[gmx-users] Minimize energy of holoprotein

[Zorzan Simone]

Hello to everybody, I am new of bioinformatics (I am a computer
scientist), but I'm trying to learn.

Actually I am working on a model of a holoprotein (protein with "other"
molecules), cp26, a transmembrane protein found in tylacoids), modelled
for homology (56% identity) from an axisting one (1rwt.pdb monomer).

I would like to minimize the energy of my model, but I have some
difficulties with ligands:

1) I can't use pdb2gmx since I get "Residue 'LUT' not found in residue
topology database" for the lutein.
So I got topology for lutein from PRODRG, and reading John E. Kerrigan
tutorial I would like to merge files.
Anyway I think it's a pity to lose information about relative positions of
apoprotein and ligands present in pdb file (I am not so sure that gromacs
can rebuild them minimizing energy, and these position are pretty
correct).
Will I have to write numerous constrains for the information lost in
dividing pdb file or is there a way to make pdb2gmx process the
holoprotein with ligands in place to generate the correct output files?

2) I have another kind of ligands: clorophylls, of two kinds (a and b).
They contain a polar head with an 'eme' group binding a Mg atom that is
essential for their binding to the active sites.
Sadly prodrg does not support such atoms (is it because of the energy
fields used?) and I really don't know how to create a topology files for
them.

It would be really useful a program to minimize the energy directly of the
holoprotein, does it exist?
If not, is there a way to treat the holoprotein as a unique protein with
pdb2gmx, so keeping all information about relative distances?

Thank you very much for your help!

Simone Zorzan.

---

Molecular and Industrial Biotechnologies Phd Student
Dept. Scientific and Technological, Università di Verona
Ca' Vignal 1
strada le Grazie 15
I-37134 Verona VR  (Italy)
Tel: +39-045-802 7915   Fax: +39-045-802 7929