through XSEDE grants).  Have you ever
>> used
>> Gromacs on a cloud service?
>>
>> Thank you for your time!
>>
>> Andrew DeYoung
>> Carnegie Mellon University
>>
>> --
>> gmx-users mailing listgmx-users@gromacs.org
>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>> * Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-users-requ...@gromacs.org.
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at 
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] problem with g_density -center

[Christopher Neale]

Sorry Ali, I can't do your work for you. If this is way over your head, then 
either forget it, find a colleague to help 
you, or spend some time learning bash and gromacs. I'n not commenting on your 
approach, but if you modified
your .gro then you need a new .top file. You can create one.

Chris.

-- original message --

I know that these commands are not similar to your code at all.

1-  trjconv -f md.xtc -o tmp.xtc -n cn.ndx

2- trjconv -center -pbc mol -f tmp.xtc -o
tmp2.xtc -s md.tpr -n cn.ndx tmp.xtc

3- trjconv -center -pbc atom -f tmp2.xtc -o
tmp3.xtc -s md.tpr -n cn.ndx tmp2.xtc

  4-  ## now make a new .tpr file in which the solute is at the
center of the box
  #first output a single frame

   trjconv -f  tmp3.xtc -dump 125000 -o
tmpgro.gro -s md.tpr -center -pbc mol -n cn.ndx

 5- #make a new .tpr file
 touch empty.mdp

 grompp -p mytop.top -c tmpgro.gro -f
empty.mdp -o centered.tpr -maxwarn 1

At this step give me an error :

 Fatal error:
Invalid line in tmpgro.gro for atom 1:
1LI  C11   0.8   1.9   0.7

I know this error related to my .top file. Because my top file did not
change while my .gro file changed.

2- I really did not find out these line in your commands. I'm not a
professional linux programmer.

 
GMXLIB=/project/pomes/cneale/GPC/exe/intel/gromacs-4.5.3_modtrjconv/exec/share/gromacs/top
echo "NE_CZ_NH1_NH2_CB_CG_CD System" |

and another lines 

   Please explain more if it is possible,
--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] problem with g_density -center

[Christopher Neale]

My point is that the old paper probably used the incorrect trjconv/g_density as 
well so to "reproduce" it you may
be best off using the standard gromacs.

Chris.

-- original message --

Thank you so much for your reply but it is very important for me that
i can reproduce

results of a paper or not.

I use new gromacs version but that paper has used old version(paper
has published: 2009)

I confused, i spent many times,

-- 
Sincerely

--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] A favor question: experience running Gromacs in the cloud

[Szilárd Páll]

Hi,

When it comes to hardware, pretty much the only thing that matters is
processors (CPUs and maybe GPUs) , memory will always be more than enough
-- unless you plan to run massive analysis in the cloud. In general, if
your simulation system runs fast enough on a single compute node (1-2
CPUs), the could could very well be worth it. When it comes to running over
a network, I'm not sure what the cloud-based options are, but I think it is
not not worth bothering with anything slower than 10G Ethernet or
Infiniband.

If you are not after a few long trajectories, but you're doing
some ensemble-simulations and need lots of independent trajectories, the
could will be quite well suited, especially if you can manage the
simulations in an elegant manner, e.g. see http://copernicus-computing.org.

I have no first hand concrete experience, but I know that some GROMACS
developers have been testing Amazon EC2 instances with pretty positive
results. You can get really low prices and good performance with the
"High-CPU Spot Instances".

Just to make it clear, I do not mean to advertise Amazon EC2, just that
I've only heard about first-hand experiences on this resource. There are
quite a few other cloud providers, so you might want compare and try before
buying!

Cheers,
--
Szilárd


On Fri, Oct 19, 2012 at 1:59 AM, Andrew DeYoung wrote:

