[gmx-users] clustering "pdb outsiders"

[Nuno R. L. Ferreira]

Hi *

I've some hundreds of pdbs (conventional format) that came out from a docking 
calculation.  

My aim is to analyze these structures, using g_cluster, to find out how's the 
cluster distribution.  I don't know of any other soft. that can do this, so I 
pointed my batteries to gmx.

Some thoughts that ocurred me ...

1- First I have to build a tpr file for  one of the input pdbs (they contain 
only regular aminoacid residues).  The input pdbs have 3 chains, and each one 
is also broken in 3 pieces.   I tryed to build a topology using opls f.f. and 
I was successful.  Just have to figure out how to resolve 2 long bond 
warnings, for each chain (possibly the bond between the chain broken parts).

2- Concatenate all the pdbs, to have a kind of trajectory, ready to be fed up 
into g_cluster (easy thing to do).

I think I have to change the nomenclature of the input pdbs, to meet the 
nomenclature stored under the produced .tpr file (residue and atom names I 
mean).
But this is rather difficult to do, cause not only I have to change some 
residue names (like LYS --> LYSH, or HYS --> HISA), atom names (like ILE HD11 
--> ILE HD1), but also I suppose that I have to change the residue number, 
cause pdb2gmx joins all those broken pieces, renumbering them.

Any suggestions ?

Best regards,
Nuno



-- 
**********************************************
Nuno Ricardo Santos Loureiro da Silva Ferreira
Grupo de Química Biológica 
Departamento Química
Universidade de Coimbra
Portugal
www.biolchem.qui.uc.pt
**********************************************
"Do not worry about your difficulties in maths.
 I can assure you that mine are still greater."
                                   emc2