Excuse me if I'm wrong but should't the correlation coefficient
between Fo and Fc (which is what the Density Fit analysis is looking
at) be independent of this issue, provided Fo and Fc are on the same
scale ? And since the map is calculated from weighted 2Fo-Fc
amplitudes and phases the scaling problem should not occur.
What seems to happen is that Coot doesn't even "look" at the map - it
immediately displays a collection of red bars. And to confuse the
issue even more, I have data from another structure refined with
PHENIX that works fine!
At 16:10 17-01-2013, Phil Evans wrote:
(hesitating to restart an old flame war)
There has been a long-standing argument about scaling of electron
density maps, with many people deprecating "sigma-scaling", for two reasons
1) in 2Fo-Fc type maps, "sigma" = RMS density is a function among
other things of the solvent content (smaller with high solvent content)
2) in difference maps Fo-Fc type, as the model improves, sigma gets
smaller (ideally = 0). Sigma contouring is useful to pick out the
largest features though.
Scaling even approximately to e/A^3 (i.e. scaling Fobs to Fcalc) is
consistent across solvent content and model quality, though still
affected by resolution and other things, and generally should be
preferred in my opinion. Sharpening map coefficients may also affect the scale
Phil
On 17 Jan 2013, at 15:48, Nat Echols <nathaniel.ech...@gmail.com> wrote:
> On Thu, Jan 17, 2013 at 3:16 AM, Pedro M. Matias
<mat...@itqb.unl.pt> wrote:
>> To clarify my previous e-mail, in both cases the maps are being calculated
>> from weighted coefficients from an MTZ file output by the program.
>> I tried calculating a weighted 2Fo-Fc map in CCP4i using the
PHENIX MTZ file
>> and the end-result (red bars) was the same.
>
> If I had to guess, I'd say that the output from Refmac is on an
> approximately absolute scale, i.e. the volume-scaled density values
> resemble the actual electron densities expected for the model. (I say
> "approximately" because the absence of F(0,0,0) and various FFT
> artifacts pretty much guarantee that the values are not actually
> accurate, just on the same order of magnitude.) This is presumably
> done by scaling F(obs) to F(calc). Maps from Phenix are definitely
> *not* on an absolute scale, however, and I guess the same must be true
> for BUSTER. Since everyone is used to looking at sigma-scaled maps
> this has not been a problem in the past; I don't think older versions
> of Coot had this problem either.
>
> You are not the first person to complain about this, for what it's worth.
>
> -Nat
Industry and Medicine Applied Crystallography
Macromolecular Crystallography Unit
___________________________________
Phones : (351-21) 446-9100 Ext. 1669
(351-21) 446-9669 (direct)
Fax : (351-21) 441-1277 or 443-3644
email : mat...@itqb.unl.pt
http://www.itqb.unl.pt/research/biological-chemistry/industry-and-medicine-applied-crystallography
http://www.itqb.unl.pt/labs/macromolecular-crystallography-unit
Mailing address :
Instituto de Tecnologia Quimica e Biologica
Apartado 127
2781-901 OEIRAS
Portugal