> Hi, Gromacs users,
>
> If you have time, I am wondering if you have any advice, as a favor.
>
> I'm a graduate student in computational chemistry in the U.S.  My
> department
> has an EXCELLENT "in-house" cluster, on which I routinely run Gromacs jobs
> on 8 to 24 cores.  However, it occasionally happens that I run into a
> "crunch time" need for some more computational resources.  For example, I
> recently found a mistake (an incorrect parameter) that I made in an .itp
> file that I wrote.  This means that I need to rerun a whole slew of
> simulations as quickly as possible so that I can continue on with the
> project.
>
> I am interested in purchasing temporary computational resources for using
> Gromacs.  Do any of you have experience running Gromacs "in the cloud" --
> on
> a service such as Amazon EC2 (http://aws.amazon.com/ec2/)?  For my quick
> need for computational resources, some sort of cloud solution looks
> interesting to me.  On the other hand, I could apply for time on XSEDE
> (https://www.xsede.org/), but that would probably be a lengthier process,
> and probably more suitable for a longterm solution.
>
> The problem with the cloud idea is that I am not an expert on computers and
> networking, so I am worried that actually successfully using cluster time
> might be difficult (whereas I have heard that, in contrast, Gromacs is
> sometimes more or less pre-installed on university/national lab-based
> supercomputer resources available through XSEDE grants).  Have you ever
> used
> Gromacs on a cloud service?
>
> Thank you for your time!
>
> Andrew DeYoung
> Carnegie Mellon University
>
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] Re: Changing md command line in a script!

[Arman Mahboubi Soufiani]

Dear Vitaly,

I appreciate your informative response.

Dear Justin,

Thank you for your kind helps.

On Fri, Oct 19, 2012 at 5:31 PM, Dr. Vitaly Chaban wrote:

> > Dear all,
> >
> > A simple question.
> > I want to change the md command line in a script. I want to prevent my
> > script to lead the simulation on parallel nodes. By this I mean I want to
> > run it on one CPU. I do not know where I should embed the flag -nt=1 in
> that
> > command lien.
> > I wonder the following is the md command? If yes, how it will become when
> > adding -nt=1 to it?
>
> You need to remove the MPI program at the very beginning of the line,
> if you have any.
>
> You may also specify "-nt 1" as the mdrun option. My new workstation,
> for instance, tries to simulate with "-nt 12", even if I do not ask
> for this.
>
> All other utilities will be executed using 1 processor, regardless of
> what you request on the command line.
>
>
> Dr. Vitaly V. Chaban
> MEMPHYS - Center for Biomembrane Physics
> Department of Physics, Chemistry and Pharmacy
> University of Southern Denmark
> Campusvej 55, 5230 Odense M, Denmark
> --
> gmx-users mailing listgmx-users@gromacs.org
> http://lists.gromacs.org/mailman/listinfo/gmx-users
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> * Please don't post (un)subscribe requests to the list. Use the
> www interface or send it to gmx-users-requ...@gromacs.org.
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] problem with g_density -center

[Christopher Neale]

Dear Ali:

Please also keep in mind that 99% of the papers published with gromacs using 
NpT have shown density profiles
along Z that were computed simply by using (the incorrect) trjconv -center and 
(the incorrect for NpT) g_density. 
So while I think that it's always important to get things right, if you're new 
to gromacs and/or scripting then maybe 
you should just use trjconv and g_density straight out of the box -- you won't 
be the first or the last to do so and
if you are seeing results that are way different than what you find in the 
literature then there is some other 
problem. A good place to start is to get 100 ns of sampling and run standard 
g_density.

Chris.

-- original message --

Dear Chris

I found your topic:

chris.neale | 21 Aug 01:30   ,   problem with g_density -center

The following text is your email. Is it correct?

I pursue your steps but :

1- I don't understand first commands that related to directories and :

   empty.mdp ?  and  command : avg=($(cat coord.xvg|grep -v '[@|#]' |
awk '{x+=$2;y+=$3;z+=$4;n++}
END{print -1*x/n,-1*y/n}'))

2- I encounter with a problem in now make a new .tpr file, my error:

Fatal error:
Invalid line in tmpgro.gro for atom 1:
1LI  C11   0.8   1.9   0.7

my top file is the same first top file in my simulation,

3- You can help me by simpler commands?

--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] problem with g_density -center

[Christopher Neale]

Sorry, Ali, it's not really possible for me to simplify those commands. That's 
about as simple as it gets.

I suggest that you start a normal post that follows the form: I did this (paste 
commands) and I got this 
(paste error message). Also, when you get a message like 

Fatal error:
Invalid line in tmpgro.gro for atom 1:
1LI  C11   0.8   1.9   0.7

it becomes interesting to know what the result of "head tmpgro.gro" is.

* Note that you will need to also use a modified version of trjconv (see my 
post on trjconv -center centering by
max-min and not the actual center).

-- original message --

Dear Chris

I found your topic:

chris.neale | 21 Aug 01:30   ,   problem with g_density -center

The following text is your email. Is it correct?

