I am now modifying the script 'dock_tor_rigid.py' from
eginput/dock_dipolar_chemshift. To perform minimization of the oligomer, I
added lines as follows:
command("""
ncs restraints
initialize
group
equi (segid A)
equi (segid B)
equi (segid C)
equi (segid D)
...
weight = $kncs
end
?
end""")
...
m.fix( select('resid 5:20') )
groupList = m.groupList()
groupList.append( select('resid 25:43') )
groupList.append( select('resid 5:20'))
For each protomer, I want to fix residue 5 to 20 and allow residue 25 to 43
moving as an alpha helix to reduce energy of the crowed oligomer center. The
output is something like this:
%WDSUB-ERR: symbol not found:
weight = $kncs
^^^^^
%WEIGht-ncs=-ERR: real number expected:
weight = $kncs
^^^^^
NOTE: The first EQUIvalence set defined in a GROUp is taken as the reference!
NCS group 1 has 18 sets of equivalent atoms:
Equivalence set 2 :
R.m.s. diff. between fitted set and comp. set for selected atoms = 0.0004
...
dock.py(163): m.setStepType( 'powell' )
dock.py(164): m.setVerbose( m.verbose() | m.printNodeDef )
dock.py(165): m.setNumSteps( 1000 )
dock.py(166): m.setDEpred( 1 )
dock.py(167): m.setETolerance( 1e-7 )
dock.py(168): m.setPrintInterval( 1 )
dock.py(169): m.fix( select('resid 5:20') ) # Rigid body minimization set
up
dock.py(170): m.fix( select('resid 500') )
dock.py(171): groupList = m.groupList()
dock.py(172): groupList.append(select('resid 25:43') )
dock.py(173): m.setGroupList( groupList )
dock.py(176): m.potList().removeAll()
dock.py(177): m.potList().add( XplorPot("VDW") )
dock.py(178): m.run() # Minimize with just NOEs and van der Waals (rigid
body)
AT_Build::buildNode: cycle link found between atoms 962 GLY 21 N and 960 PHE 20
C
removing bond.
...
AT_Build::buildNode: cycle link found between atoms 11538 GLY 21 N and 11536
PHE 20 C
removing bond.
Node info:
node 0 0: -1 (base atom), 19 SER 5 N, 20 SER 5 HN, 21 SER 5 CA, 22 SER 5
HA, 23 SER 5 CB, 24 SER 5
node 1 0: 17 ILE 4 C, translate
node 1 1: 301 GLY 21 N, translate
...
node 26 16: 11539 GLY 21 HN, translate
total number of degrees of freedom: 5784
*-- POWELL ------ step= 0 --- stepsize= 0.01000 --- energy evals= 2 -*
| E(poten)=882181.6704152 grad= 98.1839386 VDW=882181.6704152 |
*------------------------------------------------------------------------------*
...
--------------- cycle= 1 --------------------------------------------------
| Etotal =0.29E+10 grad(E)=79.202 E(VDW )=444532.967 E(NOE )=0.000 |
| E(RAMA)=0.000 E(COLL)=0.29E+10 E(SANI)=0.000 |
-------------------------------------------------------------------------------
dock.py(399): rms_bonds = command("print thres=0.05 bonds","result")
(atom-i |atom-j ) dist. equil. delta energy const.
(A 4 C |A 5 N ) 371.521 1.329 370.192 ******** 1000.000
(A 20 C |A 21 N ) 397.705 1.329 396.376 ******** 1000.000
(A 24 C |A 25 N ) ******** 1.329 ******** ******** 1000.000
(A 44 N |A 44 HN ) 75.604 0.980 74.624 ******** 1000.000
(A 44 N |A 44 CA ) 81.043 1.458 79.585 ******** 1000.000
...
I am not sure what's going on here, but apparently something has gone quite
wrong. What have I missed?
It seems that I should add the posSymmPot, is there an example?
Why is the bond between resid 20 and 21 removed?
The degree of freedom is definitely wrong, did I did the NCS part wrong?
Thank you very much for your help.
Yi Zhang
> From: Charles at Schwieters.org
> To: zhang_yi at pku.edu.cn
> CC: xplor-nih at nmr.cit.nih.gov
> Subject: Re: [Xplor-nih] Energy minimization for oligomers
> Date: Thu, 28 Oct 2010 11:12:00 -0400
>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
> Hello Yi--
>
> >
> > Thank you very much for your previous reply. The side chain
> > minimization went well now. Here is another issue, I have an oligomer
> > model generated according to symmetry. The core of the model is quite
> > crowded due to overlapping of a helical component. I would like to
> > perform rigid body refinement of the modeled structure without
> > experimental determined restraints. To do this, I need to fix the
> > peripheral part of the protomer and group the central alpha helix,
> > allowing the central alpha helix to move apart from each other during
> > energy minimization. The script I use is 'sa_mef_dipo.inp' from the
> > eginput/ mef_dna. How should I express the 'fix and group' in this
> > macro? (The expression in 'anneal.py' does not work here.)
>
> Because that old XPLOR script was written before the IVM was added, it
> does not support rigid body grouping in the same was as in the Python
> scripts (or using the dyanmics internal statement in the old XPLOR
> language). My advice is to switch to a different base script for this
> work.
>
> >
> > Another problem is that during minimization, the central helix should
> > not be thrown out from the core, can the script achieve this without
> > experimental restraints? Should I modify the simulation parameter such
> > as temperature, step, etc. and how?
> >
>
> If the nonbonded (vdw) energy term is enabled, the core portion should
> not move through other atoms to escape. However, if strain is too large
> I suppose it is possible- in this case, you might add a restraint to a
> subset of core atoms specifying that they move a bounded amount from
> their initial positions. This can be achieved with a PosDiffPot in the
> Python interface.
>
> best regards--
> Charles
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.10 (GNU/Linux)
> Comment: Processed by Mailcrypt 3.5.9 <http://mailcrypt.sourceforge.net/>
>
> iEYEARECAAYFAkzJksAACgkQPK2zrJwS/lbI4wCdFArPOmiw+POdFhNZMg5HMheQ
> gJUAnRmHgRP5FfiglH+xnrzfqFQQ8Fo+
> =wBDg
> -----END PGP SIGNATURE-----
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20101029/d7baaf8c/attachment.html
