Re: [modeller_usage] absolute position restraints

[Modeller Caretaker <modeller-care@ucsf.edu>]

Frank Böckler wrote:
> I'm trying to keep some part (ligand) fixed in space using the restraint 
> features 9, 10 and 11 and the restraint type 34. 
> Unfortunately, I was not quite successful with this approach (outlined 
> below). 
> Can anybody give me an advice, how to proceed correctly (correct syntax 
> for the restraints, other approach giving the same result, ...)?

If you don't want part of your system to move at all, you can just 
deselect it. Modeller only optimizes selected atoms. See
http://salilab.org/modeller/manual/node27.html

Otherwise, using absolute position restraints is your best option.

> MODELLER5 VERSION: MODELLER FORMAT 
> R    3   1    9   34   1   2   0  1936   46.483   0.0 

> My idea was to constrain the x, y or z coordinate (feature 9, 10, 11) of 
> the ligand with a harmonic potential (3) written into restraint type 34 
> and applied on 1 atom (using 2 parameters: the coordinate and the standard 
> deviation). I've used both 0.0 and higher values as standard deviations to 
> fix the atom position absolutely or to allow some movement, however both 
> was unsuccessful. Perhaps the whole syntax is false??? 

Your syntax looks fine, but it won't work with a stdev of 0. If it 
doesn't work with a non-zero stdev either, then I'm not sure exactly 
what's wrong. Send me your input files and I'll take a look.

	Ben Webb, Modeller Caretaker
--
modeller-care@ucsf.edu             http://www.salilab.org/modeller/
Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage