RE: angle deformation

From: CHINDEA Vlad (vchindea_at_hotmail.com)
Date: Fri Dec 19 2008 - 18:02:06 CST

Dear James
 
This was a revelation to me ! I could never imagine that angles and dihedrals might be missing from the PSF in the absence of any complaints from the MD program.
 
Since my PSF was built with AutoPSF I have no script that I can adjust but I have tryed to do as suggested by you enter the following lines at the VMD prompt:
 
topology <file>
regenerate angles dihedrals (here I got the message: regenerating all angles, dihedrals)
set sel [...all]
$sel writepsf....
 
but PSF file was generated with no angle, dihedral or improper information ! Any ideeas on what might be wrong ?
 
If all fails I imagine that I could add the missing angles and dihedrals by hand in the PSF file. Are there any issues with this approach ? Of course I would have to take care of the total no. of angles (dihedrals) and proper no. of items/line.
 
Many thanks and kind regards
 
Vlad
  From: gumbart_at_ks.uiuc.eduTo: vchindea_at_hotmail.com; namd-l_at_ks.uiuc.eduSubject: RE: namd-l: angle deformationDate: Thu, 18 Dec 2008 23:04:17 -0600

They fluctuate wildly because they are not specified in your PSF. In your psfgen script, add the following line after using the patch statement:
 
regenerate angles dihedrals
 
PSFgen autogenerates the angles and dihedrals ONLY when building the segment, or when it is explicitly told to do so like above.
 
 

From: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] On Behalf Of CHINDEA VladSent: Wednesday, December 17, 2008 5:01 PMTo: NAMD listSubject: namd-l: angle deformation

 
 
Hi everybody I am simulating a small ligand using topology/parameter files modified by analogy with other parameters available in the original files. The initial topology of the molecule is OK but as soon as MD starts the angle C-OT2-CT gets deformed to wild values (jumping between 150 and 120 degrees instead of the theoretical 109) and also the angles HT-CT-HT (jumping between 10 and 150 degrees during a simulation instead of the theoretical 110 degs). The choices I made when modeling the missing parameters looked intuitively correct to me, so now I am a bit stucked. I know that tampering with the topology/parameter files is done 'at my own risk' but any ideea about why I am getting this wild behaviour would be very appreciated. Please find attached the links to the modified topology (with new atom names HJx for the NTER patch), parameter file (the new entries are commented with 'analogy...' on the second comment line) and the psf and pdb files of the ligand. http://rapidshare.com/files/174371793/STME.pdb
http://rapidshare.com/files/174371794/STME.psf
 
http://rapidshare.com/files/174371795/par_all27_prot_lipid_na_mod.inp
 
http://rapidshare.com/files/174371796/top_all27_prot_lipid_na_mod.inp
 
Excerpts of the new entries from the parameter file:
 
HA CT1 HA 35.500 109.00 5.40 1.80200 ! TEST for test cpd ! analogy based on HA CT2 HA
HC NH1 CT1 30.000 109.50 20.00 2.07400 ! ALLOW POL ALI ! analogy from HC NH3 CT1
HC NH1 HC 39.000 106.5000 ! ALLOW POL ! analogy from HC NH2 HC
NH3 CT1 CD 43.700 110.0000 ! ALLOW PEP POL ARO ALI ! analogy from NH3 CT1 C
HC NH1 CT1 CD 0.1100 3 0.00 !analogy from HC NH2 CT2 CD
HC NH1 CT1 CT2 0.0000 1 0.00 ! ALLOW PEP ! analogy from H NH1 CT1 CT2HC NH1 CT1 HB 0.1100 3 0.00 !analogy from HC NH2 CT2 HBMany thanks and warm regards Vlad

Explore the seven wonders of the world Learn more!
_________________________________________________________________
Drag n’ drop—Get easy photo sharing with Windows Live™ Photos.

http://www.microsoft.com/windows/windowslive/photos.aspx

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:50:17 CST