<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#ffffff">
On 1/03/2011 4:58 PM, C.Y. Chang wrote:
<blockquote
cite="mid:AANLkTikQ3L72moP7O1YS5U=tnDp79A5tt_cW-3EuD4ei@mail.gmail.com"
type="cite">Hi,<br>
<br>
I have modified the dummy atom in the topology (attached files)
and pdb file.<br>
On the other side, I study the lipid system.<br>
Therefore, I constrained all bond lengths in my system.<br>
This is bond parameters in the mdp file.<br>
<br>
continuation = no ; starting up<br>
constraints = all-bonds ; constrain all bond lengths<br>
constraint-algorithm = lincs ; holonomic constraints<br>
lincs-iter = 1 ; accuracy of LINCS<br>
lincs-order = 4 ; also related to accuracy<br>
<br>
<br>
But the MDS process was performed and it showed the error msg.<br>
<br>
ERROR: Cannot have constraint (2-1) with virtual site (2)<br>
ERROR: Cannot have constraint (2-3) with virtual site (2)<br>
<br>
-------------------------------------------------------<br>
Program grompp, VERSION 4.0.5<br>
Source code file: vsite_parm.c, line: 856<br>
<br>
Fatal error:<br>
There were 2 virtual sites involved in constraints<br>
-------------------------------------------------------<br>
<br>
How should I solve the problem.<br>
</blockquote>
<br>
You made a virtual atom 2 three-quarters of the way from real atoms
1 and 3 (is an N-C bond really three times as long as an C#N bond?).
You defined bonds between 1 and 2, and between 2 and 3. Then you
told GROMACS to convert them into constraints. It didn't like that.<br>
<br>
So remove those bonds, and make a bond between 1 and 3 whose
equilibrium length is the sum of those of the old bonds. The force
constants won't matter, since you're holding the length fixed anyway
(but you may as well interpolate between the previous two force
constants). You will need to adjust your angle and dihedral
definitions accordingly - references to atom 2 now need to be to
atom 1.<br>
<br>
Mark<br>
<br>
<blockquote
cite="mid:AANLkTikQ3L72moP7O1YS5U=tnDp79A5tt_cW-3EuD4ei@mail.gmail.com"
type="cite">Thanks for your suggestions.<br>
<br>
Chia-yun Chang<br>
<br>
<br>
<div class="gmail_quote">2011/2/28 Mark Abraham <span dir="ltr"><<a
moz-do-not-send="true" href="mailto:mark.abraham@anu.edu.au">mark.abraham@anu.edu.au</a>></span><br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt
0.8ex; border-left: 1px solid rgb(204, 204, 204);
padding-left: 1ex;">
<div class="im"><br>
<br>
<span>On 02/28/11, <b>"C.Y. Chang" </b> <<a
moz-do-not-send="true"
href="mailto:chiayun.chang@gmail.com" target="_blank">chiayun.chang@gmail.com</a>>
wrote:</span>
<blockquote style="border-left: 1px solid rgb(0, 0, 255);
padding-left: 13px; margin-left: 0pt;" type="cite">
<div>Hi, <br>
<br>
I try to add a dummy atom in my small molecular pdb and
topology file. (attached files)<br>
But the MDS process still shut down.<br>
The msg. is<br>
[1]+ Exit 255 nohup mdrun -v -deffnm
npt_cmplx<br>
Could you give me some suggestions for dealing with the
problem.</div>
</blockquote>
<br>
</div>
Solving the problem requires removing the one of the three
existing co-linear atoms by treating it as a virtual atom. You
haven't done that. You've introduced a fourth co-linear atom -
the dummy. Treat the central C atom as a dummy. That reduces
your number of non-virtual co-linear atoms to 2 in all cases.
That solves this problem.<br>
<font color="#888888"><br>
Mark</font>
<div>
<div class="h5"><br>
<br>
<blockquote style="border-left: 1px solid rgb(0, 0, 255);
padding-left: 13px; margin-left: 0pt;" type="cite">
<div><br>
Thaks a lot.<br>
Best,<br>
<br>
<div> Chia-yun Chang<br>
</div>
<br>
<br>
<div class="gmail_quote">
2011/2/19 Mark Abraham <span dir="ltr"><<a
moz-do-not-send="true"
href="mailto:Mark.Abraham@anu.edu.au"
target="_blank">Mark.Abraham@anu.edu.au</a>></span><br>
<blockquote class="gmail_quote" style="margin: 0pt
0pt 0pt 0.8ex; border-left: 1px solid rgb(204,
204, 204); padding-left: 1ex;">
<div bgcolor="#ffffff" text="#000000">
<div> On 19/02/2011 5:18 PM, C.Y. Chang wrote: </div>
<blockquote type="cite">Hi,<br>
<br>
<div> The errpr msg. that I expressed in the
previous mail is unclear.<br>
It really happened in the grompp step of my
dealinw with the em.<br>
The manual shows that the viste can be used
by <br>
<br>
[ virtual sites2 ]<br>
</div>
</blockquote>
<br>
That's misspelled too. Underscores are
significant. You have to be literal when dealing
with computers.
