[gmx-users] Box generation for monoclinic crystal - regd

[Mark Abraham]

On 9/04/2012 3:22 PM, ramesh cheerla wrote:
> Dear Gromacs users,
>
>                            I am planing to simulate a polymeric 
> crystal in gromacs, which is of monoclinic unit cell with cell 
> parameters a = 0.805 nm b = 1.304 nm and c = 1.948 nm and beta = 125.4 
> deg. my crystal is of 424 type  i.e 4 unit cells along 'a' direction  
> '2' unit cells along 'b' direction and 4 unit cells along 'c' direction.
> So my box vector lengths are a = 3.22 nm, b = 2.608 nm, c = 7.792 nm.
> I  am using editconf to generate box for this crystal as:
>
> editconf -f input.gro -o out_box.gro  -bt tric -box 3.22 2.608 7.792  
> -angles 90  125.4 90
>
> Here i have some doubts:
>
> 1) Am i using the editconf in correct manner for generation of box for 
> the monoclinic crystal, though the crystal is of monoclinic  i am 
> using box type as triclinic
> i haven't found any specific method for the generation of box for the 
> monoclinic crystal.
> 2) Is there any specific method to generate the box for the monoclinic 
> crystals.
> 3) The notations for the lattice parameters of the crystal i.e 'a' , 
> 'b' , 'c' and angles alpha, beta , gamma used by crystallography and 
> the gromacs editconf are the same or different, why because if i use 
> editconf as above i am getting the following warning:
>
> WARNING: Triclinic box is too skewed.

This means equation 3.3 of the manual is not true. Maybe beta should be 
acute? I've no idea of necessity or convention here.

Inspecting the whole output of editconf is probably instructive, but 
you've kept all that information to yourself instead of re-thinking how 
your repeat post might be constructed so as to make it easier / more 
likely for people help you. Likewise the final line of out_box.gro. Make 
sure you have read and understood what documentation is available with 
editconf -h and in manual section 3.2.

Mark