<div>Hello Chris, thanks for the reply. Below his message commented walkthrough. </div><div>In your email you asked me for quantitative results. All I have so far is </div><div>a visual inspection of the pulling trajectory. All I can see are </div>
<div>the two molecules moving linearly away from each other in a rectangular box </div><div>of dimensions 3x3x6nm. However this separation does not occur on the z axis, </div><div>as I wanted, but in a diagonal way across the faces of the box.</div>
<div><br></div><div>Bests</div><div>eef</div><div><br></div><div>> 1. There is no need to use a virtual atom, the pull code will use the</div><div>> center of mass. I strongly suggest that you stop using a virtual atom</div>
<div>> and just use the entire benzene as an argument to the pull code group.</div><div>> I have had difficulties with slightly more complicated setups of this</div><div>> type.</div><div><br></div><div>Ok, now I'm using only the center of mass of the molecules, no virtual sites.</div>
<div><br></div><div>> 2. In test2.jpg, the system without pbc shows a flat PMF after the</div><div>> cut-off -- exactly what one would expect. The pbc system shows</div><div>> continued interaction -- again what I would expect. So there is</div>
<div>> nothing actually all that strange here. One would not expect to see</div><div>> such a drastic difference in a high dielectric such as water, but in</div><div>> vacuum I suspect that this is expected.</div>
<div><br></div><div>Alright! In fact the well of both PMFs are not so different.</div><div><br></div><div>> 3. Please clarify what your cutoff was. I don't see a cut-off listed</div><div>> in your .mdp options and leaving this to the default of 1.0 nm is a</div>
<div>> bad idea because it can lead to confusion a times like this. I might</div><div>> assume that it was 0.7 nm based on test2.jpg, but then see the point</div><div>> #3 below.</div><div><br></div><div>Apesar de eu nao ter incluido o valor do cutoff na mensagem anterior </div>
<div>eu usei 1.2 nm</div><div>Although I have not included the cutoff value in previous message</div><div>I used it as 1.2 nm.</div><div><br></div><div>> 4a. I have no idea what -DPOSRES is actually doing for you since I</div>
<div>> can't see your topology.</div><div><br></div><div>A ideia de usar POSRES se referia ao emprego dos sitios virtuais.</div><div>Assim eu restringia a distancia das moleculas atraves da restricao do</div><div>sitios virtuais sobre seus centros.</div>
<div>The idea of using POSRES is referred to the employment of the virtual sites.</div><div>So I restricted the distance of molecules by restricting virtual sites </div><div>on their centers.</div><div><br></div><div>></div>
<div>> 4b. Are you sure that "pull_dim = N N Y" is really what you want?</div><div>> Sometimes one wants to average over X and Y, but I am not sure that</div><div>> you do in this case.</div><div><br></div>
<div>I want to separate the two molecules from one another so that the</div><div>displacement is restricted to the z axis. I can be wrong,</div><div>but I think to do it in water, using PBC, I should use the </div><div>components distance (N N Y ) with direction geometry to get a </div>
<div>pulling in a direction z! Please correct me if it is wrong.</div><div><br></div><div>> 4c. What exactly do you believe pull_r0 and pull_r1 are doing for you?</div><div><br></div><div>Here they are worthless, because I have not used the option "cylinder". </div>
<div>I should have removed these options from the original message.</div><div><br></div><div>> ### More minor notes:</div><div>></div><div>> 5. regarding test1.jpg: a PMF is correct to an additive constant,</div>
