[gmx-users] minimisation converted into machine precision

[Mark Abraham]

nitu sharma wrote:
> Dear all
> 
>         After concatanation of protein with lipid I did inflatgro 
> processing forscaling but I am facinng problem during energy 
> minimisation step of inflatgro procedure . The minimisation terminated 
> after few step and it converted into machine precision-
> Steepest Descents converged to machine precision in 349 steps,
> but did not reach the requested Fmax < 1000.
> Potential Energy  =  4.2808670e+06
> Maximum force     =  4.9091840e+04 on atom 7290
> Norm of force     =  8.9087488e+02
> 
> As I searched its becoz of clashes between lipid and protein .
> Do anyone have idea about how can we avoid these clashes ?whats the 
> parameter required to avoid this?

We can't tell. If you have atomic clashes they'll normally be apparent 
if you *look* at the structure. Atom 7290 might be a good place to 
start. Only then can you have an idea where the problem might have 
arisen in the method to arrive at that structure.

Mark