[gmx-users] triclinic box too skewed

[Erik Lindahl]

>> Dear all,
>>
>> I want to run a simulation on a polyamide crystal structure to test  
>> some force field
>> parameter changes. The crystal unit cell angles are alpha=48.5,  
>> beta=77, and gamma=63.5.
>> If I try to start any simulation with this
>> triclinic system I always get the error: Triclinic box is too skewed.
>> Does anybody know a way to solve this problem?

Yes,

Berk is the expert in this, but IIRC, a box can be equally well  
represented if you translate the z vector
one unit cell. We check this every step, since the box could otherwise  
be infinitely skewed.

Check if you can rotate it, or translate the z box vector coordinates  
by x/y box vector unit lengths.

Cheers,

Erik
------------------------------------------------------------------------ 
-----
Erik Lindahl, MSc, PhD     <lindahl at stanford.edu>
D109, Fairchild Building
Dept. Structural Biology, Stanford University School of Medicine
Tel. 650-7250754    Fax. 650-7238464