[gmx-users] bad box error
jalemkul at vt.edu
Tue Jan 29 21:51:12 CET 2013
On 1/29/13 3:46 PM, amna khan wrote:
> i am using the best conformation of protein ligand file obtained after
> docking results
> i made the .gro file by
> pdb2gmx -f pro.pdb -o proceed.gro -water spc -missing -ignh
> then the ligand codinates from ligand,gro was obtained from prodrug sever,
Beware the quality of PRODRG topologies - they are never sufficiently accurate
without manual reparameterization. Using the coordinates from PRODRG may or may
not cause problems. Make sure you don't allow PRODRG to run EM on the
structure, or else the docked pose will be altered.
> then i pasted ligand coordinates in this file
> the first three lines are
> Gallium Rubidium Oxygen Manganese Argon Carbon Silicon
> 705SER N 1 -3.373 -1.131 1.295
> 705SER H1 2 -3.442 -1.166 1.232
> 705SER H2 3 -3.339 -1.043 1.261
> 705SER H3 4 -3.298 -1.196 1.303
> last three lines are
> 1LIG OAR 33 0.425 -0.911 1.194
> 1LIG CAC 34 0.593 -1.073 1.167
> 1LIG CAT 35 0.500 -1.191 1.198
> 9.21266 2.49507 2.95990
> and the output for command is
> wc -l proceed.gro
> 2100 proceed.gro
This shows your problem. You didn't account for the fact that you pasted in a
ligand by properly incrementing the atom count on the second line of the file.
Thus, when editconf reaches atom 2059, it expects to find box vectors on the
next line, but instead finds another atom and gives you the "bad box" message.
Your file actually has 2097 atoms (2100 - 3).
Note that there is a protein-ligand tutorial available if these concepts are
unfamiliar to you.
Justin A. Lemkul, Ph.D.
Department of Biochemistry
jalemkul[at]vt.edu | (540) 231-9080
More information about the gromacs.org_gmx-users