I pursue your steps but :

1- I don't understand first commands that related to directories and :

   empty.mdp ?  and  command : avg=($(cat coord.xvg|grep -v '[@|#]' |
awk '{x+=$2;y+=$3;z+=$4;n++}
END{print -1*x/n,-1*y/n}'))

2- I encounter with a problem in now make a new .tpr file, my error:

Fatal error:
Invalid line in tmpgro.gro for atom 1:
1LI  C11   0.8   1.9   0.7

my top file is the same first top file in my simulation,

3- You can help me by simpler commands?
--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] Changing md command line in a script!

[Justin Lemkul]

On 10/19/12 7:16 AM, Arman M. Soufiani wrote:

Dear all,

A simple question.
I want to change the md command line in a script. I want to prevent my
script to lead the simulation on parallel nodes. By this I mean I want to
run it on one CPU. I do not know where I should embed the flag -nt=1 in that
command lien.
I wonder the following is the md command? If yes, how it will become when
adding -nt=1 to it?



Like any mdrun options, the argument can appear anywhere on the mdrun command 
line, i.e.:


mdrun -nt 1 -s md.tpr

-Justin

--


Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

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

[gmx-users] Error on mpi installation

[Rajiv Gandhi]

Dear Gromacs,

I am trying to install gromacs with mpi enabled in my cluster and i end
up with following error:

libtool: link: cannot find the library `../mdlib/libmd_mpi.la' or unhandled
argument `../mdlib/libmd_mpi.la'
make[1]: *** [libgmxpreprocess_mpi.la] Error

I can able to install gromacs in single precision without having trouble.
However, when i use mpi enabled by using lam-7.4.1 it gives above mentioned
error, I have tried openmpi as it installed in yum install manner but it
shows different error like not even compile it by giving this following
error. configure: error: cannot compute sizeof (off_t)



I have used the command line

./configure --program-suffix=_mpi --enable-mpi --enable-shared


I do not know how to solve this problem. Any suggestion would be
much appreciated.



Regards

Rajiv
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

[gmx-users] Changing md command line in a script!

[Arman M. Soufiani]

Dear all,

A simple question. 
I want to change the md command line in a script. I want to prevent my
script to lead the simulation on parallel nodes. By this I mean I want to
run it on one CPU. I do not know where I should embed the flag -nt=1 in that
command lien.
I wonder the following is the md command? If yes, how it will become when
adding -nt=1 to it?

Thank you in advance.

Arman



--
View this message in context: 
http://gromacs.5086.n6.nabble.com/Changing-md-command-line-in-a-script-tp5002237.html
Sent from the GROMACS Users Forum mailing list archive at Nabble.com.
-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

Re: [gmx-users] remd jobs failed

[Justin Lemkul]

On 10/19/12 2:37 AM, Albert wrote:

hello:

I am trying to submit replica exchange jobs to cluster by following command, but
failed:

g_tune_pme_d -x -np 128 mdrun -s remd_.tpr -multi 16 -replex 1000 -reseed -1
-launch

Here is the log file:

---
Program g_tune_pme_d, VERSION 4.5.5
Source code file: gmx_tune_pme.c, line: 1579

Fatal error:
File remd_.tpr not found.
For more information and tips for troubleshooting, please check the GROMACS
website at http://www.gromacs.org/Documentation/Errors
---


I've got all necessary .tpr file in the same directory (remd_1.tpr, remd_2.tpr
..)



You should be executing mdrun instead of g_tune_pme.

-Justin

--


Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

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

Re: [gmx-users] Dipole moment

[Chandan Choudhury]

Dear Rajiv,

Please start a different thread, if you want to get help. I suppose
you need to be informative regarding your error(s).

