<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Tahoma
}
--></style>
</head>
<body class='hmmessage'><div dir='ltr'>
I have used a cubic box of dimensions 8 x 8 x 14 (nm), and my total pulling was 5nm along the z direction. I dontīt think there could be a problem with the PBC, since the layer of solvent around the protein is quite big, although I suppose that using a dodecahedron box for this system would have been better.<br>I will try now the "pull_dim = Y Y Y" and see what it happens.<br>Thanks a lot for the suggestions!<br>Best wishes,<br>Rebeca.<br><br><div>> Date: Fri, 22 Jul 2011 08:23:18 -0400<br>> From: chris.neale@utoronto.ca<br>> To: gmx-users@gromacs.org<br>> Subject: [gmx-users] large error bars in PMF<br>> <br>> I see now what you mean. As it happens, I doubt that this would have <br>> caused the problem since no force was applied on X and Y dimensions, <br>> so it would require that there was a PBC-based distance degeneracy <br>> along Z, although this is of course possible and hopefully Rebecca <br>> will answer this part.<br>> <br>> Also, thanks for the pull_dimension/pull_vec fix.<br>> <br>> Chris.<br>> <br>> -- original message --<br>> <br>> <br>> chris.neale@utoronto.ca wrote:<br>> <br>> [Hide Quoted Text]<br>> I don't see why the box-type makes any difference whatsoever. It is<br>> possible that if you use a rhombic dodecahedron, you may reduce the<br>> system size, thus simulate more ns/day, thus converge faster, but that<br>> should be the only effect. I would be interested to hear more from<br>> Justin about how the box-shape is expected to affect peptide rotation...<br>> perhaps I misunderstand this point.<br>> My point was not that the box shape has any effect on protein rotation. That<br>> will happen regardless of the box shape, of course. My suggestion for <br>> this box<br>> type was that since Rebeca has a system that is essentially spherically<br>> symmetric (i.e. two proteins connected by some arbitrary vector, which are at<br>> the same time rotating freely), then she must use a suitable box shape that<br>> reflects this type of symmetry. I never got a clear answer to whether or not<br>> the weird interactions she cited were due to PBC or not, but if one uses a<br>> rectangular box for a system like this one, there can be artificial <br>> interactions<br>> very easily.<br>> <br>> [Hide Quoted Text]<br>> I have a few other comments.<br>> <br>> 1. If you allow the peptide to rotate freely, then you do indeed need to<br>> converge all of their different rotational interactions. An alternative<br>> is to apply orientational restraints during the pulling (assuming that<br>> you know the bound state) and then to correct to an unrestrained state<br>> at large separations. You can see, for instance, D. L. Mobley, J. D.<br>> Chodera, K. A. Dill. "On the use of orientational restraints and<br>> symmetry number corrections in alchemical free energy calculations", ...<br>> <br>> 2. You are using "pull_dim = N N Y" which, if I haven't entirely<br>> forgotten how the pull-code works, means that the distance along Z is<br>> restrained but the distance along X and Y is free to change. What you<br>> end up with by sampling in this way is pretty strange and will require a<br>> really weird volumetric correction in the absence of infinite sampling<br>> time. You must decide to either: (i) pull to a spherical distance:<br>> <br>> pull_dim = Y Y Y<br>> pull_geometry = distance<br>> <br>> or (ii) to pull along a defined vector<br>> <br>> pull_dim = Y Y Y<br>> pull_geometry = direction<br>> Just a note here - if you set direction geometry, the pull_dim keyword is not<br>> used, but pull_vec is.<br>> <br>> -Justin<br>> <br>> [Hide Quoted Text]<br>> What you have done:<br>> <br>> pull_dim = N N Y<br>> pull_geometry = distance<br>> <br>> is only really useful when the system is isotropic along the XY plane<br>> (at least in the time averaged sense), such as for a lipid bilayer, or<br>> when the freedom in X and Y is very low, such as in a channel.<br>> <br>> 3. Finally, just because you sampled *more* doesn't mean that your<br>> values are converged. Look into block averaging and test to see if your<br>> binding free energy is drifting over time.<br>> <br>> Good luck,<br>> Chris.<br>> <br>> -- original message --<br>> <br>> Hi again,<br>> I have one doub about the suggestion of using a dodecahedral box for my<br>> umbrella sampling to remove the problems I am having with the peptides<br>> rotating. I cannot see why a dodecaheral box is going to avoid this.<br>> Would it be better a truncated octahedron?<br>> Thanks a lot for your help.<br>> Best wishes,<br>> Rebeca.<br>> <br>> <... snip...><br>> <br>> <br>> -- <br>> gmx-users mailing list gmx-users@gromacs.org<br>> http://lists.gromacs.org/mailman/listinfo/gmx-users<br>> Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!<br>> Please don't post (un)subscribe requests to the list. Use the <br>> www interface or send it to gmx-users-request@gromacs.org.<br>> Can't post? Read http://www.gromacs.org/Support/Mailing_Lists<br></div> </div></body>
</html>