<div>> meaning that you can shift two PMFs relative to one another. These 2</div><div>> PMFs are therefore less different than they appear in your compaison</div><div>> plot, but they do differ in the slope between 1.0 - 2.0 nm. This is</div>
<div>> probably just a convergence issue and you will always need to do tests</div><div>> like this.</div><div><br></div><div>Surely.</div><div><br></div><div>> 6. regarding histo.png: can you confirm that the few very short</div>
<div>> gaussians are due to less sampling in a few windows? In any event, the</div><div>> overlap looks good.</div><div><br></div><div>Yes, there were two histograms whose simulations had crashed for some reason.</div>
<div>It was corrected.</div>_______________________________________<br>Eudes Eterno Fileti<br>Centro de Ciências Naturais e Humanas<br>Universidade Federal do ABC — CCNH<br>Av. dos Estados, 5001<br>Santo André - SP - Brasil<br>
CEP 09210-971<br>+55.11.4996-0196<br><a href="http://fileti.ufabc.edu.br">http://fileti.ufabc.edu.br</a><br>
<br><br><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
4. PMF in vacuum and pull direction (<a href="mailto:chris.neale@utoronto.ca">chris.neale@utoronto.ca</a>)<br>
<br><br>
<br>
Message: 4<br>
Date: Tue, 08 Jun 2010 12:14:34 -0400<br>
From: <a href="mailto:chris.neale@utoronto.ca">chris.neale@utoronto.ca</a><br>
Subject: [gmx-users] PMF in vacuum and pull direction<br>
To: <a href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>
Message-ID: <<a href="mailto:20100608121434.u6rhqhz400gc8cg8@webmail.utoronto.ca">20100608121434.u6rhqhz400gc8cg8@webmail.utoronto.ca</a>><br>
Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes";<br>
format="flowed"<br>
<br>
Dear Eudes:<br>
<br>
This is better, but I'm afraid that I'm going to have to insist on<br>
getting everything I asked for. Please see my previous message.<br>
<br>
Chris.<br>
<br>
-- original message --<br>
<br>
Dear Chris, I'm so sorry for surperficial email.<br>
You are correct because in the previous message<br>
I forgot to mention that now the simulation was<br>
performed in water (naturally with PBC).<br>
The main parameters of the simulation are given below.<br>
As I said, I wish to keep the displacement of the pulled molecule at<br>
the z-axis, but I'm not getting with this protocol.<br>
<br>
For the vacuum I followed the tips you gave me earlier and had success!<br>
Thanks<br>
eef<br>
<br>
<br>
<br>
title = Umbrella pulling simulation<br>
integrator = sd<br>
dt = 0.002<br>
nsteps = 3000000<br>
constraint_algorithm = lincs<br>
constraints = all-bonds<br>
nstlist = 5<br>
ns_type = simple<br>
rlist = 1.2<br>
rcoulomb = 1.2<br>
rvdw = 1.2<br>
coulombtype = Cut-off<br>
Tcoupl = V-rescale<br>
tc_grps = system<br>
tau_t = 0.1<br>
ref_t = 298<br>
Pcoupl = no<br>
pcoupltype = isotropic<br>
tau_p = 1.0<br>
compressibility = 4.5e-5<br>
ref_p = 1.0<br>
pbc = xyz<br>
<br>
; Pull code<br>
pull = umbrella<br>
pull_geometry = direction<br>
pull_dim = N N Y<br>
pull_start = yes<br>
pull_ngroups = 1<br>
pull_group0 = BEN1<br>
pull_group1 = BEN2<br>
pull_vec1 = 0 0 1<br>
pull_rate1 = 0.001<br>
pull_k1 = 1600<br>
<br>
<br>
<br>
> Message: 4<br>
> Date: Tue, 08 Jun 2010 08:05:23 -0400<br>
> From: chris.neale at <a href="http://utoronto.ca" target="_blank">utoronto.ca</a><br>
> Subject: [gmx-users] PMF in vacuum and pull direction<br>
> To: gmx-users at <a href="http://gromacs.org" target="_blank">gromacs.org</a><br>
> Message-ID: <20100608080523.i6oo62he1wwc0wkg at <a href="http://webmail.utoronto.ca" target="_blank">webmail.utoronto.ca</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes";<br>