<div><br>
<br>
<blockquote type="cite"> ; Site from funct a<br>
5 1 2 1 0.7439756<br>
<br>
What should I choose the vsite for the -CN
group? (2, 3fd or others?)<br>
</blockquote>
<br>
</div>
The problems come when the force from normal
angle and dihedral functions aren't numerically
stable when there are 3 collinear atoms. The way
to avoid this is to have one of those atoms
"virtual". Forces on it get projected onto the
real atoms from which it is defined, the
integration of forces proceeds only on the real
atoms, and the position of the virtual site gets
reconstructed later. So, which type do you think
will work best here?
<div><br>
<br>
<blockquote type="cite">I read the manual and
the columns of the vsite need some
parameters.<br>
The "Site" seems to be the atom number.<br>
</blockquote>
<br>
</div>
Yes.
<div><br>
<br>
<blockquote type="cite"> The "from" seems some
atom group, but which atoms should be
included?<br>
</blockquote>
<br>
</div>
The real atoms that form the virtual site. The
number and order depends on the type of virtual
site. Which atom is which is covered back in 4.7
You need to read that together with 5.2.2, like
I said several emails ago.
<div><br>
<br>
<blockquote type="cite">How could obtain the
"func" values?<br>
</blockquote>
<br>
</div>
That's situation-dependent. Here, you want to
pick a position along the line that will lead to
the three atoms being a sensible distance apart.
<div><br>
<br>
<blockquote type="cite">Could you give me a
molecule for my example?<br>
</blockquote>
<br>
</div>
No, I don't have one. Your example above will
construct atom 5 from 1 and 2 75% of the way
along the line from 1 to 2, per Figure 4.16 and
equation 4.134.<br>
<font color="#888888"> <br>
Mark</font>
<div>
<div><br>
</div>
<div><br>
<br>
<blockquote type="cite">Thanks for your
help.<br>
<br>
Best,<br>
<br>
Chia-yun<br>
<br>
<br>
<div class="gmail_quote">2011/2/19 Mark
Abraham <span dir="ltr"><<a
moz-do-not-send="true"
href="mailto:Mark.Abraham@anu.edu.au"
target="_blank">Mark.Abraham@anu.edu.au</a>></span><br>
<blockquote class="gmail_quote"
style="margin: 0pt 0pt 0pt 0.8ex;
border-left: 1px solid rgb(204, 204,
204); padding-left: 1ex;">
<div bgcolor="#ffffff" text="#000000">
<div> On 19/02/2011 3:02 PM, C.Y.
Chang wrote: </div>
<blockquote type="cite">Hi,<br>
<br>
<div> I have tried to add<br>
<br>
[ position restraints ]<br>
</div>
</blockquote>
<br>
This is misspelled. Surely grompp
warned about this?
<div><br>
<br>
<blockquote type="cite">2 1 1000
1000 1000 ; Restrain to a point<br>
1 1 1000 0 1000 ; Restrain to a
line (Y-axis)<br>
3 1 1000 0 0 ; Restrain to a
plane (Y-Z-plane)<br>
<br>
in the end of the topology file.<br>
</blockquote>
<br>
</div>
This does not even approach the
solution to your problem. Position
restraints inhibit diffusion and
structural changes. You need a
vsite, like you knew earlier.
<div><br>
<br>
<blockquote type="cite">In the em.
process, I get the eroor msg.<br>
</blockquote>
<br>
</div>
No, this error happened in grompp.
<div><br>
<br>
<blockquote type="cite">Fatal
error:<br>
Invalid dihedral type 1000<br>
</blockquote>
<br>
</div>
GROMACS didn't recognise the
mis-spelled directive, and so it's
trying to make sense of your
position restraint lines as
dihedrals.<br>
<font color="#888888"> <br>
Mark</font>
<div>
<div><br>
<br>
<blockquote type="cite">
Chia-yun<br>
<br>
<br>
<div class="gmail_quote">2011/2/18
Mark Abraham <span
dir="ltr"><<a
moz-do-not-send="true"
href="mailto:Mark.Abraham@anu.edu.au"
target="_blank">Mark.Abraham@anu.edu.au</a>></span><br>
<blockquote
class="gmail_quote"
style="margin: 0pt 0pt 0pt
0.8ex; border-left: 1px
solid rgb(204, 204, 204);
padding-left: 1ex;">
<div>On 18/02/2011 1:13
PM, C.Y. Chang wrote:<br>
<blockquote
class="gmail_quote"
style="margin: 0pt 0pt
0pt 0.8ex;
border-left: 1px solid
rgb(204, 204, 204);
padding-left: 1ex;">
Hi,<br>
<br>
I am dealing with the
lipid bilayer
permeation simulation.<br>
Most compounds can be
finished, but the
compounds with CN
can't be performed
simulation.<br>
I have searched the
discussion in the
gmx-users discussion.<br>
The "vsite" has been
mentioned, and I have
refered to the gromacs
manual.<br>
But I don't understand
that how I can use the
"vsite".<br>
</blockquote>
<br>
</div>
The theory is discussed in
chapter 4 and there's a
brief example in 5.2.2.
What have you tried and
what went wrong?<br>
<br>
Mark
<div>
<div><br>
<br>
<blockquote
class="gmail_quote"
style="margin: 0pt
0pt 0pt 0.8ex;
border-left: 1px
solid rgb(204, 204,
204); padding-left:
1ex;"> (add the
toplogy file or use
the command line?)<br>
I attach the PDB and
toplogy file of the
molecule.<br>
First, I performed
the molecular
dynamic simulation
for the pure lipid
bilayer, and the
step was been
finished.<br>
After I insert the
molecule, these
command lines are
performed.<br>
<br>
grompp -f minim.mdp
-c cmplx.pdb -p
topol_dppc.top -o
em.tpr<br>
mdrun -v -deffnm em<br>
grompp -f
npt_cmplx.mdp -c
em.gro -p
topol_dppc.top -o
npt_cmplx.tpr<br>
nohup mdrun -v
-deffnm npt_cmplx
&<br>
<br>
Thanks for your
help.<br>
Best,<br>
<br>
Chia-yun<br>
</blockquote>
<br>
</div>
</div>
<font color="#888888"> --
<br>
gmx-users mailing list
<a
moz-do-not-send="true"
href="mailto:gmx-users@gromacs.org" target="_blank">gmx-users@gromacs.org</a><br>
<a
moz-do-not-send="true"
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
moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists/Search"
target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a>
before posting!<br>
Please don't post
(un)subscribe requests
to the list. Use the www
interface or send it to
<a
moz-do-not-send="true"
href="mailto:gmx-users-request@gromacs.org" target="_blank">gmx-users-request@gromacs.org</a>.<br>
Can't post? Read <a
moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
</font></blockquote>
</div>
<br>
</blockquote>
<br>
</div>
</div>
</div>
<br>
--<br>
gmx-users mailing list <a
moz-do-not-send="true"
href="mailto:gmx-users@gromacs.org"
target="_blank">gmx-users@gromacs.org</a><br>
<a moz-do-not-send="true"
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
moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists/Search"
target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a>
before posting!<br>
Please don't post (un)subscribe
requests to the list. Use the<br>
www interface or send it to <a
moz-do-not-send="true"
href="mailto:gmx-users-request@gromacs.org"
target="_blank">gmx-users-request@gromacs.org</a>.<br>
Can't post? Read <a
moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists"
target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
</blockquote>
</div>
<br>
</blockquote>
<br>
</div>
</div>
</div>
<br>
--<br>
gmx-users mailing list <a
moz-do-not-send="true"
href="mailto:gmx-users@gromacs.org"
target="_blank">gmx-users@gromacs.org</a><br>
<a moz-do-not-send="true"
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
moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists/Search"
target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a>
before posting!<br>
Please don't post (un)subscribe requests to the
list. Use the<br>
www interface or send it to <a
moz-do-not-send="true"
href="mailto:gmx-users-request@gromacs.org"
target="_blank">gmx-users-request@gromacs.org</a>.<br>
Can't post? Read <a moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists"
target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
</blockquote>
</div>
<br>
</div>
</blockquote>
</div>
</div>
<br>
--<br>
gmx-users mailing list <a moz-do-not-send="true"
href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>
<a moz-do-not-send="true"
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 moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists/Search"
target="_blank">http://www.gromacs.org/Support/Mailing_Lists/Search</a>
before posting!<br>
Please don't post (un)subscribe requests to the list. Use the<br>
www interface or send it to <a moz-do-not-send="true"
href="mailto:gmx-users-request@gromacs.org">gmx-users-request@gromacs.org</a>.<br>
Can't post? Read <a moz-do-not-send="true"
href="http://www.gromacs.org/Support/Mailing_Lists"
target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
</blockquote>
</div>
<br>
</blockquote>
<br>
</body>
</html>