On Tue, Nov 3, 2015 at 9:39 AM, Gerald Monard <
Gerald.Monard.univ-lorraine.fr> wrote:
> Hello,
>
> On 11/01/2015 11:52 AM, Jason Swails wrote:
> > On Sat, Oct 31, 2015 at 1:21 PM, Gerald Monard <
> > Gerald.Monard.univ-lorraine.fr> wrote:
> >
> >> As a follow-up, I can partially solve the problem using "iat":
> >> # d2
> >> &rst iat = 11, 48, r0 = 3.2, k0 = 0., /
> >> # d3
> >> &rst iat = 48, 49, r0 = 3.2, k0 = 0., /
> >> # d4
> >> &rst iat = 49, 16, r0 = 3.2, k0 = 0., /
> >> # d2-(d3-d4)
> >> &rst restraint = "coordinate(distance( :1.C :3.O ), 1.0,
> distance(
> >> :3.O :3.H1 ), -1.0, distance( :3.H1 :2.OP2 ), 1.0 )", r0 = 3.2, k0 =
> >> 0., /
> >>
> >
> > What happens if you use the "iat" approach here as well? I've often
> > observed rather funny behavior with the "natural language" restraints,
> and
> > there are a number of peculiar things about that code (for instance -- it
> > doesn't use the standard mask parser, it implements its own highly
> limited
> > version).
> >
> >
>
> Well, the problem is that I want to use the 'coordinate' mode because I
> want to restrain the difference between 2 distances. I don't see how to
> do that with iat.
> I tried the 'igr' command also (it is mentionned in the manual an
> unclear relationship between the two), but with no success.
> Thus I've started to check the code...
>
It works just fine. To use igr1...8, you need to put "-1" in the iat
column for that atom. Then you need to use "rstwt" to apply restraint
weights. For example:
&rst
iat=-1,-1,-1,-1, igr1=1, 2, 3, 4, 5, igr2=6, 7, 8, 9, 10,
igr3=11, 12, 13, 14, 15, igr4=16, 17, 18, 19, 20,
rstwt=1,-1, r1=-10, r2=-5, r3=-5, r4=10, rk1=10, rk2=10,
/
This defines a coordinate defined by (1)*D1+(-1)*D2, where D1 is the
distance between the COM of igr1 and igr2, and D2 is the distance between
the COM of igr3 and igr4. The flat-well potential is defined as per usual
by the r1, r2, r3, r4 and rk1/2 variables.
This just becomes one of your restraints, and can be used alongside any
number of other simple restraints.
All the best,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Nov 03 2015 - 07:00:04 PST