Chandan

--
Chandan kumar Choudhury
NCL, Pune
INDIA


On Fri, Oct 19, 2012 at 2:07 PM, Rajiv Gandhi  wrote:
> Dear Gromacs,
>
> I am trying to install gromacs with mpi enabled in my cluster and i end
> up with following error:
>
> libtool: link: cannot find the library `../mdlib/libmd_mpi.la' or unhandled
> argument `../mdlib/libmd_mpi.la'
> make[1]: *** [libgmxpreprocess_mpi.la] Error
>
> I can able to install gromacs in single precision without having trouble.
> However, when i use mpi enabled by using lam-7.4.1 it gives above mentioned
> error, I have tried openmpi also but its not even compile it and ends up
> with error like
>
> configure: error: cannot compute sizeof (off_t)
>
> I have used the command line ./configure --program-suffix=_mpi --enable-mpi
> --enable-shared
>
>
> I do not know how to solve this problem. Any suggestion would be
> much appreciated.
>
>
> Regards
>
> Rajiv
>
>
> On Fri, Oct 19, 2012 at 4:17 PM, Javier Cerezo  wrote:
>
>> When calculated during the md run, the corresponding options in g_energy
>> for the dipole moment are MU-X, MU-Y and MU-Z.
>>
>> However, taking into account the previous mail by David van der Spoel,
>> it's possible that the dipole moment could not be calculated in your
>> system. Options in g_energy vary depending on the actual computation.
>>
>> If you want to calculate the dipole moment by your own, I think it's
>> easier to use the xtc file exported every 3fs (it'll be considerably
>> heavier than the edr file, however) and writing an specific code for your
>> purposes. In order to read the trajectories, you can use the xtc library (
>> http://www.gromacs.org/**Developer_Zone/Programming_**Guide/XTC_Library).
>> Anyway, if the code didn't compute it, there should be a reason.
>>
>> Javier
>>
>>
>> El 19/10/12 04:42, Nilesh Dhumal escribió:
>>
>>  Is it possible to calcualte the dipole moment at 3fs by modifing the
>>> source code?
>>>
>>> Nilesh
>>>
>>>
>>>  How can I save the total dipole moment extacted by .edr file?

 I could not find the option in g_energy.

 Nilesh


  Justin is right. I've slipped a bit, that was not the way to set
> nstenergy. But still, you can save the dipole directly in the edr file.
> Not from your existing trajectory, but in a new run. Say your dt=1fs,
> then you should set nstenergy=3 to get your dipole moment every 3fs
> which you can retrieve with g_energy.
>
> Javier
>
>
> El 18/10/12 20:43, Nilesh Dhumal escribió:
>
>> In short, I can not save dipole moment directly. I can calculate dipole
>> moment only using .trr file.
>>
>> Nilesh
>>
>>> On 10/18/12 1:45 PM, Javier Cerezo wrote:
>>>
 Hi

 The dipole is stored in the edr file, which output frequency is
 controlled by
 "nstenergy" option in the mdp parameter file. Set it appropriately
 during your
 run (nstenergy=0.003) but keep in mind that it should be a multiple
 of
 the time
 step.

  nstenergy cannot be set this way; it is an integer indicating the
>>> interval
>>> (in
>>> number of steps) for writing energy values.  If the values have been
>>> saved
>>> every
>>> 2 ps, there is no way to retrieve them every 3 fs; if more frequent
>>> information
>>> is needed, it needs to be saved more frequently.
>>>
>>> -Justin
>>>
>>>  El 18/10/12 18:09, Nilesh Dhumal escribió:

> Hello,
>
> I am calculating the dipole moment auto-correlation function for my
> system
> which have 128 cation and 128 anion.
>
> I am saving the trajectory at each 2 ps and using this trajectory
> for
> further analysis.
>
> Can I save the dipole moment and three vectors at each 3 fs?
>
>
>
> Thanks
>
> Nilesh
>
>
>
>
>  --
>>> ==**==
>>>
>>> Justin A. Lemkul, Ph.D.
>>> Research Scientist
>>> Department of Biochemistry
>>> Virginia Tech
>>> Blacksburg, VA
>>> jalemkul[at]vt.edu | (540) 231-9080
>>> http://www.bevanlab.biochem.**vt.edu/Pages/Personal/justin
>>>
>>> ==**==
>>> --
>>> gmx-users mailing listgmx-users@gromacs.org
>>> http://lists.gromacs.org/**mailman/listinfo/gmx-users
>>> * Please search the archive at
>>> http://www.gromacs.org/**Support/Mailing_Lists/Searchbefore

Re: [gmx-users] Dipole moment

[Rajiv Gandhi]

Dear Gromacs,

I am trying to install gromacs with mpi enabled in my cluster and i end
up with following error:

libtool: link: cannot find the library `../mdlib/libmd_mpi.la' or unhandled
argument `../mdlib/libmd_mpi.la'
make[1]: *** [libgmxpreprocess_mpi.la] Error

I can able to install gromacs in single precision without having trouble.
However, when i use mpi enabled by using lam-7.4.1 it gives above mentioned
error, I have tried openmpi also but its not even compile it and ends up
with error like

configure: error: cannot compute sizeof (off_t)

I have used the command line ./configure --program-suffix=_mpi --enable-mpi
--enable-shared


I do not know how to solve this problem. Any suggestion would be
much appreciated.


Regards

Rajiv


On Fri, Oct 19, 2012 at 4:17 PM, Javier Cerezo  wrote:

> When calculated during the md run, the corresponding options in g_energy
> for the dipole moment are MU-X, MU-Y and MU-Z.
>
> However, taking into account the previous mail by David van der Spoel,
> it's possible that the dipole moment could not be calculated in your
> system. Options in g_energy vary depending on the actual computation.
>
> If you want to calculate the dipole moment by your own, I think it's
> easier to use the xtc file exported every 3fs (it'll be considerably
> heavier than the edr file, however) and writing an specific code for your
> purposes. In order to read the trajectories, you can use the xtc library (
> http://www.gromacs.org/**Developer_Zone/Programming_**Guide/XTC_Library).
> Anyway, if the code didn't compute it, there should be a reason.
>
> Javier
>
>
> El 19/10/12 04:42, Nilesh Dhumal escribió:
>
>  Is it possible to calcualte the dipole moment at 3fs by modifing the
>> source code?
>>
>> Nilesh
>>
>>
>>  How can I save the total dipole moment extacted by .edr file?
>>>
>>> I could not find the option in g_energy.
>>>
>>> Nilesh
>>>
>>>
>>>  Justin is right. I've slipped a bit, that was not the way to set
 nstenergy. But still, you can save the dipole directly in the edr file.
 Not from your existing trajectory, but in a new run. Say your dt=1fs,
 then you should set nstenergy=3 to get your dipole moment every 3fs
 which you can retrieve with g_energy.

 Javier


 El 18/10/12 20:43, Nilesh Dhumal escribió:

> In short, I can not save dipole moment directly. I can calculate dipole
> moment only using .trr file.
>
> Nilesh
>
>> On 10/18/12 1:45 PM, Javier Cerezo wrote:
>>
>>> Hi
>>>
>>> The dipole is stored in the edr file, which output frequency is
>>> controlled by
>>> "nstenergy" option in the mdp parameter file. Set it appropriately
>>> during your
>>> run (nstenergy=0.003) but keep in mind that it should be a multiple
>>> of
>>> the time
>>> step.
>>>
>>>  nstenergy cannot be set this way; it is an integer indicating the
>> interval
>> (in
>> number of steps) for writing energy values.  If the values have been
>> saved
>> every
>> 2 ps, there is no way to retrieve them every 3 fs; if more frequent
>> information
>> is needed, it needs to be saved more frequently.
>>
>> -Justin
>>
>>  El 18/10/12 18:09, Nilesh Dhumal escribió:
>>>
 Hello,

 I am calculating the dipole moment auto-correlation function for my
 system
 which have 128 cation and 128 anion.

 I am saving the trajectory at each 2 ps and using this trajectory
 for
 further analysis.

 Can I save the dipole moment and three vectors at each 3 fs?



 Thanks

 Nilesh




  --
>> ==**==
>>
>> Justin A. Lemkul, Ph.D.
>> Research Scientist
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.**vt.edu/Pages/Personal/justin
>>
>> ==**==
>> --
>> gmx-users mailing listgmx-users@gromacs.org
>> http://lists.gromacs.org/**mailman/listinfo/gmx-users
>> * Please search the archive at
>> http://www.gromacs.org/**Support/Mailing_Lists/Searchbefore
>>  posting!
>> * Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-users-requ...@gromacs.org.
>> * Can't post? Read 
>> http://www.gromacs.org/**Support/Mailing_Lists
>>
>>
>>  --
 Javier CEREZO BASTIDA
 Ph.D. Student
 Physical Chemistry
 Universidad de Murcia
 

Re: [gmx-users] Dipole moment

[Javier Cerezo]

When calculated during the md run, the corresponding options in g_energy 
for the dipole moment are MU-X, MU-Y and MU-Z.


However, taking into account the previous mail by David van der Spoel, 
it's possible that the dipole moment could not be calculated in your 
system. Options in g_energy vary depending on the actual computation.


If you want to calculate the dipole moment by your own, I think it's 
easier to use the xtc file exported every 3fs (it'll be considerably 
heavier than the edr file, however) and writing an specific code for 
your purposes. In order to read the trajectories, you can use the xtc 
library 
(http://www.gromacs.org/Developer_Zone/Programming_Guide/XTC_Library). 
Anyway, if the code didn't compute it, there should be a reason.


Javier


El 19/10/12 04:42, Nilesh Dhumal escribió:

Is it possible to calcualte the dipole moment at 3fs by modifing the
source code?

Nilesh



How can I save the total dipole moment extacted by .edr file?

