[gmx-users] hand made rtp entry: grompp doesn't find the dihedral parameters i included explicitly
mbx0009 at yahoo.com
Sun Nov 6 15:02:12 CET 2005
I made myself an additional entry for an organic
molecule in ffoplsaa.rtp. I use existing atom types
from opls-aa and everything works fine with additional
explicit bonds/angle parameters i included to the rtp entry.
however, for some of the proper dihedrals grompp complains:
"No default Ryckaert-Bell. types, using zeroes"
although I DID include these parameters explicitely
in the rtp entry.
what i did was:
1) look up a similar dihedral angle in the existing opls-aa
2) put a #define statement into ffoplsaabon.rtp, as in:
"#define dih_my_parameters" followed by the six numbers
i got from that similar angle
3) include a "[ dihedrals ]" section into the rtp entry
with a line starting with the 4 atom names followed by
the keyword "dih_my_parameters"
4) after pdb2gmx i obtain a topol.top file that does
include this angle followed by the word "dih_my_parameters"
which is what i expected, however, when using this topol.top
file with grompp i still get the abovementioned warning(s)
what am i missing ? is there somnething hard-coded here ??
i also tried to include the paramters directly
into the ffoplsaabon.itp file but grompp wouldn't find them,
e.g., i have a bond between two atom-types that correspond
to the bond-types CA and CV.
if i include a corresponding line into the "[ bondtypes ]" section
of ffoplsaabon.itp, using force constant and bondlength from a
similar bond in an AA, grompp keeps complaining about not finding
parameters for this bond.
i also found that some (not all) of the dihedrals grompp is missing
should actually be there since they are covered by definitions
including wild cards, such as X CA CA X, in ffoplsaabon.rtp.
grompp would not find them either ...
thanks in advance for any help!
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
More information about the gromacs.org_gmx-users