[gmx-users] PRODRG again

[Ing. Vojtech Spiwok]

>I have a "drug" greater than 300 atoms what should I do to get the
>topology
>I don know if exist another program or server that do something similar 
>My drug have 310 atoms 
>Help please  :-(
>

Such drug must be a nightmare of bioavailability experts, anyway :-)
I was succesful to generate a topology of large compounds by linking
of PRODRUG topologies of smaller moieties. I.e. if you have two
compounds linked by aliphatic chain you can generate two itp files
and put them together, renumber atoms in the second file, and add
some more parameters. In example of aliphatic linkage it is one bond,
two angles, three dihedrals, three pairs etc (as far as I remember).

if the compound is:
R-CH2-CH2-CH2-CH2-R'
then
1.) generate itp for:
R-CH2-CH3 and CH3-CH2-R'
2.) renumber atoms in CH3-CH2-R' to start with (number of atoms in first
compound + 1), cocatenate itp files and put together [atom], [bonds] .. and
other fields.
3.) Change types CH3 to CH2, add CH2-CH2 bonds and all neccesary
angles, diherdals, pairs etc.
4.) Cocatenate corresponding gro files and renumber them, update number
of atoms and possibly change a box size.
5.) Minimize your "monster-drug". If there is some error you are likely to
see it in a final structure.

If the drug is not R-CH2-CH2-CH2-CH2-R' then it will be more
complicated, but idea is the same. It is definitely also possible
to build some larger compounds from scafolds in a similar manner like
protein from aminoacids.
I hope it will help

Vojtech Spiwok