[gmx-users] Re: Problem in AFM pulling, the parameter unit and direction definition!

[Emily Walton]

>
> Dear all,
> I am confused about the unit of some parameters in AFM pulling  
> function. In manual 3.3, page 117, the unit of "afm_rate" is nm/ps,  
> and unit of afm_k is kJ mol-1 nm-2. But in the output file(*.log),  
> there is another unit:
>
> Initializing pull groups. Inv. mass of group 1: 0.024950
> Initial coordinates center of mass:    3.966    3.338    3.648
> Initializing reference group. Inv. mass: 0.000132
> Initial coordinates center of mass:    3.162    3.127    3.097
> Initial distance of group 1:    0.998
>
> Pull rate: 5.000000e-01 nm/ns. Force constant: 1.000000e+03 kJ/(mol  
> nm)
> Pull direction:    0.806    0.212    0.552
>
> what makes these different?
>
> and "Pull direction" seems to be gotten by subtracting coordinates  
> of referencePgroup from coordinates of pull group. This value of  
> "Pull direction:    0.806    0.212    0.552" is different from what  
> I set in pull.ppa files,which is "afm_dir1	= 0.588 0.428 -0.686"??
>
> Could anybody here tell me why? I am crazy about these problem!


Hi,

As far as the units on the spring constant go, only the kJ/(mol nm^2)  
make sense. I'm guessing (but haven't checked the code) that the kJ/ 
(mol nm) is just a typo.

The nm/ps rate is correct. I believe that's also a typo in the log code.

As for why your pull.ppa and md.log pull directions are different,  
I'm not sure. In my simulations, they are the same. If you post your  
pull.ppa file, that would help. Also, which version of the code are  
you running?

-Emily Walton