I could not find the option in g_energy.

Nilesh



Justin is right. I've slipped a bit, that was not the way to set
nstenergy. But still, you can save the dipole directly in the edr file.
Not from your existing trajectory, but in a new run. Say your dt=1fs,
then you should set nstenergy=3 to get your dipole moment every 3fs
which you can retrieve with g_energy.

Javier


El 18/10/12 20:43, Nilesh Dhumal escribió:

In short, I can not save dipole moment directly. I can calculate dipole
moment only using .trr file.

Nilesh

On 10/18/12 1:45 PM, Javier Cerezo wrote:

Hi

The dipole is stored in the edr file, which output frequency is
controlled by
"nstenergy" option in the mdp parameter file. Set it appropriately
during your
run (nstenergy=0.003) but keep in mind that it should be a multiple
of
the time
step.


nstenergy cannot be set this way; it is an integer indicating the
interval
(in
number of steps) for writing energy values.  If the values have been
saved
every
2 ps, there is no way to retrieve them every 3 fs; if more frequent
information
is needed, it needs to be saved more frequently.

-Justin


El 18/10/12 18:09, Nilesh Dhumal escribió:

Hello,

I am calculating the dipole moment auto-correlation function for my
system
which have 128 cation and 128 anion.

I am saving the trajectory at each 2 ps and using this trajectory
for
further analysis.

Can I save the dipole moment and three vectors at each 3 fs?



Thanks

Nilesh





--


Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin


--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists



--
Javier CEREZO BASTIDA
Ph.D. Student
Physical Chemistry
Universidad de Murcia
30100, Murcia (SPAIN)
T: (0034)868887434

--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists




--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists






--
Javier CEREZO BASTIDA
PhD Student
Physical Chemistry
Universidad de Murcia
Murcia (Spain)
Tel: (+34)868887434
--
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.

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

Re: [gmx-users] remd jobs failed

[Davide Mercadante]

Hi Albert, 

Please accept my apologies, I must have misread your e-mail. Effectively
everything seems to be right in the command line and I am not able to
understand either why you get the error.

Good luck.

Cheers,
Davide

On 19/10/12 7:54 PM, "Albert"  wrote:

>hello David:
>
>thanks for kind reply.
>The .tpr file was created by grompp in cluster and there is no problem
>for that.
>
>thank you very much
>Albert
>
>On 10/19/2012 08:52 AM, Davide Mercadante wrote:
>> Hello,
>>
>> Basically is telling you that the output (.tpr) file that grompp should
>> have created is not there to be read.
>>
>> Check if grompp ran correctly and produced the wanted output. I suspect
>> that it may have failed for some reasons.
>>
>> Hope this helps,
>>
>> Cheers,
>> Davide
>>
>> On 19/10/12 7:37 PM, "Albert"  wrote:
>>
>>> hello:
>>>
>>> I am trying to submit replica exchange jobs to cluster by following
>>> command, but failed:
>>>
>>> g_tune_pme_d -x -np 128 mdrun -s remd_.tpr -multi 16 -replex 1000
>>> -reseed -1 -launch
>>>
>>> Here is the log file:
>>>
>>> ---
>>> Program g_tune_pme_d, VERSION 4.5.5
>>> Source code file: gmx_tune_pme.c, line: 1579
>>>
>>> Fatal error:
>>> File remd_.tpr not found.
>>> For more information and tips for troubleshooting, please check the
>>> GROMACS
>>> website at http://www.gromacs.org/Documentation/Errors
>>> ---
>>>
>>>
>>> I've got all necessary .tpr file in the same directory (remd_1.tpr,
>>> remd_2.tpr ..)
>>>
>>> thank you very much
>>> Albert
>>> -- 
>>> gmx-users mailing listgmx-users@gromacs.org
>>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>>> * Please search the archive at
>>> http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>>> * Please don't post (un)subscribe requests to the list. Use the
>>> www interface or send it to gmx-users-requ...@gromacs.org.
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>
>-- 
>gmx-users mailing listgmx-users@gromacs.org
>http://lists.gromacs.org/mailman/listinfo/gmx-users
>* Please search the archive at
>http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
>* Please don't post (un)subscribe requests to the list. Use the
>www interface or send it to gmx-users-requ...@gromacs.org.
>* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists


-- 
gmx-users mailing listgmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
* Please search the archive at 
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
* Please don't post (un)subscribe requests to the list. Use the 
www interface or send it to gmx-users-requ...@gromacs.org.
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists