[gmx-users] Error message: Cut-off length is longer than half the shortest box vector

[Mark Abraham]

Lum Nforbi wrote:
> Hello everyone,
>       Thanks to Tsjerk, Mark Abraham, Justin and Dr. Vitaly for the very 
> useful inputs to my quest for generating a .gro file from a .pdb file.
>       I am now trying to minimize the energy of the system so I can do 
> an mdrun but I keep having the error message:
> the cut-off length is longer than half the shortest box vector or longer 
> than the smallest box diagonal element. Increase the box size or 
> decrease rlist.
>       I have changed the values of rlist, rvdw and rcoulomb, 

Don't, unless you know what you're doing. Your force field is 
parametrized under certain conditions, and varying these quantities by 
much will remove you from conditions where the force field might be 
supposed to work well. It does sound like you would benefit from doing 
some general background reading about molecular mechanics force fields 
and molecular dynamics simulations, and/or some more tutorial material.

> and even 
> the box size several times but I still keep having this message. Please, 

You can change your box size with editconf, and then you will need to 
use genconf (or other) to increase the amount of solvent.

> I need help to figure out what to do to fix this problem.
> The command line I am using is:grompp -f waters.mdp -c waters_b.gro -p 
> water.top -o watersinput.tpr

This command generates a (binary) run input file. It does not do 
anything to the box size, which it reads from the -c file.

>       Also, is there a way to convert atom types from one format to 
> another? I also have the following warnings:
> 
> "Warning: atom name 1 in water.top and waters_b.gro does not match (OW - O)
> Warning: atom name 2 in water.top and waters_b.gro does not match (HW1 - H)
> Warning: atom name 3 in water.top and waters_b.gro does not match (HW2 - H)
> atom names from water.top will be used
> atom names from waters_b.gro will be ignored."
> 
> I had drawn my molecule in ghemical and exported as a pdb file and so 
> the atom types O and H were automatically generated.

Using the coordinate file output from pdb2gmx for subsequent operations 
is the correct way to avoid these warnings. If you're not using pdb2gmx 
then you need to really understand what you're doing.

Mark