<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
On 9/01/2012 10:35 PM, Guido Leoni wrote:
<blockquote
cite="mid:CAMXQpORuWM+_gnbf9T8KDw4A0GFpMSv6ExbpWZ5oM1z-pDq8hw@mail.gmail.com"
type="cite">
<div>Dear List </div>
<div>I'm newbie to MD .</div>
<div>I'd like to simulate a protein protein binding dinamic
between 2 proteins of 500 aa each one. I have some very basic
questions:</div>
<div>1) In order to setup my dinamic i'm following the basic
tutorial on spider toxin peptide (<a moz-do-not-send="true"
href="http://www.gromacs.org/Documentation/Tutorials#Protein-Ligand_Systems">http://www.gromacs.org/Documentation/Tutorials#Protein-Ligand_Systems</a>)
starting from a pdb with the 2 structures near each other and
oriented according to some docking calculations. Is this
procedure a reasonable starting point? <br>
</div>
</blockquote>
<br>
I guess so. It tests only that orientation, of course. Using the
pull code to measure the free energy of (un)binding might be
fruitful, but do read the literature here.<br>
<br>
<blockquote
cite="mid:CAMXQpORuWM+_gnbf9T8KDw4A0GFpMSv6ExbpWZ5oM1z-pDq8hw@mail.gmail.com"
type="cite">
<div><br>
</div>
<div>2) I tryed to run the MD with the same parameters of
tutorial in my md.mdp file (listed at the end of the mail).
Should i set the contraints parameter to "none"? In order to
allow more flexibility of the side chains?</div>
<div>Finally if the previous points are reasonable ...</div>
</blockquote>
<br>
Constraints pertain to individual bonds and angles. Not using them
requires a larger integration time step. While it's conceivable that
some extra flexibility at bond-and-angle level in the bound
configuration might affect the binding energy a little, the error in
that quantity is likely to be limited by the fixed-point-charge
approximation. A suitable point charge for a near-surface atom when
bound would differ from that when unbound. So you may as well have
constraints and virtual sites and have a large time step.<br>
<br>
<blockquote
cite="mid:CAMXQpORuWM+_gnbf9T8KDw4A0GFpMSv6ExbpWZ5oM1z-pDq8hw@mail.gmail.com"
type="cite">
<div>3) In your opinion which could be a reasonable time to
explore the possibility that the 2 proteins bind each other? I'm
making some attempts using 500 ps for now <br>
</div>
</blockquote>
<br>
Off the cuff, I'd think that at least an order of magnitude too low,
but you should check out the existing literature for similar work.<br>
<br>
<blockquote
cite="mid:CAMXQpORuWM+_gnbf9T8KDw4A0GFpMSv6ExbpWZ5oM1z-pDq8hw@mail.gmail.com"
type="cite">
<div><br>
</div>
<div>Sorry for my silly questions and thank you very much for any
kind of help </div>
<div>Best </div>
<div>Guido</div>
<div>
<div>
<div>define = -DPOSRES</div>
</div>
</div>
</blockquote>
<br>
Position restraints will prevent relative COM motion.<br>
<br>
Mark<br>
<br>
<blockquote
cite="mid:CAMXQpORuWM+_gnbf9T8KDw4A0GFpMSv6ExbpWZ5oM1z-pDq8hw@mail.gmail.com"
type="cite">
<div>
<div>
<div>constraints = all-bonds</div>
<div>integrator = md</div>
<div>dt = 0.002 ; ps !</div>
<div>nsteps = 50000 ; total 100.0 ps.</div>
<div>nstcomm = 10</div>
<div>nstxout = 500 ; collect data every 1.0 ps</div>
<div>nstxtcout = 500</div>
<div>nstvout = 5000</div>
<div>nstfout = 0</div>
<div>nstlog = 10</div>
<div>nstenergy = 50</div>
<div>nstlist = 10</div>
<div>
ns_type = grid</div>
<div>rlist = 1.0</div>
<div>coulombtype = PME</div>
<div>rcoulomb = 1.0</div>
<div>vdwtype = cut-off</div>
<div>rvdw = 1.4</div>
<div>
pme_order = 4 </div>
<div>ewald_rtol = 1e-5</div>
<div>optimize_fft = yes</div>
<div>DispCorr = no</div>
<div>; Berendsen temperature coupling is on</div>
<div>Tcoupl = v-rescale</div>
<div>tau_t = 0.1 0.1</div>
<div>
tc-grps = protein non-protein</div>
<div>ref_t = 300 300 </div>
<div>; Pressure coupling is off</div>
<div>Pcoupl = no</div>
<div>Pcoupltype = isotropic</div>
<div>tau_p = 1.0</div>
<div>compressibility = 4.5e-5</div>
</div>
<div><br>
</div>
<div><br>
</div>
-- <br>
Guido Leoni<br>
National Research Institute on Food and Nutrition<br>
(I.N.R.A.N.)<br>
via Ardeatina 546<br>
00178 Rome<br>
Italy<br>
<br>
tel + 39 06 51 49 41 (operator)<br>
+ 39 06 51 49 4498 (direct)<br>
</div>
<br>
<fieldset class="mimeAttachmentHeader"></fieldset>
<br>
</blockquote>
<br>
</body>
</html>