Help! There is no angle or dihedral definitions for patches in .psf!

From: Edward Patrick Obrien (edobrien_at_Glue.umd.edu)
Date: Thu Apr 29 2004 - 10:43:09 CDT

I tried sending this message yesterday but it appears it didn't go
through...

On Wed, 28 Apr 2004, Ed wrote:

 Hi All,
    Serious problem here (I think), I create a .pdb and .psf with patches
 of CT3 (methyl amide) and ACE, but in the .psf file there is no angle
 definitions nor dihedral definitions for the patched atoms!

    This leads to serious problems... when I minimize, or run MD, the atoms
 in the patches contort and float around like they don't feel each other,
 outrageous bond angles can occur! And finally the system crashes at some
 point because RATTLE can't take it.

    Why are angle and dihedral not being defined? How can I correct it?

    Just in case you don't believe me check this out:

    If I just use the Standard NTER and CTER in charmm (less atoms then ACE
 and CT3) the number of angles and dihedrals are:

         Info: writing psf file test.psf
         total of 172 atoms
         total of 174 bonds
         total of 314 angles
         total of 461 dihedrals
         total of 21 impropers

    Now if I patch with CT3 and ACE:

         Info: writing psf file test.psf
         total of 181 atoms
         total of 183 bonds
         total of 307 angles
         total of 453 dihedrals
         total of 24 impropers

    So there are more Atoms but less angles and dihedrals? Sounds crazy to
 me.

    Also here are the positions of the hydrogens on the carbon of the
 amide-methyl BEFORE minimization:

 ATOM 170 HT1 SER 11 19.167 -0.569 0.788 1.00 0.00
 ATOM 171 HT2 SER 11 19.311 -2.153 1.177 1.00 0.00
 ATOM 172 HT3 SER 11 20.181 -1.533 -0.064 1.00 0.00

    Here are the same hydrogens AFTER 2000 steps of minimization:

 ATOM 170 HT1 SER 11 20.509 -1.046 0.825 1.00 0.00
 ATOM 171 HT2 SER 11 20.560 -1.180 0.961 1.00 0.00
 ATOM 172 HT3 SER 11 20.432 -1.047 0.906 1.00 0.00

    Notice anything strange!? The atoms are practically on top of each
 other!?

    Here is the TCL script I use to create the .pdb and .psf:

 package require psfgen
 topology top_all27_prot_lipid.inp
         segment X {
         first none
         last none
         pdb X_start.pdb}
         patch ACE X:1
         patch CT3 X:11
         coordpdb X_start.pdb X
         guesscoord
         writepdb test.pdb
         writepsf test.psf

 Any suggestions would be much appreciated,
 Ed

 PS - I've tried using 'auto' to no avail.
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:38:37 CST