Hi Bob,
Yes, you are correct-non-identical altloc identifiers will usually address
this. My concern is that this prioritizes a notation (altloc) over a
legitimate bit of modeled information (occupancy) and leads to certain
types of model features that I think are problematic. For example,
providing altlocs to waters in conflict with discretely disordered amino
acids strongly implies that the model builder knows something about the
location of the other alternative location for that water, which is almost
never true. It also can fail in the other way, by concealing a legitimate
clash (atoms whose occ sum >1.0) if they have different altlocs. In all,
I would prefer supplementing the altloc criterion with an occupancy-based
one, which solves some of these problems using a refined feature of the
model (occs) rather than the a naming convention (altloc).
Best regards,
Mark
Mark A. Wilson
Associate Professor
Department of Biochemistry/Redox Biology Center
University of Nebraska
N118 Beadle Center
1901 Vine Street
Lincoln, NE 68588
(402) 472-3626
[log in to unmask]
On 2/1/16 10:41 AM, "CCP4 bulletin board on behalf of Robert Immormino"
<[log in to unmask] on behalf of [log in to unmask]> wrote:
>I'm guessing the issue is with reduce and probe. I used to work for the
>Richardsons, and there is a bit of a trick with partially occupied
>waters. If you put a partially occupied residue in AltA and a partially
>occupied water in AltB or vice-versa,
> then you shouldn't get a clash. If you have the residue as AltA and the
>water with no AltLoc you will get a clash. This all comes out of how
>Reduce decides what to use as "your model", and last I knew this was done
>with AltLocs and not occupancies. If your
> model already has the residue and the water in different AltLocs then
>this may be a bug that the Richardson's can address. Methyl rotation is
>possible in reduce, but is off by default. I think the idea behind this
>was that reduce doesn't currently, but may
> in the future have the power to do the sort of energy minimizations that
>balances each methyl "giving a little".
>Hope this helps,
>
>-bob
|