> format="flowed"<br>
><br>
> Dear Eudes:<br>
><br>
> You can make my job a whole lot easier! First, please go back through<br>
> all of the comments that I gave you last time and reply to them one by<br>
> one. Did you do them? What did you see? Second, please include your<br>
> new .mdp and some quantitative results to better explain what you see<br>
> (e.g. your .px file and a description of why the values are not what<br>
> you expect).<br>
><br>
> Sorry if this sounds annoying, but there's no point in my working in the<br>
> dark.<br>
><br>
> Chris.<br>
><br>
> -- original message --<br>
><br>
> Hello Chris, thanks for the tips, they were very helpful.<br>
> Now a new problem appeared. I'm trying to separate the<br>
> two benzene molecules from one another while maintaining<br>
> the displacement vector aligned with the z axis. For that I<br>
> have used "pull_geometry = direction". However, at the end of the<br>
> simulation, I notice that the desired alignment (with z) was<br>
> not maintained. The molecule is pulled away from z and eventually<br>
> crosses the PBC. I took a good look at gmx-list and noticed<br>
> that you rode conducting tests with this option. Could you give<br>
> me a light on this problem too? Grateful<br>
> eef<br>
> _______________________________________<br>
> Eudes Eterno Fileti<br>
> Centro de Ciências Naturais e Humanas<br>
> Universidade Federal do ABC ? CCNH<br>
> Av. dos Estados, 5001<br>
> Santo André - SP - Brasil<br>
> CEP 09210-971<br>
> +55.11.4996-0196<br>
> <a href="http://fileti.ufabc.edu.br" target="_blank">http://fileti.ufabc.edu.br</a><br>
><br>
><br>
> Message: 2<br>
> Date: Thu, 27 May 2010 12:12:03 -0400<br>
> From: chris.neale at <a href="http://utoronto.ca" target="_blank">utoronto.ca</a><br>
> Subject: [gmx-users] PMF in vacuum<br>
> To: gmx-users at <a href="http://gromacs.org" target="_blank">gromacs.org</a><br>
> Message-ID: <20100527121203.orb2sq228s4ow0wg at <a href="http://webmail.utoronto.ca" target="_blank">webmail.utoronto.ca</a>><br>
> Content-Type: text/plain; charset=ISO-8859-1; DelSp="Yes";<br>
> format="flowed"<br>
><br>
> Dear Eudes:<br>
> To answer your pbc vs no-pbc question, I suggest that you use pbc=no<br>
> and set nstlist=0 rlist=0 rvdw=0 rcoulomb=0 so that you calculate all<br>
> interactions in direct space with no cutoffs.<br>
><br>
> ## Major comments that you should still investigate<br>
><br>
> 1. There is no need to use a virtual atom, the pull code will use the<br>
> center of mass. I strongly suggest that you stop using a virtual atom<br>
> and just use the entire benzene as an argument to the pull code group.<br>
> I have had difficulties with slightly more complicated setups of this<br>
> type.<br>
><br>
> 2. In test2.jpg, the system without pbc shows a flat PMF after the<br>
> cut-off -- exactly what one would expect. The pbc system shows<br>
> continued interaction -- again what I would expect. So there is<br>
> nothing actually all that strange here. One would not expect to see<br>
> such a drastic difference in a high dielectric such as water, but in<br>
> vacuum I suspect that this is expected.<br>
><br>
> 3. Please clarify what your cutoff was. I don't see a cut-off listed<br>
> in your .mdp options and leaving this to the default of 1.0 nm is a<br>
> bad idea because it can lead to confusion a times like this. I might<br>
> assume that it was 0.7 nm based on test2.jpg, but then see the point<br>
> #3 below.<br>
><br>
> 4a. I have no idea what -DPOSRES is actually doing for you since I<br>
> can't see your topology.<br>
><br>
> 4b. Are you sure that "pull_dim = N N Y" is really what you want?<br>
