[gmx-users] more questions

[David van der Spoel]

On Tue, 9 Oct 2001, Daan van Aalten wrote:

>
>Hi Luke
>
>This is going to be quite difficult. Usually the way these topology
>builders (like pdb2gmx) work is that they need to be able to identify
>certain atoms by their name. For instance, to add the psi torsion
>angle between residues X and Y, one needs X_CA,X_C,Y_N,Y_CA for the next
>psi angle, between Y and Z, one needs Y_CA,Y_C,Z_N,Z_CA
>The problem now is that if you think of your entire complicated 3-amino
>acid thingy as "residue" Y, the atoms Y_CA in the first and second psi
>angle are *not* the same, and as such you cannot build a .rtp entry for
>this.
Not so pessimistic. There is another method using the special bonds file
that is also used for Cys-Cys bridges and e.g. Heme-Protein interconnects.
This looks for atoms in two residue types (Ser-GLy in this case) within a
certain distance, and if found makes a bond.

>Once the size limit on PRODRG is gone, one should be able to build entire
>topologies for protein which contain as much of these modifications as you
>want. Perhaps you can test a small section of your protein (< 100 atoms)
>with this chromophore in the middle? If it works I'll try to free some
>time to up the size limit to 10000 or so.
Obviously most users would really welcome this. Hope your algorithm scales
reasonably.

Groeten, David.
________________________________________________________________________
Dr. David van der Spoel, 	Biomedical center, Dept. of Biochemistry
Husargatan 3, Box 576,  	75123 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://zorn.bmc.uu.se/~spoel
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++