> Sometimes one wants to average over X and Y, but I am not sure that<br>
> you do in this case.<br>
><br>
> 4c. What exactly do you believe pull_r0 and pull_r1 are doing for you?<br>
><br>
> ### More minor notes:<br>
><br>
> 5. regarding test1.jpg: a PMF is correct to an additive constant,<br>
> meaning that you can shift two PMFs relative to one another. These 2<br>
> PMFs are therefore less different than they appear in your compaison<br>
> plot, but they do differ in the slope between 1.0 - 2.0 nm. This is<br>
> probably just a convergence issue and you will always need to do tests<br>
> like this.<br>
><br>
> 6. regarding histo.png: can you confirm that the few very short<br>
> gaussians are due to less sampling in a few windows? In any event, the<br>
> overlap looks good.<br>
><br>
> Chris<br>
><br>
> -- original message --<br>
><br>
> Hello Chris, thanks for your attention.<br>
> I'm sending you some links to some tests<br>
> I performed. As I said you will notice that<br>
> depending on the parameter used my simulation<br>
> shows PMF profiles quite different. Especially what<br>
> concerns to the difference between the use or not of the PBC.<br>
><br>
> <a href="https://sites.google.com/site/fileti/files/test1.jpg" target="_blank">https://sites.google.com/site/fileti/files/test1.jpg</a><br>
> <a href="https://sites.google.com/site/fileti/files/test2.jpg" target="_blank">https://sites.google.com/site/fileti/files/test2.jpg</a><br>
> <a href="https://sites.google.com/site/fileti/files/histo.png" target="_blank">https://sites.google.com/site/fileti/files/histo.png</a><br>
><br>
> I have constructed two very similar topologies (ben-a.itp and ben-b.itp)<br>
> where I put a virtual site in the center of benzene.<br>
> This sites were restrained to keep my molecules<br>
> fixed distance desired.<br>
><br>
> The basic details of the simulations are given below:1000<br>
><br>
> define = -DPOSRES<br>
> integrator = sd<br>
> tinit = 0<br>
> dt = 0.002<br>
> nsteps = 5000000 or 500000<br>
> comm-mode = angular<br>
> nstcomm = 1<br>
> comm-grps = System<br>
> bd-fric = 1<br>
> ld-seed = 1993<br>
> nstlist = 5<br>
> ns_type = simple<br>
> pbc = no or xyz<br>
> periodic_molecules = no<br>
> rlist = 0<br>
> coulombtype = Cut-off<br>
> vdw-type = Cut-off<br>
> rvdw = 0<br>
> DispCorr = no<br>
> Tcoupl = Nose-Hoover<br>
> tc_grps = system<br>
> tau_t = 0.1<br>
> ref_t = 300<br>
> Pcoupl = no<br>
> pcoupltype = isotropic<br>
> tau_p = 1.0<br>
> compressibility = 4.5e-5<br>
> ref_p = 1.0<br>
> constraints = all-bonds<br>
> constraint_algorithm = lincs<br>
><br>
> ; COM PULLING<br>
> pull = umbrella<br>
> pull_geometry = distance<br>
> pull_dim = N N Y<br>
> pull_r1 = 1<br>
> pull_r0 = 1.5<br>
> pull_constr_tol = 1e-06<br>
> pull_start = yes<br>
> pull_nstxout = 10<br>
> pull_nstfout = 10<br>
> pull_ngroups = 1<br>
> pull_group0 = BENa<br>
> pull_weights0 =<br>
> pull_pbcatom0 = 0<br>
> pull_group1 = BENb<br>
> pull_weights1 =<br>
> pull_pbcatom1 = 0<br>
> pull_vec1 = 0 0 1<br>
> pull_init1 = 0<br>
> pull_rate1 = 0.0<br>
> pull_k1 = 1700 or<br>
><br>
><br>
><br>
<br>
<br>
<br>
<br>
------------------------------<br>
<font color="#888888"><br>
--<br>
gmx-users mailing list<br>
<a href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>
<a href="http://lists.gromacs.org/mailman/listinfo/gmx-users" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-users</a><br>
Please search the archive at <a href="http://www.gromacs.org/search" target="_blank">http://www.gromacs.org/search</a> before posting!<br>
<br>
End of gmx-users Digest, Vol 74, Issue 53<br>
*****************************************<br>
</font></blockquote></div><br>